ITT: People not getting the joke when he says not to trust someone with their book on a shelf facing out.... And then adds more and more books and rubik's cube behind him for the entire video.
I loved how he said not to trust people filming in front of a bookcase with their own book facing out. And as his video progressed, more and more of his books were facing out. I love this guy.
@@nickguman192 Do that mean that he's getting more and more Rubik's cubes with the assumption that previous one was broken because it was hard to solve?
His reveal of having all his own book facing out surrounded by Rubik's cubes after saying he couldn't trust someone who does that made me laugh out loud. Well done Matt!
@@dargtagnan3696 I count 11. 4 on the top row, 5 on the middle, and 2 on the bottom. (All are Humble Pi except for Things to Make in the Fourth Dimension on the second row.)
1:49 3 Rubik's cubes and 2 Humble Pis 8:41 5 Rubik's cubes and 2 Humble Pis 9:34 6 Rubik's cubes and 4 Humble Pis 13:58 10+ Rubik's cubes and 7 Humble Pis 16:24 the whole shelf is now just Rubik's cubes and Humble Pis well played sir 👏👏👏
There is another way to describe what they showed -the result shows that voters who vote by candidate are more likely to dislike the presidential candidates.
"I don't trust a person with a bookcase as a background with their own book facing outward" -Matt Parker, author of humble pi. Lol 😂 Edit: I just noticed over the course of the video he adds more copies of his book. Absolute based legend.
@@filipsperl he's using the reference to the video he shows stills from and expanding on it, also he mentions that he doesn't mention his book... While knowing he is totally mentioning his book
I was literally thinking why they subtracted two different percentages that aren't equal in populations. Thank you Engineering Statistics in college for this omen lmao
I don’t even have a background in stats and although I couldn’t place why I also realized that something was off about the subtracting of the two percentages. It just felt kinda wrong but I wouldn’t have been able to tell you why.
The reason is that you must weight averages before performing any aggregate function on them. A percentage is essentially an average. An extremely common mistake that statisticians shouldn't make.
@@totovader actually it wasn’t math that had my alarm bells going. In a republican precinct what are reasons someone would not check the box for all republicans? They didn’t want t vote for a specific republican. The most main republican is the president. So as someone who still smells some fish but am more skeptical currently of “stollen” election or full coup that is what went through my head.
Better than that, as Naim Kabir pointed out in an article on Medium, some of his predicted straight lines involve points which are impossible. The headline of the article involves the words "Doubles Down on his Con" so Naim was even more brutal in describing Shiva that Matt was!
@@57thorns true that, this is about as apolitical as you can get while dis/proving a political argument - unfortunately less referenced bc its less inflammatory, but a better video for it.
17:26 "I'm meant to make silly, fun, educational, mathematical videos" Mate, using maths to help people understand election results and the debate about them is exactly what the world need right now.
"If your theories can live by the spreadsheet, then I'm afraid they can also die by the spreadsheet" was a much more cool and powerful sentence than i ever expected any sentence containing the word "spreadsheet" to be. hell yeah
@Bobby Jones No I think Matt's made an honest mistake. Yes his math is completely wrong, but it's kind of hard to spot. Shiva is 100% wrong too. Meanwhile both sides are just parroting misinformed talking points and yelling at each other. I
“I don’t know if I can trust someone with their own book face out” you dog you. Love how you managed to fit some Rubik’s cubes in there too. Taking troll to a whole other level.
@@inyobillBahaha. One of the more important lessons that I teach my students, the unscrupulous action of cherry picking data or information has not only ended relationships, it' has also started wars! Geesh, next thing they are going to tell us is that popsicle sales are the cause of the spikes of drownings!
Really? I thought it was more of a case of "I have a hypothesis, where can I find evidence to support it? I can't? I'll make some then." Or was I the only looking at the initial graph thinking "They're using WHAT for x and y axis? Why?"
@@gordonrichardson2972 Important corollary: You can never rebut every detail because there is always one more. After rebutting arguments for hours and hours, eventually you will have to stop to eat or sleep or go to work. Whatever argument you didn't get to is now the clincher that proves the idiot right. I learned this lesson the hard way from arguing with moon landing conspiracy theorists.
@@JohnnyAdroit and even still, if you could hypothetically rebut every single detail, both you and the person you're arguing with would have to agree to every single definition of every single word you use, which is just bonkers
@Yuri Bezmenov Yeah, I don't trust people who call correlation coefficients "strong R" or quantify them as "lowered". Take a stats class before you embarrass yourself any further. P.S. I've interpreting a very vague comment incorrectly. Unfortunately for Yuri, the spirit of comment still lives since he hasn't really shown any evidence.
@@gags730 Indeed, go ahead and plot them yourself. All the data is there and Mat and Dr Shiva have given you the tools to do so. Mat wants to be proven wrong (I am not so sure about Dr Shiva though).
I was a bit frustrated at first that I couldn't understand the beginning of this video. Now I get why I was confused: It is a video explaining why something that doesn't make sense, doesn't make sense lol.
@@George_Taylor_ The y axis is just the % of votes that each candidate got. Biden no having any data points below 0.3 just means that he didn't get less than 30% of the vote in any off the districts
@@George_Taylor_ Dots being more tightly clustered on a graph with a larger scale only makes sense because objects look closer together when you zoom out. If it was at the same scale, the distance between dots would look similar on both graphs.
Isn't a bigger issue the hypothesis that the straight ticket voters should be identical to the ticket-splitters in their preferences? Presumably people split their tickets to vote against someone on the ticket of the party they otherwise prefer.
Exactly - Shiva's plot is a reference to his own prediction, not proof of anything (beyond his prediction not meeting reality). As you've said, this prediction of %president-%party=0 isn't exactly compatible with logic!
Exactly. We can assume that if someone didn't elect to vote straight ticket then they split their ticket. Meaning their was some Democrat on that ticket they elected to vote for instead. I dont have a source for this, but I believe it's not uncommon for people to vote for conservative representatives and liberal presidents.
@@Draenal , The Lincoln Project wanted Trump Gone... but would otherwise vote Republican on the rest of the ticket... because they only want Trump Gone... as they are Republicans.
@@Draenal It may not even be about voting for the opposing side, it could be abstaining from a particular race if you feel you couldn't back either candidate. There was an Economist/YouGov survey that get mention in an article about a Republican Governor endorsing Biden (Charlie Baker maybe?) where 16% of Republicans and 11% of Democrats stated they wouldn't back their candidate in the presidnetial race. It also mentioned 3% of Republicans choosing to vote for Biden, and 1% of Democrats choosing to vote for Trump, but it's not clear from how it was written if they were already part of the 16 and 11 per cents mentioned earlier And none of this offers any statistical method to account for people who aren't affiliated to a party, just to scuff up the linear prediction that bit more...
Basically, either these guys had no idea how to do the math properly, and therefore can’t be trusted; or they knew *exactly* how to do the math properly, and did it improperly on purpose so they could argue a faulty conclusion, in which case they *definitely* can’t be trusted.
The problem is that math often isn't easy to understand so people will go with "Trump will win, a mathematician has already proved that votes are wrong everywhere".
This guy would have to flop the Biden graph and put the strongest blue precincts to the right to be fair. He didn't. When you do, Biden goes up. Not down.
As noted in this vid, you can do that if the set sizes were equal-and Dr Shiva specified they were. So, embarrassing mistake on you part? However a problem remains as far as the claim of equal set sizes.
@@jrettetsohyt1 - I think you can kind of do it when you know the 'weight' of the percentages, there is such a thing as a weighted mean, so in theory you could try reverse engineering that calculation, but would it prove something, or just be someone messing around with numbers?
Propagation is only natural and is to be expected... Propagation is only natural and is to be expected... Propagation is only natural and is to be expected... Propagation is only natural and is to be expected... Propagation is only natural and is to be expected... Propagation is only natural and is to be expected...
@@George_Taylor_ that scale difference is very small and likely the scale the graph generates with to fill the space it was allocated. yes the data points are more tightly clustered but this is fairly minor and this proves nothing
@@George_Taylor_ I’m unsure whether you’re trolling or not, but in the case you’re sincere, I’d like to point out the simple fact that any scatter plot will get *less* clustered as you zoom in. Because Biden’s plot is “larger” (i.e. the grid behind is larger due to a smaller range of values on each axis), If you were to resize Biden’s plot to be on the same scale as Trump’s, it would look even more tightly clustered than it does in the video.
And this is why every single person needs more education in statistics. To be able to spot inconsistencies in data and to spot when they are being misled using data is of vital importance to literally everyone.
A good publication on this is "How to lie with Statistics" by Darrell Huff who explains in simple terms the tricks people use to lie with statistics (including different types of graph).
One thing to think about is that when we know the percentage of people voting for one candidate, we approximately know the percentage of people voting for the candidate. This is not like flipping a coin where we would know this exactly since there are more than 2 candidates in the race. The consequence of the 2 percentages being related mathematically means that when we create a plot for 1 candidate using the percentage, the other plot is mathematically determined. Briefly, it would be more surprising if the Trump and Biden scatter plots didn't show the same trend. Here is a brief video explaining why the 2 scatter plots have to be similar. th-cam.com/video/uMApwgGFVlw/w-d-xo.html
@@Tokahax I did spot Matt's switch of y-scale when comparing Trump's vs Biden's slope. This is a common tactic used to misrepresent the similarities between two charts. Matt did mention that Biden's slope was steeper in the audio, but that doesn't carry over visually in the charts and is therefore misleading.
As far as I can see the two scatter plots at the brgining of the video show exactly what Dr Shiva was describing. The ratios according to the plots are; for trump +10% in democrat areas to -30% in Republican areas,whereas Biden is +30% in Republican areas to -10% in democrat areas. Isn't this exactly what Dr shiva was saying? Biden is getting up to 30% more than expected in Republican areas and Trump is getting up to 30% less than expected in Republican areas.
Your restraint and respect to people who could justifiably be called charlatans is admirable and a good example to us all, Mr Parker. 👍 Thank you for being the hero we need rather than the demagogue we deserve. 🙏🤝
He's just being careful, as he said. If he starts making statements like that he could be taken to court and accused of slander and it could get messy for no benefit to him or anyone he is trying to reach. He's being street smart as well as book smart. That is what I applaud.
Idk if I can trust this Matt Parker guy, he has his own book prominently displayed on his bookcase Though on the other hand he does have cubes... hmm...
@@scienceworksinmysteriouswa9463 Matt's previous video about chicago addresses Bonavito's concerns as well. There's not enough variance in order of magnitude in his datasets for benford's law to be applicable. A handful of counties submitting hundreds of thousands of votes per candidate, and a handful submitting 1,000-9,999 with the rest submitting 10's of thousands means that Benford's law doesn't really apply. Also worth noting that benfords law is a red flag indicator, and not proof of wrong doing. The red flag is allowed to be raised, but once you look at the data (and to his credit, Bonavito does show his data sets) you can see exactly why you can disregard the red flag in this scenario.
I was very puzzled when you initially showed the method used to make the scatter chart and just went along with the percentages being subtracted from one another. You left me hanging for like 5 minutes, arguing in my head who was wrong, me or you!
Subtracting percentages is always a bad idea. Imagine situation where one metric goes from 1% to 2% (that is, it doubles) vs another that goes from 55% to 56%. Both changes are identical if you simply subtract the percentages!
I watch both videos...and this is a complete misrepresentation of what Dr. Shiva was showing with the graph. Shiva was showing how precinct's that voted low for Democrats, also voted high for Biden, and as you go further to the viewers right on the graph you are getting more and more Republican but also strangely more and more Biden votes...and as you go viewers left you are getting into more Democrat districts and more equal distribution of either less or more Trump votes by those Republican voters in Democrats districts. This guy is a complete liar, and thinks that he can pull the wool over your eyes because you won't look at the other video Now we know why he won't use any of Dr. Shivas video.
@@giant9833 You can't subtract percentages if they are not based on the exact same size population. That's the golden rule of percentages. If you do that your data will no long represent reality.
There's also the fact that all of the non-party votes for Biden + all of the non-party votes for Trump won't equal the total non-party votes, because there are other candidates. Therefore, even with perfect correlation you won't have a perfectly flat line of best fit.
To be fair, that data would be mostly negligible as it would amount to less than 5% of the total data in most places. But yes, it is important to keep note of because it is not negligible everywhere. Although depending on the exact information released in the voting data, it is possible you could account for it. (if they gave the votes for every possible candidate, you could subtract that from the total to get just the total votes of Trump and Biden votes only.)
Little do you know that the head of security for Dominion was a member of Antifa. Yes it sounds ridiculous but ppl have the receipts. And this whole, the data doesn't show anything this is not evidence is put to bed. But you won't accept it nor will anyone else because this isn't about looking at this accurately, its about trying to debunk it to get your way. I will await the backpedalling
Antifa makes receipts? I see them more as a hand scrawled note kind of operation. Plus they would favor trump because he gives them more excuses to riot and stir up problems. Both types of extremists need each other or else there is no contraversey to try to use to your advantage.
I watch both videos...and this is a complete misrepresentation of what Dr. Shiva was showing with the graph. Shiva was showing how precinct's that voted low for Democrats, also voted high for Biden, and as you go further to the viewers right on the graph you are getting more and more Republican but also strangely more and more Biden votes...and as you go viewers left you are getting into more Democrat districts and more equal distribution of either less or more Trump votes by those Republican voters in Democrats districts. This guy is a complete liar, and thinks that he can pull the wool over your eyes because you won't look at the other video Now we know why he won't use any of Dr. Shivas video.
@@beebait1464 I watch both videos...and this is a complete misrepresentation of what Dr. Shiva was showing with the graph. Shiva was showing how precinct's that voted low for Democrats, also voted high for Biden, and as you go further to the viewers right on the graph you are getting more and more Republican but also strangely more and more Biden votes...and as you go viewers left you are getting into more Democrat districts and more equal distribution of either less or more Trump votes by those Republican voters in Democrats districts. This guy is a complete liar, and thinks that he can pull the wool over your eyes because you won't look at the other video Now we know why he won't use any of Dr. Shivas video.
I appreciate the time and effort you have spent on this subject, I think it's in "above and beyond expectations" territory. I also appreciate the multiple copies of your own book in the background to emphasize the joke.
It's baffling why they plotted the difference but all that it proved was that people who ordinarily vote Republican voted almost a complete Republican ticket except for the president. A nonscientific observation from my own life showed me that people who voted Republican their whole lives did not vote Republican for president last time though they voted all the other Republican candidates in local and lower ranked federal offices that they normally would. I really appreciate it that you plotted the same correlation for Biden as for Trump. I wish I could think of a more interesting topic to tackle but the only other big issue is covid-19 and the statistics on it don't seem to matter much once people get tired of dealing with it.
I'm not the smartest guy so you're saying, some people might go to Trump Rallies, ware the shirts and hat tell their other Trump friends Trump all the way, but once in the Voting booth, vote for Biden and just not tell you're friends and family that you did?
It doesn't even prove that. It shows (under their stupid transformation) that people not doing straight parts votes are more likely to not vote the same party for president, than those who chose not to split their votes. Which is not very surprising, given that those who didn't split their votes, well, in fact, didn't split their votes. It says nothing about the other races by the way. The people may have plot votes in one more more races. Or select to split votes and not split it anyway. But let's be real here. They just tried to find a graph that would slooe downwards to lie about the "election being stolen" and is at least complicated enough not to stand out to everyone foe being wrong. There's zero chance that they did this in good faith. Which, by the way, is masterfully suggested by Parker without actually saying so.
@@BlackShogun1 No, he's saying people who normally voted Republican chose not to vote for tRump because they actually paid attention to the real world and see tRump for the liar and possible traitor that he is. But you guys don't actually care what the truth is, so long as you can continue to worship your orange god.
@@BlackShogun1 No. They're saying there are republicans who normally vote straight ticket republican that didn't like Trump and didn't vote for him. There's nothing about those people going to Trump rallies or wearing Trump shirts.
He does only focus on the maths, and doesn't care about anything political. So there's that. He's a great TH-camr in this regard. This is unlike Tom Scott who says you're wrong for having a certain political view and made a videe of how he was upset about the result. It's so unlike this normal content he makes.
@@Liggliluff A lot of science communicators came out against Trump, and in hindsight they were absolutely right to do so; the Trump administration has been extremely anti-science. It's not really a political opinion to express that, it's just a fact.
The rage boiling right behind his eyes from the theory he was suggesting, but not explicitly stating: Dr. Shiva intentionally used erroneous math to mislead people about an already volatile situation. The disgust he was fighting back that someone would use his favorite subject for evil was palpable. Despite this, though, he kept his cool and let his math do the arguing. Bravo!
I watched his response, and I have to say I'm still team Matt. His math was all over the place, and he relied heavily on people believing the assumption that other mathematicians were misguided because they didn't understand politics. He subtracted percentages. No kind of politics makes this good math. Likewise, towards the end of his response video, it seems he abandons this straight line theory for a curve theory instead. Perhaps, he realized he was wrong (even though he didn't admit to it).
@@MrCallidus no shivas response ignores the whole argument Parker made... Parker used the same method to plot trumps data for Biden... And this shows a slope just line with biden... Shiva argued in the beginning that this slope means that trumps vote was stolen... So it should mean the same for Biden. Now Shiva made some excuse that the Biden graph is Parker made is just the inverse of the Trump graph... But thats just an excuse, because It's actually not the inverse because you also changed
Guys, look at the comments here and elsewhere - your objections have been fully dealt with. I agree that Dr Shiva's analysis was sloppy around the edges - that's partly what made me skeptical initially - but his fundamental point is sound.
Fun little experiment I just tried: 500 lines of randomly generated numbers between 0 and 100, A & B (essentially random percentages). Like this, I plotted A against B-A and, like magic, my negative slope line emerges. It's just the nature of the way they're plotting it.
that is not the actual formula for the value they got i think he was trying to give a simplified example but i don't think it is actually a good one just causes confusion. what they did was normalize for the straight ticket vote not the percentage of the precinct that voted republican which is what their x value was, so not y-x
@@beartankoperator7950 except Matt shows that using that simple subtraction method he can directly recreate the graphs from the video - that shouldn't work, if they actually used a different method (and you can see the method Matt is using at 11:54 in the video when he goes to change the calculation)
@@steelfittermasta1715 you clearly don’t understand Benford’s law, or you are deliberately being obtuse. The key problem was with the data not spanning several orders of magnitude.
The belief that the line should be flat seems to come from an assumption that m=1. It's completely unsurprising that the party votes and candidate votes are correlated, but a two minute conversation with a party voter and a candidate voter should disabuse you of the notion that m=1.
This is why the phrase "lies, damned lies, and statistics" exists. Also the reason why you should check if your reasoning agrees with other data sets, but Dr. Shiva has a history of dishonesty.
@@tyttuut I’m not disagreeing with you, but trying to convince someone their favourite celebrity isn’t lying with a Wikipedia article probably isn’t the beat idea, haha.
I really enjoy the increase in your books and Rubik’s cubes throughout your video after you pointed out you shouldn’t trust someone that front faces their own book and that you liked that Rubik’s cubes were a props in the other video. Brilliant I love your sense of humor!!!
I am SO glad that professional mathematicians are standing up to combat the "mathemagic" nonsense that are being circulated around, even though you're not an American. Also, I really enjoy your other videos, and I always learn something new.
To be fair, it's always better to have a real topic when making educational videos on maths - the details can seem arcane so if you can give people a concrete way to understand them, so much the better
@@Ziggy_Moonglow No, it genuinely is political. In theory, everyone agrees that "we want clean water", but they don't have a single point of agreement of what that means and how much money they are willing to pay for it.
If you plot the percentage of the length of time of the video against the difference in percentages of books+Rubik's cubes and the length of time it shows that there were books and Rubik's cubes taken away and added to Shiva's video. Undeniable mathematical proof!
The amount of times I double click it at work and my colleagues go “wait, how did you do that so fast?” Or I see people painfully drag it all the way down, I have to stop them and show the double click.
Hey Rafael welcome to the club, I had a major in mathematics in university, so I have been corrupted as such for a few decades. If you can find a maths topic that clicks with you, well that is fantastic. Cheers from Canada.
If the benefit of everything that happened includes things like people taking a bigger interest in subjects like maths and science, and shows people the importance of voting, then I think that's great
@@George_Taylor_ You keep pasting that comment everywhere but keep ignoring everyone telling you what you’re going on about isn’t relevant to the issue.
"They subtracted one from the gradient of the plot, and then were amazed that the gradient had changed." That line had me laughing way too hard. And it wasn't even a joke. It was just a statement of fact.
@Yuri Bezmenov Just a question - Imagine you have a graph with a correlation line that went through the point [50,70] (Republican straight ticket percentage,Trump single vote percentage), which would be representing his "Trump more popular than party" slide, right? Now if the slope of the correlation was ~1, what would the y-value, meaning the expected percentage of trump-single-person-voters be, when we look at counties with 85%, or even 90% Republican straight ticket, to fit in with that correlation line? x would grow from 50% by 35%, meaning y also grows by 35%, from 70%. Talk about exaggerated expectations... Now you might say, he leaves out those areas greater than 80% in most of his plots, maybe he's just not telling us the limits of definition to not water down his point, right? Then just remember that these very numbers would also tell us all about Biden's performance. We'd have the correlation line on the Dem/Biden-graph at [50,30]; we got towards the 15%-Dem position; 15% is always included in his plots, right? And once more, we might realize that a result of MINUS 5% for Biden among split votes would seem a bit unexpected. He basically tells you that in f(x) = mx+b, you can expect to play around with b, but never with m. Which will always result in impossible results around the extremes as we've seen, and for which assumption he provides no proof - to the contrary, he provides lots of real world examples where m is below 1, not a single one where m~1, and expects us to believe that this is not normal.
@Yuri Bezmenov I don't even understand how they would come to that conclusion? Trump is a very divisive candidate, and there is good reason to assume socially/fiscally conservative voters might choose to vote for Republican state representatives but not Trump as president, considering the decidedly unchristian and not fiscally conservative things he's done.
@Yuri Bezmenov I can't exactly tell what you're referring to. Could you answer my example first, please? What would be the expected value ofTrump split voter percentage in counties where Republican straight tickets is >80%, if we assumed the correlation line to go through 50/70 Republican/Trump split and had a slope ~1? I think we'll have to agree that this would be an impossible prediction by that correlation line, right?
@Yuri Bezmenov Not with the "normal case"-Trump more liked than GOP candidate- slide he presents: th-cam.com/video/Ztu5Y5obWPk/w-d-xo.html As he explains before, we add the x value to the y value to derive Trump votes - and arrive at what Trump voter percentage for x=50%? I mean, we can argue where the linear regression over this sparse synthetic dataset would land - but clearly somewhere between 60% and 70%, DEFINITELY not at 50%. Again, this is a linear function f(x) = mx+b. You're saying m~1 in all legitimate cases, which would leave at least b variable, right? That's exactly what he proposes as his expectation in his Trump more/less popular than the other candidate slides. What's f(50%), for any b != 0?
And conservatives still pretend to be all about freedom of speech. It's always been that conservatives feel entitled to freedom from consequences for whatever false or bigoted things they say.
Brave (in the American sense) move to discuss maths in politics. I respect it. I also respect the "I AM A MATHEMATICIAN PLEASE DON'T SUE ME" energy throughout.
@@andrejnawoj8471 this is news to me, a Brit. Maybe the confusion is that we are condescending and sarcastic all the live-long-day? Calling stupid things "brave" in a condescending and sarcastic way until Americans think we've changed the meaning of the word.
@@andrejnawoj8471 That would only be true for British English if there was sarcastic context applied. Then again, that would also be true for American English? I've never heard of "brave" meaning "foolish"
I think it would be in the context of a person does something kind of unorthodox and you might say “That’s brave.” I’ve seen brits use it in that context but I am American so you definitely know more than I do.
Plotting the difference (as matt points out) is just a transformation of plotting the two percentages...it's another way of looking at the same trend. It's actually not a mistake to look at it this way as Matt is suggesting.
I've been hating the use of statistics to claim a conclusion where the audience don't really understand the method used. Might as well just say you get it by magic. Thank you for being neutral and informative on political problems.
Hrrm, this whole year sure has been a statistic fest. First we get all the corona charts and data sets, now it's the elections. Very cool to see mathematics get all this attention, but yes all the attention should be on the correct maths.
Funny, went over to the Shiva video and he starts off bragging about his 4 degrees and PhD. Meanwhile Matt lets the content of his video speak for itself without self aggrandisement. I wonder who more intellectually insecure out of the two of them...
I live in Silicon Valley. Trust me when I say that the people that boast the most about their academic performance and amount of certifications are the most rigid and uncreative people I've ever met. Education does not equal intelligence.
@@zoltandober Education requires opportunity. Intelligence does not. In poor countries, there are loads of extremely smart people, but few who receive education. Yet they still seem to pop out once in a while regardless. Without any "proper" education.
Lost humble points for copies of 'humble pi" multiplying like rabbits, and what I assume to be Rubik's cube "taunting," subliminally, in the background. :-)
You were awfully kind to these people, since it's pretty obvious to me they did some elaborate nonsense to generate a graph that supported their conclusion, then withheld relevant data from their viewers. Like, them saying, "Feel free to challenge our ideas!" is definitely just a bluff to gain confidence.
I think it's important in arguments to have respect to the person you're arguing with. People too often on the internet treat their opponent like an idiot.
The fact that he was kind kept me watching, and I'm probably in the group of people that would have believed the faulty conclusions on face value. I think its more effective this way at crossing divides.
I think the fact that he was so kind and showed them more respect than they deserve makes it harder for them to just dismiss this video as a “hater” or something.
10:49 "If your theories live by the spreadsheet, then I'm afraid they can also die by the spreadsheet" has got to be one of the hardest lines I've ever heard.
I watched Dr. Shiva's video first. A couple weeks later I came across Mr. Parker's critique. I wish my math was better or that I was smarter, because I still feel like I need more information. Great video Mr. Parker.
You're probably affected by anchoring bias, having watched Dr. Shiva's video first. Why should the split-ticket voters be voting in the exact same way, or exact same numbers, as straight-party voters? Also, the conclusion with the downward slope only proves that split-ticket and straight-party voters in political precinct are correlated (but not identical). The initial analysis doesn't make sense.
-"Look at me, i got two rubik's cubes, I'm smart!" . -"Look at me, there's a book with my name on it, so you can't discuss my authority" . Stand-up Math: "Ooooh, that's how it works? Got it" .
having a rubik cube on your desk does not prove you ever used them, they seem to be in exactly the same state as as the were when bought., same goes for books on the shelf, it only shows you have them but doesn't proove you read them.
I watch both videos...and this is a complete misrepresentation of what Dr. Shiva was showing with the graph. Shiva was showing how precinct's that voted low for Democrats, also voted high for Biden, and as you go further to the viewers right on the graph you are getting more and more Republican but also strangely more and more Biden votes...and as you go viewers left you are getting into more Democrat districts and more equal distribution of either less or more Trump votes by those Republican voters in Democrats districts. This guy is a complete liar, and thinks that he can pull the wool over your eyes because you won't look at the other video Now we know why he won't use any of Dr. Shivas video.
@@giant9833 Gee, maybe a lot of Republicans decided that Trump sucks, and split their ticket so they could vote for Biden. (And with more Republicans in a district, you see more more of these voters.) It's not hard to explain, if you think about it... but you have to be willing to think about it.
@@jpdemer5 Then how does Matt get the same plot for Biden...It makes no sense because Matt is misrepresenting Dr. Shivas video.. Why don't you just go watch Dr.Shivas video to compare? You won't because you like Matt too much
@@giant9833 but Shiva's graph doesn't show the precinct that voted low for democrats, Matt's graph does, and Matt's graph (based on the same data but plotted more honestly) does not show anything suspicious
@@giant9833 not indication of anything malicious going on. It could just be that republican voters dislike Trump, so they split and vote for certain republicans in Senate and Biden for president. And that this behaviour is more common among republicans then democrats, which is expected when checking how Trump behaves and really tries to divide the country. You have to think wether the assumption Shiva made is probable or not, and it just isn't. So Shiva whole argument is based on a flawed assumption. Which is the problem with persons lacking connection with reality outside their own bubble.
Me, someone who's not nearly as good at math as I should be: I'm not really sure, but I think I kinda get what Matt's saying *Matt shows the equation of a line* Me: I know what m and b represent! Yeah, I'm smart
The best thing for me in the video was to find out that in Excel you can double click the edge of the cell to replicate the formula to whole stack! THANK YOU!
Worth mentioning that the reason the slope is less than 1, ie the reason the difference is negatively correlated, is likely that Candidate votes will tend to lean the same way as the precinct overall, but will lean less strongly one way or the other because people choosing to vote in each race separately are more likely to be more centrist (since they presumably want to select candidates from a mix of parties across the different races). The more heavily the precinct leans toward one party, the more the candidate voters will deviate from the party voters on average because those are the more centrist voters.
It seems like this is the actual refutation of the original argument, and I'm still not sure how Matt's addresses it exactly. They were expecting a flat line, which IS what you'd get if you subtract 1 from the slope of a y=x line (which they expected, if they think the same vote split among party/nonparty voters). What Matt shows here seems (though I expect I'm wrong) to sort of "transpose" the problem. The slope is still more negative than they claim to expect. In one graph it's negative instead of flat, and in the "fixed" on it's less than 1 instead of 1. What am I missing?
@@binarysmile As others have pointed out, the Issue is they were assuming y=x which was false. Matt DOES address this in the video. The number of people who voted party line was NOT equal to the number of people who individually chose their candidates. What the original video did was assume that in every county, the number of people who voted party line and the number of people who individually choose their candidates were equal, which was factually wrong. That is why the data appeared that way. It is ALSO why Biden's graph of the same data had the same exact issue present.
Actually impressed beyond belief at your neutral, impartial and balanced stance in such a nasty politically charged mess. Love the man who loves the maths. Keep it clean and brilliant, and don't stop with the beautiful critical thinking and anyone-can-do-it explanations. Don't ever change!
A neutral person would be fuming at the threat to democracy being levelled by a narcissist. I cannot muster this level of placidness in the face of something so wrong.
@@wizard7314 nice narrative you got there. A neutral person would say "calm tf down Trump but sure let's see if you have any sound arguments and evidence" they shouldn't just presume that he's an evil fascist dictator who hates democracy and he can't POSSIBLY be right.
@@thisaccountisdead168 It's called projection. They watch Fox and Newsmax and Breitbart nonstop and regurgitate their claims uncritically, so *obviously* anyone who disagrees with them must be doing the same thing but with different news sources.
Statistics are constant, the problem lies in the flexibility of being able to pull them from misleading base data, or display them in a misleading way.
@@elmerbeltshire7599 "Facts: Google-TH-cam pushed this to the top of people's feed to make sure everyone saw it. " A trumpist's last excuse if all others fall on deaf ears.
@Vinny Mac Only in the minds of crayon eating illiterate goons. For people who understand mathematics, the claim that he makes that what he is doing isn't maths is the final proof that emperor is naked. Not only is it explicitly maths, he is making mistakes that you teach 17/18 year olds not to make...
"I've debunked a thing and I got forty-odd comments saying “Too easy, debunk this!”, so here I am!" Oh no. Oh no, I've seen this before. Don't do it, Matt. You'll be spending way too much time chasing the shifting goalposts. I'll trust you that this plot is fascinating (haven't watched til the end yet), but please just ignore those comments, for sanity's sake.
@@durnsidh6483 then don;t keep chasing when they arn't running through interesting plots. it not hard problem to solve keep your eyes on the prize. an excuse to talk about plots that people care about.
@@Royvan7 Fair point, I just worry that this will turn into a "Matt vs. The Conspiracy Theorists" slug fest. Though if he keeps his eye on the prize as you said, he'll be fine.
Thank you Matt, I'm far from a Biden supporter, but I just had an online discussion where this video was brought up and exactly conjectured that this "anomaly" would also show up in Biden's results. There was no explanation for why one would expect a horizontal line in those graphs and no comparison to voting data of Biden.
@Benjamin Maiorella a clear red herring used to stoke their base. It's no coincidence every single suite they pushed was strongly put down for lack of evidence and may of those by republican judges, and the two lawfirms decided to bail because they got nothing to bring forward. But sadly their base does not care and necer will. That being said it's quite scarry even from Canada seeing the road the GOP is taking right now this is an unprecedented attack on democracy and in the US of all places.
Congrats, you failed the test! Yes, Matt didn't claim it's a mystery, I claimed Matt's underhanded manipulation and misrepresentation of results is a "mystery" (but not really). His explanation is just plain wrong, there's much simpler mathematical explanation why they look the same, and nowhere Matt even hinted at it. But I'll give you another hint it's because those graphs are strongly dependent, to the point they are bound to have the same slope, no matter what manipulations were used in counting. But lets try again, If I were to plot 1-gop% against 1-trump% , I would get 180 degree rotation of Trump's graph, but what else? What that graph would show? Could that result somehow correlate with Biden's graph in two party system?
Since you've probably bailed, I'll tell you what it means, the plot would show not Trump votes vs not Republican votes, in two party system like US, vast majority of those vote for dems and Biden. Therefore Biden's graph is extremely close to 180 degree rotation of Trump's, THAT'S WHY THEY LOOK THE SAME. They would always look the same no matter what Russian hackers would do! Now, why would supposedly apolitical math enthusiast keep silent about that fact and pretend as if it wasn't simple 180 degree rotation? This is the "mystery" I was talking about. But since I have to spell everything out, it's because: a) Matt made a politically motivated lie under guise of math b) Matt doesn't understand simple math concepts like rotation transformation and what it does to linear graphs. c) a and b.
ITT: People not getting the joke when he says not to trust someone with their book on a shelf facing out.... And then adds more and more books and rubik's cube behind him for the entire video.
Thanks for pointing that out. That's hilarious.
Poe's Law, it's inevitable...
and he added some Rubik's cubes xD
@@jonathankay849, what's Poe's Law?
Came here for this comment!
There seems to also be a positive correlation between number of Humble Pi books and time elapsed in video 🤔
Along with Rubik’s cubes
But if you subtract the number of rubik's cubes.... :)
Waiting for a scatterplot of this data to be posted
Would be interesting to see this transposed onto a spreadsheet...
y+n vs t+ 9:13 where n=(#HumblePi) and t=time
I loved how he said not to trust people filming in front of a bookcase with their own book facing out. And as his video progressed, more and more of his books were facing out. I love this guy.
Also more and more magic cubes
Also rubix cube-like things
*cut to the wide shot*
You can tell he's not as smart as the other guy because the Rubik's cubes aren't solved.
@@nickguman192 Do that mean that he's getting more and more Rubik's cubes with the assumption that previous one was broken because it was hard to solve?
His reveal of having all his own book facing out surrounded by Rubik's cubes after saying he couldn't trust someone who does that made me laugh out loud. Well done Matt!
The bookshelf had me dying.
The cubes and humble pi's were replicating.
The trouble with tribbles
Just the cheer amount of copies of Humble Pi facing out. totally lost it.
That was a good touch on his part
Loved it! 😄
We'd better hurry and buy some then.
I mean, honestly, who trusts a guy with just ONE copy of his book on the bookcase, facing out?
I lost it when I realized that he had 3(4?) of his own books facing out 🤣
EDIT: OMG THEY'RE MULTIPLYING. I'M DYING 😂
@@zachk7305 at the end there are 7 of them 😂
@@dargtagnan3696 I count 11. 4 on the top row, 5 on the middle, and 2 on the bottom. (All are Humble Pi except for Things to Make in the Fourth Dimension on the second row.)
@Jon Lauer it was a joke... in fact,he adds more and more throughout the video
LOL!
1:49 3 Rubik's cubes and 2 Humble Pis
8:41 5 Rubik's cubes and 2 Humble Pis
9:34 6 Rubik's cubes and 4 Humble Pis
13:58 10+ Rubik's cubes and 7 Humble Pis
16:24 the whole shelf is now just Rubik's cubes and Humble Pis
well played sir 👏👏👏
Awesome counting. I was hoping for Fibonacci sequence, but no dice.
At the very end there's 9 Humble Pi copies. And one copy of Things to Make and Do in the Fourth Dimension.
+1 Things to Make and Do in the Fourth Dimension
This is a special kind of annoying that I admire.
I counted 8 red cover books at the end, the last one's behind his head but it sneaks out occasionally :) I only noticed the change at the end lol
16:14 They took the gradient, subtracted 1 from it, and were amazed that the gradient had changed! LOL
Surprised pikachu face
Not only that, a gradient less than one becoming negative.
There is another way to describe what they showed -the result shows that voters who vote by candidate are more likely to dislike the presidential candidates.
Thats how data works. What blockchain do you even use to secure your identity or data? probably none.
No Blockchain No peace
Doesn't it point to an irregularity when you see that the county was "more republican" and the inversion greater?
"I don't trust a person with a bookcase as a background with their own book facing outward" -Matt Parker, author of humble pi. Lol 😂
Edit: I just noticed over the course of the video he adds more copies of his book. Absolute based legend.
And also the number of Rubik's cubes increase as well
Based on what?
@@filipsperl he's using the reference to the video he shows stills from and expanding on it, also he mentions that he doesn't mention his book... While knowing he is totally mentioning his book
Has anyone plotted the number of books and Rubik's cubes against video runtime?
Based on science ;-)
“Live by the spreadsheet, die by the spreadsheet” is my new motto
+
Same
That's going to be my new tramp stamp. I want anyone who's looking back there to know what's up.
@Julian LSA and it should be in that fake "heavy metal" appearance too
Now we also need some math variation of "understanding is a triple edged sword".
I was literally thinking why they subtracted two different percentages that aren't equal in populations. Thank you Engineering Statistics in college for this omen lmao
I don’t even have a background in stats and although I couldn’t place why I also realized that something was off about the subtracting of the two percentages. It just felt kinda wrong but I wouldn’t have been able to tell you why.
The reason is that you must weight averages before performing any aggregate function on them. A percentage is essentially an average. An extremely common mistake that statisticians shouldn't make.
@@totovader actually it wasn’t math that had my alarm bells going. In a republican precinct what are reasons someone would not check the box for all republicans? They didn’t want t vote for a specific republican. The most main republican is the president. So as someone who still smells some fish but am more skeptical currently of “stollen” election or full coup that is what went through my head.
@@totovader to be fair, it's only honest statisticians who should be expected to not make such an error.
Cue the Mark Twain quote...
Same here... I tripped over that step in his explanation of their video because it makes no sense why you’d do that. Glad it wasn’t me being a noob.
The strangest part is that he tried to get away with claiming that a line that clearly slopes down started out horizontal.
Better than that, as Naim Kabir pointed out in an article on Medium, some of his predicted straight lines involve points which are impossible. The headline of the article involves the words "Doubles Down on his Con" so Naim was even more brutal in describing Shiva that Matt was!
It makes no sense. It's not some time dependent or experimental prediction thing; that's just a segment of the data.
@@ishoottheyscore8970 Nauim Kabir can probably afford to call Shiva out. Matt has an educational mathematics channel to protect against rabid idiots.
@@57thorns true that, this is about as apolitical as you can get while dis/proving a political argument - unfortunately less referenced bc its less inflammatory, but a better video for it.
17:26 "I'm meant to make silly, fun, educational, mathematical videos"
Mate, using maths to help people understand election results and the debate about them is exactly what the world need right now.
+
Unfortunately, Matt's math is wrong here for a number of reasons.
@@LukeSchiefelbein can you elaborate?
@@LukeSchiefelbein How so?
@@LukeSchiefelbein Would love to hear your basis for this claim
"If your theories can live by the spreadsheet, then I'm afraid they can also die by the spreadsheet" was a much more cool and powerful sentence than i ever expected any sentence containing the word "spreadsheet" to be. hell yeah
Never underestimate the power of a nerd with a little time and a spreadsheet.
Matt's theories need to die by the spreadsheet too. This video is 100% wrong for many reasons
imgflip.com/i/4mo61f
@Bobby Jones No I think Matt's made an honest mistake. Yes his math is completely wrong, but it's kind of hard to spot. Shiva is 100% wrong too. Meanwhile both sides are just parroting misinformed talking points and yelling at each other. I
@Bobby Jones It's about selling lies to people who are grasping for any excuse to believe that their fantasy world is real.
“I don’t know if I can trust someone with their own book face out” you dog you. Love how you managed to fit some Rubik’s cubes in there too. Taking troll to a whole other level.
@ oooopsie, someone forgot to take its medication? ;-)
@CapnAce79 wait 1 more hour then maybe you tiny brain will get it
Kung-fu figthing!!!
"I've found the evidence supporting my hypothesis, I can stop looking."
Can you say cherry picking? Lol
@@kimberlymonsini2604 Bite your tongue, madam!
@@inyobillBahaha.
One of the more important lessons that I teach my students, the unscrupulous action of cherry picking data or information has not only ended relationships, it' has also started wars!
Geesh, next thing they are going to tell us is that popsicle sales are the cause of the spikes of drownings!
Really? I thought it was more of a case of "I have a hypothesis, where can I find evidence to support it? I can't? I'll make some then."
Or was I the only looking at the initial graph thinking "They're using WHAT for x and y axis? Why?"
@@tskmaster3837 One of my pet peaves, unlabelled axes.
The fact that the negative gradient can be explained so easily is kind of funny
It's almost like he makes a convoluted argument, in order to explain the obvious. Never argue with an idiot, unless you can rebut every detail.
It would be funny if it wasnt a sign that people are starving to classify themselves as victims and the oppressed.
@@gordonrichardson2972 Important corollary: You can never rebut every detail because there is always one more. After rebutting arguments for hours and hours, eventually you will have to stop to eat or sleep or go to work. Whatever argument you didn't get to is now the clincher that proves the idiot right. I learned this lesson the hard way from arguing with moon landing conspiracy theorists.
@@JohnnyAdroit True. Fortunately the election results are able to be plotted, which rebuts the original argument (though not all possible theories).
@@JohnnyAdroit and even still, if you could hypothetically rebut every single detail, both you and the person you're arguing with would have to agree to every single definition of every single word you use, which is just bonkers
almost sounds deliberate, how could they possibly "forget" the second part of the two part graph
@Yuri Bezmenov No it does not. Much like the first part, it only shows what happens if you wrongly subtract percentages.
Remainder not to feed the troll.
@@gags730 why wouldnt he plot those himself to show the anomaly?
@Yuri Bezmenov Yeah, I don't trust people who call correlation coefficients "strong R" or quantify them as "lowered". Take a stats class before you embarrass yourself any further.
P.S. I've interpreting a very vague comment incorrectly. Unfortunately for Yuri, the spirit of comment still lives since he hasn't really shown any evidence.
@@gags730 Indeed, go ahead and plot them yourself. All the data is there and Mat and Dr Shiva have given you the tools to do so. Mat wants to be proven wrong (I am not so sure about Dr Shiva though).
I was a bit frustrated at first that I couldn't understand the beginning of this video. Now I get why I was confused: It is a video explaining why something that doesn't make sense, doesn't make sense lol.
I feel you so much.. the explanation of the plot is so confusing and weird.. turns out.. it's because it makes 0 sense ahhaha
Trumpers here looking for proof and coming out even more confused
@@George_Taylor_ The y axis is just the % of votes that each candidate got. Biden no having any data points below 0.3 just means that he didn't get less than 30% of the vote in any off the districts
@@George_Taylor_ Dots being more tightly clustered on a graph with a larger scale only makes sense because objects look closer together when you zoom out. If it was at the same scale, the distance between dots would look similar on both graphs.
@@George_Taylor_ i didnt notice that, what does the tight clustering imply?
Isn't a bigger issue the hypothesis that the straight ticket voters should be identical to the ticket-splitters in their preferences? Presumably people split their tickets to vote against someone on the ticket of the party they otherwise prefer.
Exactly - Shiva's plot is a reference to his own prediction, not proof of anything (beyond his prediction not meeting reality). As you've said, this prediction of %president-%party=0 isn't exactly compatible with logic!
Exactly. We can assume that if someone didn't elect to vote straight ticket then they split their ticket. Meaning their was some Democrat on that ticket they elected to vote for instead.
I dont have a source for this, but I believe it's not uncommon for people to vote for conservative representatives and liberal presidents.
@@Draenal ,
The Lincoln Project wanted Trump Gone... but would otherwise vote Republican on the rest of the ticket... because they only want Trump Gone... as they are Republicans.
I live in Kent County, there is a very large percentage of folks who vote one party for president and the other party for senate/rep.
@@Draenal It may not even be about voting for the opposing side, it could be abstaining from a particular race if you feel you couldn't back either candidate. There was an Economist/YouGov survey that get mention in an article about a Republican Governor endorsing Biden (Charlie Baker maybe?) where 16% of Republicans and 11% of Democrats stated they wouldn't back their candidate in the presidnetial race. It also mentioned 3% of Republicans choosing to vote for Biden, and 1% of Democrats choosing to vote for Trump, but it's not clear from how it was written if they were already part of the 16 and 11 per cents mentioned earlier
And none of this offers any statistical method to account for people who aren't affiliated to a party, just to scuff up the linear prediction that bit more...
I grew up and I still work in Kent county, I feel like I’ve basically made it into one of Matt’s videos! Love the bookshelf and cube troll!
You’re famous now!
SAME!
Basically, either these guys had no idea how to do the math properly, and therefore can’t be trusted; or they knew *exactly* how to do the math properly, and did it improperly on purpose so they could argue a faulty conclusion, in which case they *definitely* can’t be trusted.
I hope is the first, but having his Wikipedia page suggesting he is a very competent conspiracy theory spreader, I incline to think it's the second.
The problem is that math often isn't easy to understand so people will go with "Trump will win, a mathematician has already proved that votes are wrong everywhere".
This guy would have to flop the Biden graph and put the strongest blue precincts to the right to be fair. He didn't. When you do, Biden goes up. Not down.
so do you maybe think they said here is the evidence now give me my million dollars?
I think they definitely did it on purpose- look at that bent "best fit" line!
Geez, subtracting percentages -- what an embarrassing mistake.
I immediately knew there was a problem with the data once I saw it
As noted in this vid, you can do that if the set sizes were equal-and Dr Shiva specified they were. So, embarrassing mistake on you part? However a problem remains as far as the claim of equal set sizes.
the "mistake" was probably done on purpose tbh
@@jrettetsohyt1 The set sizes weren't equal though 13:14
@@jrettetsohyt1 - I think you can kind of do it when you know the 'weight' of the percentages, there is such a thing as a weighted mean, so in theory you could try reverse engineering that calculation, but would it prove something, or just be someone messing around with numbers?
Took me a bit to notice your book multiplying in the background.
He also had rubik cubes
Dear God, the books are multiplying.
@Centauri It’s natural, but he should really get them spayed.
I think there is a risk of them going "viral"
Propagation is only natural and is to be expected...
Propagation is only natural and is to be expected...
Propagation is only natural and is to be expected...
Propagation is only natural and is to be expected...
Propagation is only natural and is to be expected...
Propagation is only natural and is to be expected...
Atleast they're not mutating and heating up the planet
"Multiplying"?
I'm sure if he put a box of condoms up there it'll stop.
This is the most gentlemanly roast I've seen
I've seen academics get far more vicious over way less important things, so I was happy to see Matt not completely rip into them
@@George_Taylor_ that scale difference is very small and likely the scale the graph generates with to fill the space it was allocated. yes the data points are more tightly clustered but this is fairly minor and this proves nothing
@@George_Taylor_ I’m unsure whether you’re trolling or not, but in the case you’re sincere, I’d like to point out the simple fact that any scatter plot will get *less* clustered as you zoom in. Because Biden’s plot is “larger” (i.e. the grid behind is larger due to a smaller range of values on each axis), If you were to resize Biden’s plot to be on the same scale as Trump’s, it would look even more tightly clustered than it does in the video.
But isn't the point of this graph is to show it's the same? Both slopes are having gradient of > 1
How many times have you spouting that exact bullshit on this video?
And this is why every single person needs more education in statistics. To be able to spot inconsistencies in data and to spot when they are being misled using data is of vital importance to literally everyone.
I studied statistics in college... That was first semester of 2020, so I dare say it was a timely education...
A good publication on this is "How to lie with Statistics" by Darrell Huff who explains in simple terms the tricks people use to lie with statistics (including different types of graph).
misled 😜
One thing to think about is that when we know the percentage of people voting for one candidate, we approximately know the percentage of people voting for the candidate. This is not like flipping a coin where we would know this exactly since there are more than 2 candidates in the race. The consequence of the 2 percentages being related mathematically means that when we create a plot for 1 candidate using the percentage, the other plot is mathematically determined. Briefly, it would be more surprising if the Trump and Biden scatter plots didn't show the same trend.
Here is a brief video explaining why the 2 scatter plots have to be similar. th-cam.com/video/uMApwgGFVlw/w-d-xo.html
@@Tokahax I did spot Matt's switch of y-scale when comparing Trump's vs Biden's slope. This is a common tactic used to misrepresent the similarities between two charts. Matt did mention that Biden's slope was steeper in the audio, but that doesn't carry over visually in the charts and is therefore misleading.
I love how my initial thought about it being strange that they subtracted percentages ended up being right.
It's not even the abstracting percentages part (which is wrong). But also that they're doing subtraction at all.
yes, me too. & I am merely an engineer.
"If your theory lives by the spreadsheet it can also die by the spreadsheet." This must be made into a t-shirt!
Live by the spreadsheet
Die by the spreadsheet
Spreadshirt
As far as I can see the two scatter plots at the brgining of the video show exactly what Dr Shiva was describing. The ratios according to the plots are; for trump +10% in democrat areas to -30% in Republican areas,whereas Biden is +30% in Republican areas to -10% in democrat areas. Isn't this exactly what Dr shiva was saying? Biden is getting up to 30% more than expected in Republican areas and Trump is getting up to 30% less than expected in Republican areas.
You should go to Dr Shiva's channel.
Due to Matt's video, he is trashing him now.
He called him "an adolescent math learner" hahaha
It's already a Field Mantra!
Your restraint and respect to people who could justifiably be called charlatans is admirable and a good example to us all, Mr Parker. 👍 Thank you for being the hero we need rather than the demagogue we deserve. 🙏🤝
This touched me. Couldn't have said it better myself.
I came here to say "people say Canadians are polite, but this was an Aussie saying 'you guys are idiots' " 🙂
He's just being careful, as he said. If he starts making statements like that he could be taken to court and accused of slander and it could get messy for no benefit to him or anyone he is trying to reach. He's being street smart as well as book smart. That is what I applaud.
@@pmcgee003 Spoken like someone who has never spoken to a real Australian. Lmao
@@RayJnocap spoken like a Westy.
Idk if I can trust this Matt Parker guy, he has his own book prominently displayed on his bookcase
Though on the other hand he does have cubes... hmm...
You know you can trust him because his rubik's cubes are actually used and unfinished, unlike those unused, finished cubes in the other video
I guess that means "Trust, but verify"
But Matt's face is not on the book.
He talks to much to get his point across, seems like he likes to hear himself speak
@@scienceworksinmysteriouswa9463 Matt's previous video about chicago addresses Bonavito's concerns as well. There's not enough variance in order of magnitude in his datasets for benford's law to be applicable. A handful of counties submitting hundreds of thousands of votes per candidate, and a handful submitting 1,000-9,999 with the rest submitting 10's of thousands means that Benford's law doesn't really apply. Also worth noting that benfords law is a red flag indicator, and not proof of wrong doing. The red flag is allowed to be raised, but once you look at the data (and to his credit, Bonavito does show his data sets) you can see exactly why you can disregard the red flag in this scenario.
I was very puzzled when you initially showed the method used to make the scatter chart and just went along with the percentages being subtracted from one another. You left me hanging for like 5 minutes, arguing in my head who was wrong, me or you!
Bring me laugh:)
Same, I literally thought at the beginning "Why wouldn't you just plot one over the other? Matt why are you doing this!?"
Subtracting percentages is always a bad idea. Imagine situation where one metric goes from 1% to 2% (that is, it doubles) vs another that goes from 55% to 56%. Both changes are identical if you simply subtract the percentages!
@@MikkoRantalainen Also, 1% of 2000 is larger than 10% of 100.
Super impressed that you didn't mention your book at all.... Crazy how it didn't make it into this video in Any way.
Once again, they fell for one of the classic blunders - never go up against a mathematician when STATISTICS is on the line!
Nice reference, sir! Incoceivable.
@@noamz9527 You keep using that word. I do not believe it means what you think it means.
The main thing I learnt from this video is that you can double click the corner of an Excel cell instead of dragging it all the way down.
The main thing I learned from this comment is that you can double click the corner of a cell instead of dragging it down.
Agreed! This will save a not insignificant percentage of time in my life.
I use that feature all the time, it's quite a time-saver!
@@helderboymh hahahhahaahaa 😄😂😂😂😂😂😂😂😂😂
It's called Flash Fill.
Love what you did with the books in the back. Gotta like just for that
I take it you're contractually forbidden from using brown kraft paper in the Working Out Zone if you don't have Brady behind the camera.
Lol 🤣
Well, he could, but then Matt's body would be found the next day with an oddly cricket bat shaped bruise on his head
@@AnirudhGiri You don't mess with Brady "Tough as nails" Haran.
I watch both videos...and this is a complete misrepresentation of what Dr. Shiva was showing with the graph. Shiva was showing how precinct's that voted low for Democrats, also voted high for Biden, and as you go further to the viewers right on the graph you are getting more and more Republican but also strangely more and more Biden votes...and as you go viewers left you are getting into more Democrat districts and more equal distribution of either less or more Trump votes by those Republican voters in Democrats districts.
This guy is a complete liar, and thinks that he can pull the wool over your eyes because you won't look at the other video
Now we know why he won't use any of Dr. Shivas video.
@@giant9833 You can't subtract percentages if they are not based on the exact same size population. That's the golden rule of percentages. If you do that your data will no long represent reality.
There's also the fact that all of the non-party votes for Biden + all of the non-party votes for Trump won't equal the total non-party votes, because there are other candidates. Therefore, even with perfect correlation you won't have a perfectly flat line of best fit.
Like comparing apples versus oranges, and then plotting slices of apples versus slices of oranges, and forgetting to label the axes on the chart.
trump landslide :)
To be fair, that data would be mostly negligible as it would amount to less than 5% of the total data in most places.
But yes, it is important to keep note of because it is not negligible everywhere.
Although depending on the exact information released in the voting data, it is possible you could account for it. (if they gave the votes for every possible candidate, you could subtract that from the total to get just the total votes of Trump and Biden votes only.)
Little do you know that the head of security for Dominion was a member of Antifa. Yes it sounds ridiculous but ppl have the receipts. And this whole, the data doesn't show anything this is not evidence is put to bed. But you won't accept it nor will anyone else because this isn't about looking at this accurately, its about trying to debunk it to get your way. I will await the backpedalling
Antifa makes receipts? I see them more as a hand scrawled note kind of operation. Plus they would favor trump because he gives them more excuses to riot and stir up problems. Both types of extremists need each other or else there is no contraversey to try to use to your advantage.
“I wouldn’t trust anyone who has their own book face out” lol love it.
I think it's also a matter of what message the book has.
@@ehsnils Look in his background. He set up a ton of his own books to face forward in what I think is a pretty funny and subtle self-own Parker style.
I felt, perhaps unreasonably, very excited for spreadsheet time.
Spreadsheets are like a warm blanket a cold night
"Prove us wrong"
Matt: "So you have chosen... death.
... by spreadsheet"
I would happily die this way. May the spreadsheets always prevail!!
I appriciate this comment. Also regardless their conclusion they drew, their argument and conclusion seemed to be blinded by confirmation bias.
As soon as I read this I laughed way more than I should 🤣🤣🤣
I watch both videos...and this is a complete misrepresentation of what Dr. Shiva was showing with the graph. Shiva was showing how precinct's that voted low for Democrats, also voted high for Biden, and as you go further to the viewers right on the graph you are getting more and more Republican but also strangely more and more Biden votes...and as you go viewers left you are getting into more Democrat districts and more equal distribution of either less or more Trump votes by those Republican voters in Democrats districts.
This guy is a complete liar, and thinks that he can pull the wool over your eyes because you won't look at the other video
Now we know why he won't use any of Dr. Shivas video.
@@beebait1464 I watch both videos...and this is a complete misrepresentation of what Dr. Shiva was showing with the graph. Shiva was showing how precinct's that voted low for Democrats, also voted high for Biden, and as you go further to the viewers right on the graph you are getting more and more Republican but also strangely more and more Biden votes...and as you go viewers left you are getting into more Democrat districts and more equal distribution of either less or more Trump votes by those Republican voters in Democrats districts.
This guy is a complete liar, and thinks that he can pull the wool over your eyes because you won't look at the other video
Now we know why he won't use any of Dr. Shivas video.
I appreciate the time and effort you have spent on this subject, I think it's in "above and beyond expectations" territory. I also appreciate the multiple copies of your own book in the background to emphasize the joke.
Yup, could have just called them morons (which would have been correct), but took the high-ground and explained everything in great detail.
I was convinced by Dr Shiva’s presentation. Thanks for unraveling this for me. You’ve earned yourself a subscriber!
It's baffling why they plotted the difference but all that it proved was that people who ordinarily vote Republican voted almost a complete Republican ticket except for the president. A nonscientific observation from my own life showed me that people who voted Republican their whole lives did not vote Republican for president last time though they voted all the other Republican candidates in local and lower ranked federal offices that they normally would. I really appreciate it that you plotted the same correlation for Biden as for Trump. I wish I could think of a more interesting topic to tackle but the only other big issue is covid-19 and the statistics on it don't seem to matter much once people get tired of dealing with it.
Simpleton, wake up
I'm not the smartest guy so you're saying, some people might go to Trump Rallies, ware the shirts and hat tell their other Trump friends Trump all the way, but once in the Voting booth, vote for Biden and just not tell you're friends and family that you did?
It doesn't even prove that. It shows (under their stupid transformation) that people not doing straight parts votes are more likely to not vote the same party for president, than those who chose not to split their votes. Which is not very surprising, given that those who didn't split their votes, well, in fact, didn't split their votes.
It says nothing about the other races by the way. The people may have plot votes in one more more races. Or select to split votes and not split it anyway.
But let's be real here. They just tried to find a graph that would slooe downwards to lie about the "election being stolen" and is at least complicated enough not to stand out to everyone foe being wrong.
There's zero chance that they did this in good faith. Which, by the way, is masterfully suggested by Parker without actually saying so.
@@BlackShogun1 No, he's saying people who normally voted Republican chose not to vote for tRump because they actually paid attention to the real world and see tRump for the liar and possible traitor that he is. But you guys don't actually care what the truth is, so long as you can continue to worship your orange god.
@@BlackShogun1 No. They're saying there are republicans who normally vote straight ticket republican that didn't like Trump and didn't vote for him. There's nothing about those people going to Trump rallies or wearing Trump shirts.
Matt: I should avoid getting into foreign politics on my TH-cam channel
Also Matt: Someone is using Maths wrong on the internet!
Righting wrong maths on the internet is the most noble of quests.
He does only focus on the maths, and doesn't care about anything political. So there's that. He's a great TH-camr in this regard. This is unlike Tom Scott who says you're wrong for having a certain political view and made a videe of how he was upset about the result. It's so unlike this normal content he makes.
@@Liggliluff There's nothing wrong with a TH-camr expressing their political opinion.
@@Liggliluff A lot of science communicators came out against Trump, and in hindsight they were absolutely right to do so; the Trump administration has been extremely anti-science. It's not really a political opinion to express that, it's just a fact.
He’s careful not to express any opinions. Just does the math!
It feels like they failed to understand the "Missing dollar riddle", and are using nice graphs to back it up.
Well said, sir.
n1
Matt never needs to print more copies of his book or buy more rubik's cubes, He just puts them on his bookshelf and they multiply naturally.
You can't multiply pi, that'd be irrational.
Spontaneous generation-it does exist
lol, i didnt even notice.
Here is Dr Shiva's answer for the arguments of this video (between 15:30-21:30)
th-cam.com/video/umNbSvqozwQ/w-d-xo.html
The rage boiling right behind his eyes from the theory he was suggesting, but not explicitly stating: Dr. Shiva intentionally used erroneous math to mislead people about an already volatile situation. The disgust he was fighting back that someone would use his favorite subject for evil was palpable. Despite this, though, he kept his cool and let his math do the arguing. Bravo!
See Dr Shiva's response. I thought Matt was right too until I saw Dr Shiva kick his butt to kingdom come.
I watched his response, and I have to say I'm still team Matt. His math was all over the place, and he relied heavily on people believing the assumption that other mathematicians were misguided because they didn't understand politics. He subtracted percentages. No kind of politics makes this good math. Likewise, towards the end of his response video, it seems he abandons this straight line theory for a curve theory instead. Perhaps, he realized he was wrong (even though he didn't admit to it).
@@MrCallidus no shivas response ignores the whole argument Parker made... Parker used the same method to plot trumps data for Biden... And this shows a slope just line with biden... Shiva argued in the beginning that this slope means that trumps vote was stolen... So it should mean the same for Biden.
Now Shiva made some excuse that the Biden graph is Parker made is just the inverse of the Trump graph... But thats just an excuse, because It's actually not the inverse because you also changed
@Brian Carlile I work in statistical neuroscience, and so use mathematics every day.
Guys, look at the comments here and elsewhere - your objections have been fully dealt with. I agree that Dr Shiva's analysis was sloppy around the edges - that's partly what made me skeptical initially - but his fundamental point is sound.
Hard to think Shiva made his argument in good faith. Anyone willing to sue someone who disagrees with them is usually a grifter.
im gonna sue
@@loluser124 Ok mr. grifter
@@Glorc72000 You saying "ok" to them means you disagree with me
-Accept- Expect a call from my edit:parents lawyer
@@scienceworksinmysteriouswa9463 Dude. Literally, the previous video Stand-up Maths made addresses this.
@@scienceworksinmysteriouswa9463 That's about Benford's law, which Matt already covered in the previous video.
We need “live by the spreadsheet die by the spreadsheet” merch
I would like this comment, but the no. of likes is at 69 and i can't bring myself to change this...
We really do!
Fun little experiment I just tried: 500 lines of randomly generated numbers between 0 and 100, A & B (essentially random percentages). Like this, I plotted A against B-A and, like magic, my negative slope line emerges. It's just the nature of the way they're plotting it.
his previous vid argued that elections don't plot as random data..hence benford law don't apply vid, he mentions that in the beginning
@@steelfittermasta1715 I think they meant that no matter what you put in, it will always result in a negative slope.
that is not the actual formula for the value they got i think he was trying to give a simplified example but i don't think it is actually a good one just causes confusion. what they did was normalize for the straight ticket vote not the percentage of the precinct that voted republican which is what their x value was, so not y-x
@@beartankoperator7950 except Matt shows that using that simple subtraction method he can directly recreate the graphs from the video - that shouldn't work, if they actually used a different method (and you can see the method Matt is using at 11:54 in the video when he goes to change the calculation)
@@steelfittermasta1715 you clearly don’t understand Benford’s law, or you are deliberately being obtuse. The key problem was with the data not spanning several orders of magnitude.
The belief that the line should be flat seems to come from an assumption that m=1. It's completely unsurprising that the party votes and candidate votes are correlated, but a two minute conversation with a party voter and a candidate voter should disabuse you of the notion that m=1.
Lincoln project is republican on everything except trump
m=1 would not make a flat line though.. a flat line would be m = 0
@@koffieenthee1156 yeah, but (m-1) =0 when m =1. So if there was a one-to-one correlation the graph they used should be flat, as now y=x, so y-x =0.
This is why the phrase "lies, damned lies, and statistics" exists. Also the reason why you should check if your reasoning agrees with other data sets, but Dr. Shiva has a history of dishonesty.
@Chairman Mao Poe's Law or...
@Chairman Mao does claiming he won the election count as dishonesty?
@Chairman Mao "Trump has never been dishonest" is a blatant lie. en.m.wikipedia.org/wiki/Veracity_of_statements_by_Donald_Trump
@@tyttuut I’m not disagreeing with you, but trying to convince someone their favourite celebrity isn’t lying with a Wikipedia article probably isn’t the beat idea, haha.
@Chairman Mao "trump has never been dishonest" made me laugh
I really enjoy the increase in your books and Rubik’s cubes throughout your video after you pointed out you shouldn’t trust someone that front faces their own book and that you liked that Rubik’s cubes were a props in the other video. Brilliant I love your sense of humor!!!
I am SO glad that professional mathematicians are standing up to combat the "mathemagic" nonsense that are being circulated around, even though you're not an American. Also, I really enjoy your other videos, and I always learn something new.
That "doctor" Shiva needs his title taken away from him
I actually think him not being American is actually an advantage because he doesn't personally have any stake in whoever wins.
@@1TakoyakiStore Oh, you'd be surprised at Australians *cough cough* Sky News Australia (worse than Fox News)...
@@jonathankay849 Oh that's interesting. I wonder why they would care?
To be fair, it's always better to have a real topic when making educational videos on maths - the details can seem arcane so if you can give people a concrete way to understand them, so much the better
Sorry Matt, but believing water should be clean *is* political in the US.
If you can deny it or claim it, then it's a political issue. That's how politics work in the US
Nope, it's just that idiots believe the 'other guys' don't want clean water so they make it political.
@@Ziggy_Moonglow No, it genuinely is political. In theory, everyone agrees that "we want clean water", but they don't have a single point of agreement of what that means and how much money they are willing to pay for it.
The good news is that Flint technically has clean water now. Well over six goddamn years later.
Actually true.
_Has both rubik's cubes and a bookcase behind him, with a plethora of his own books face out_
Clever.
Pretty sure the number of books facing out even increases throughout the video!
@@LucasHutyler Omg you are right, and the number of rubik's cubes too!
And the number of his own books face out increase as the video progresses.
@@LucasHutyler Yup. I noticed that too.
They appear to be replicating their numbers throughout the video.
I like how the number of Rubik's Cubes and Humble Pis correlate to the length of the video.
*Roobix
If you plot the percentage of the length of time of the video against the difference in percentages of books+Rubik's cubes and the length of time it shows that there were books and Rubik's cubes taken away and added to Shiva's video. Undeniable mathematical proof!
If you divide the number of seconds in the video by the number of Rubik's Cubes multiplied by the number of book faces you get 3.14159 🤔
I'm pretty sure the cubes and books cause the length of the video
12:01 You can just double click the dot? My life has been changed.
That was so smooth I missed it until you pointed it out.
The amount of times I double click it at work and my colleagues go “wait, how did you do that so fast?” Or I see people painfully drag it all the way down, I have to stop them and show the double click.
Wow!
Now wait until you discover CTRL+Enter.
@@kain0m What does Ctrl + Enter do?
-Guy who got blown away by the double click
The best thing of this election is that it has me watching videos about math. Who would've thunk it?
These false fake Conspiracy theories. BIDEN/HARRIS2020
Agreed!
I love your perspective!
Hey Rafael welcome to the club, I had a major in mathematics in university, so I have been corrupted as such for a few decades. If you can find a maths topic that clicks with you, well that is fantastic. Cheers from Canada.
If the benefit of everything that happened includes things like people taking a bigger interest in subjects like maths and science, and shows people the importance of voting, then I think that's great
I LOL'd heartily at "the elephant (and indeed donkey) in the room..."
Omg I just got it...
@@George_Taylor_ the scale doesn’t change the fact that the slopes match (at least insofar as they’re both positive), and that was the point
@@George_Taylor_
You keep pasting that comment everywhere but keep ignoring everyone telling you what you’re going on about isn’t relevant to the issue.
"They subtracted one from the gradient of the plot, and then were amazed that the gradient had changed."
That line had me laughing way too hard. And it wasn't even a joke. It was just a statement of fact.
@Yuri Bezmenov And it isn’t, on the graph for the county in their original presentation where the use of the vote-adjusting algorithm wasn’t detected.
@Yuri Bezmenov Just a question - Imagine you have a graph with a correlation line that went through the point [50,70] (Republican straight ticket percentage,Trump single vote percentage), which would be representing his "Trump more popular than party" slide, right?
Now if the slope of the correlation was ~1, what would the y-value, meaning the expected percentage of trump-single-person-voters be, when we look at counties with 85%, or even 90% Republican straight ticket, to fit in with that correlation line?
x would grow from 50% by 35%, meaning y also grows by 35%, from 70%. Talk about exaggerated expectations...
Now you might say, he leaves out those areas greater than 80% in most of his plots, maybe he's just not telling us the limits of definition to not water down his point, right? Then just remember that these very numbers would also tell us all about Biden's performance. We'd have the correlation line on the Dem/Biden-graph at [50,30]; we got towards the 15%-Dem position; 15% is always included in his plots, right? And once more, we might realize that a result of MINUS 5% for Biden among split votes would seem a bit unexpected.
He basically tells you that in f(x) = mx+b, you can expect to play around with b, but never with m. Which will always result in impossible results around the extremes as we've seen, and for which assumption he provides no proof - to the contrary, he provides lots of real world examples where m is below 1, not a single one where m~1, and expects us to believe that this is not normal.
@Yuri Bezmenov I don't even understand how they would come to that conclusion? Trump is a very divisive candidate, and there is good reason to assume socially/fiscally conservative voters might choose to vote for Republican state representatives but not Trump as president, considering the decidedly unchristian and not fiscally conservative things he's done.
@Yuri Bezmenov I can't exactly tell what you're referring to. Could you answer my example first, please? What would be the expected value ofTrump split voter percentage in counties where Republican straight tickets is >80%, if we assumed the correlation line to go through 50/70 Republican/Trump split and had a slope ~1?
I think we'll have to agree that this would be an impossible prediction by that correlation line, right?
@Yuri Bezmenov Not with the "normal case"-Trump more liked than GOP candidate- slide he presents: th-cam.com/video/Ztu5Y5obWPk/w-d-xo.html
As he explains before, we add the x value to the y value to derive Trump votes - and arrive at what Trump voter percentage for x=50%?
I mean, we can argue where the linear regression over this sparse synthetic dataset would land - but clearly somewhere between 60% and 70%, DEFINITELY not at 50%.
Again, this is a linear function f(x) = mx+b.
You're saying m~1 in all legitimate cases, which would leave at least b variable, right? That's exactly what he proposes as his expectation in his Trump more/less popular than the other candidate slides.
What's f(50%), for any b != 0?
"I can' trust anyone who has a copy of his own book face out." Has two copies of his own book face out. And multiple Rubiks Cubes. Genius.
Two? I think you missed a few.
The number increased as the video went along. But at that point, he had 2 UK editions AND 2 US editions.
I had scratched my head and thought i was in a parallel universe lol your not alone...
@@Sam_on_TH-cam That is ... hilarious. Thanks for pointing that out! I had to scroll back and forth to see. Super clever.
th-cam.com/video/xECUrlnXCqk/w-d-xo.html
Dear algorithm, why did you not recommend me this as both a subscriber, and notification beller? Glad I stumbled.
Seriously, the books and rubiks cubes is possibly the funniest thing I've seen in a TH-cam video, well played.
The subtraction of percentages was a huge red flag.
Yep. Full stop. Had me asking what the heck was going on there... 😂
That maths breakdown at the end was savage
Wasn’t he using exactly the same plot axes as Ayyadurai was using?
@@George_Taylor_ The sales do not matter when the slopes are the same, which was the main point.
Getting sued for debunking a lie... America 101
Check out Shiva vs TechDirt, that's pretty revealing! Just not someone who copes well with being fact-checked...
wait, what happened?
And conservatives still pretend to be all about freedom of speech.
It's always been that conservatives feel entitled to freedom from consequences for whatever false or bigoted things they say.
@@dannyboy4682 how so?
@@dannyboy4682 what bigoted things are u talkin about?
The working out zone contained far less gym equipment than I was prepared for
Brave (in the American sense) move to discuss maths in politics. I respect it. I also respect the "I AM A MATHEMATICIAN PLEASE DON'T SUE ME" energy throughout.
Curious: What does “brave” mean elsewhere?
@@courtney-ray From what I know, British English can use it as "foolish", I don't know about the other English speaking nations.
@@andrejnawoj8471 this is news to me, a Brit. Maybe the confusion is that we are condescending and sarcastic all the live-long-day? Calling stupid things "brave" in a condescending and sarcastic way until Americans think we've changed the meaning of the word.
@@andrejnawoj8471 That would only be true for British English if there was sarcastic context applied. Then again, that would also be true for American English?
I've never heard of "brave" meaning "foolish"
I think it would be in the context of a person does something kind of unorthodox and you might say “That’s brave.” I’ve seen brits use it in that context but I am American so you definitely know more than I do.
I was very confused on why on earth they were plotting that difference in the y axis, I´m glad I stayed to watch the entire video.
Plotting the difference (as matt points out) is just a transformation of plotting the two percentages...it's another way of looking at the same trend. It's actually not a mistake to look at it this way as Matt is suggesting.
Matt is the true pro. He's got Rubik's cubes AND his own book behind him.
He also has some Rubik's Dodecahedrons.
And he also increased the total count of both during the video to increase his merit!
“In the last video, I promoted my book 'humble pi', which I have avoided bringing up this time” Hohoho~
What was the name of that book again?
@@GaryGrumble I don't know, but he said that he promoted "humble pi" in last video, so probably there you could find the title of "humble pi"
I've been hating the use of statistics to claim a conclusion where the audience don't really understand the method used.
Might as well just say you get it by magic.
Thank you for being neutral and informative on political problems.
Hrrm, this whole year sure has been a statistic fest. First we get all the corona charts and data sets, now it's the elections. Very cool to see mathematics get all this attention, but yes all the attention should be on the correct maths.
@@jinjunliu2401 Never trust graphs you haven't faked yourself, as they say.
Friendly reminder that anybody can get a PhD by putting in some time and generally avoiding death.
Don't forget a lot of hard work. Doesn't make you honest though.
@@Henriktranoy or smart
🧐 I'd like to see your data on that.
Thanks! Just graduated might a maths degree but this video reminded me how important the basics are.
Funny, went over to the Shiva video and he starts off bragging about his 4 degrees and PhD. Meanwhile Matt lets the content of his video speak for itself without self aggrandisement. I wonder who more intellectually insecure out of the two of them...
I live in Silicon Valley. Trust me when I say that the people that boast the most about their academic performance and amount of certifications are the most rigid and uncreative people I've ever met.
Education does not equal intelligence.
@@zoltandober Education requires opportunity. Intelligence does not. In poor countries, there are loads of extremely smart people, but few who receive education. Yet they still seem to pop out once in a while regardless. Without any "proper" education.
Lost humble points for copies of 'humble pi" multiplying like rabbits, and what I assume to be Rubik's cube "taunting," subliminally, in the background. :-)
If you have to brag about how many PhDs and other degrees you have that's a good sign you don't have anything else to say
@@pgtmr2713 That part is a joke specifically against himself though, from him saying not to trust someone who has a book behind them.
You were awfully kind to these people, since it's pretty obvious to me they did some elaborate nonsense to generate a graph that supported their conclusion, then withheld relevant data from their viewers.
Like, them saying, "Feel free to challenge our ideas!" is definitely just a bluff to gain confidence.
I think it's important in arguments to have respect to the person you're arguing with. People too often on the internet treat their opponent like an idiot.
They know their viewers won't check it themselves and if they do and speak up, they can be silenced by them it just the general mob.
@@Yolwoocle Idk about that, the guy is a republican candidate and I didn't notice him disclose that in the beginning of the video...
The fact that he was kind kept me watching, and I'm probably in the group of people that would have believed the faulty conclusions on face value. I think its more effective this way at crossing divides.
I think the fact that he was so kind and showed them more respect than they deserve makes it harder for them to just dismiss this video as a “hater” or something.
10:49 "If your theories live by the spreadsheet, then I'm afraid they can also die by the spreadsheet" has got to be one of the hardest lines I've ever heard.
Shots fired, hits taken, none spared
I watched Dr. Shiva's video first. A couple weeks later I came across Mr. Parker's critique. I wish my math was better or that I was smarter, because I still feel like I need more information. Great video Mr. Parker.
You're probably affected by anchoring bias, having watched Dr. Shiva's video first. Why should the split-ticket voters be voting in the exact same way, or exact same numbers, as straight-party voters? Also, the conclusion with the downward slope only proves that split-ticket and straight-party voters in political precinct are correlated (but not identical). The initial analysis doesn't make sense.
-"Look at me, i got two rubik's cubes, I'm smart!" .
-"Look at me, there's a book with my name on it, so you can't discuss my authority" .
Stand-up Math: "Ooooh, that's how it works? Got it" .
having a rubik cube on your desk does not prove you ever used them, they seem to be in exactly the same state as as the were when bought., same goes for books on the shelf, it only shows you have them but doesn't proove you read them.
@@kamion53 Having a brain in your head, same...
@@kamion53 How can you tell which state they were bought in?
I see no "Rubik'R'Us, Oregon" or similar on any of them...
@@RogerBarraud that *is* a joke, right? A pun on the word 'state'? Surely?
@@RogerBarraud I remember that when I bought one _ I think the American flag still had 48 stars - it came in the state you see in the video.
"You should look at a swing state or somewhere important".
American democracy summed up perfectly.
I watch both videos...and this is a complete misrepresentation of what Dr. Shiva was showing with the graph. Shiva was showing how precinct's that voted low for Democrats, also voted high for Biden, and as you go further to the viewers right on the graph you are getting more and more Republican but also strangely more and more Biden votes...and as you go viewers left you are getting into more Democrat districts and more equal distribution of either less or more Trump votes by those Republican voters in Democrats districts.
This guy is a complete liar, and thinks that he can pull the wool over your eyes because you won't look at the other video
Now we know why he won't use any of Dr. Shivas video.
@@giant9833 Gee, maybe a lot of Republicans decided that Trump sucks, and split their ticket so they could vote for Biden. (And with more Republicans in a district, you see more more of these voters.) It's not hard to explain, if you think about it... but you have to be willing to think about it.
@@jpdemer5 Then how does Matt get the same plot for Biden...It makes no sense because Matt is misrepresenting Dr. Shivas video..
Why don't you just go watch Dr.Shivas video to compare?
You won't because you like Matt too much
@@giant9833 but Shiva's graph doesn't show the precinct that voted low for democrats, Matt's graph does, and Matt's graph (based on the same data but plotted more honestly) does not show anything suspicious
@@giant9833 not indication of anything malicious going on.
It could just be that republican voters dislike Trump, so they split and vote for certain republicans in Senate and Biden for president.
And that this behaviour is more common among republicans then democrats, which is expected when checking how Trump behaves and really tries to divide the country.
You have to think wether the assumption Shiva made is probable or not, and it just isn't. So Shiva whole argument is based on a flawed assumption. Which is the problem with persons lacking connection with reality outside their own bubble.
the fear of getting absolutely legal action’d into the ground is palpable
With Trump still in charge it might not be paranoid to worry about the espionage act.
Me, someone who's not nearly as good at math as I should be: I'm not really sure, but I think I kinda get what Matt's saying
*Matt shows the equation of a line*
Me: I know what m and b represent! Yeah, I'm smart
as a math teacher who deal with a lot of parents, you then know more then average
@@iamverita but the sad thing is . . . I'm also a math teacher. I teach high school geometry
The background is killing me. Matt is one cheeky bugger.
At least it isn't his own face showing on his own books xD
So funny... xD
hUmBle piE hUmBle piE hUmBle piE hUmBle piE hUmBle piE hUmBle piE hUmBle piE hUmBle piE hUmBle piE hUmBle piE hUmBle piE hUmBle piE
@@JasperNLxD
It doesn't make a difference. It's still the same and he was being funny.
Yeah very funny, say someone can't be trusted then do the same. How to discredit anything you just said by "joking"
The one question that any decent scientist must always ask: “But what if I’m wrong?”
These last four years have clearly demonstrated that it doesn't matter if you're right or wrong.
The ever-increasing number of rubex cubes and your books in the background was a nice touch
The best thing for me in the video was to find out that in Excel you can double click the edge of the cell to replicate the formula to whole stack! THANK YOU!
Worth mentioning that the reason the slope is less than 1, ie the reason the difference is negatively correlated, is likely that Candidate votes will tend to lean the same way as the precinct overall, but will lean less strongly one way or the other because people choosing to vote in each race separately are more likely to be more centrist (since they presumably want to select candidates from a mix of parties across the different races). The more heavily the precinct leans toward one party, the more the candidate voters will deviate from the party voters on average because those are the more centrist voters.
You could do a whole political science video on this one chart.
It seems like this is the actual refutation of the original argument, and I'm still not sure how Matt's addresses it exactly. They were expecting a flat line, which IS what you'd get if you subtract 1 from the slope of a y=x line (which they expected, if they think the same vote split among party/nonparty voters). What Matt shows here seems (though I expect I'm wrong) to sort of "transpose" the problem. The slope is still more negative than they claim to expect. In one graph it's negative instead of flat, and in the "fixed" on it's less than 1 instead of 1. What am I missing?
@@binarysmile The point you are missing is that y is not equal to x.
The fact that the same trend is seen for both candidates suggests its not as simple as y=x.
@@binarysmile As others have pointed out, the Issue is they were assuming y=x which was false.
Matt DOES address this in the video. The number of people who voted party line was NOT equal to the number of people who individually chose their candidates.
What the original video did was assume that in every county, the number of people who voted party line and the number of people who individually choose their candidates were equal, which was factually wrong.
That is why the data appeared that way. It is ALSO why Biden's graph of the same data had the same exact issue present.
Matt Parker - defender of democratic process, and cheeky promoter of his own book! Well done. :)
Dont forget Trey Stone!
@@nullvoid564 Who TF is Trey Stone?
@@nullvoid564 is this a south park joke?
@@robindelabra3874 yes
Not even the democratic process, just against bad maths.
Actually impressed beyond belief at your neutral, impartial and balanced stance in such a nasty politically charged mess. Love the man who loves the maths. Keep it clean and brilliant, and don't stop with the beautiful critical thinking and anyone-can-do-it explanations. Don't ever change!
Unbiased??????! Bahahaha
A neutral person would be fuming at the threat to democracy being levelled by a narcissist. I cannot muster this level of placidness in the face of something so wrong.
Agreed, Matt's presentations are always cogent and lucid. A pleasure to follow along.
@@zaphoddog3878 The guy is an amateur
@@wizard7314 nice narrative you got there. A neutral person would say "calm tf down Trump but sure let's see if you have any sound arguments and evidence" they shouldn't just presume that he's an evil fascist dictator who hates democracy and he can't POSSIBLY be right.
You have exposed them for frauds very quickly and efficiently, I'm impressed.
Another naive Democrat who watches CNN non stop
They really are frauds. I don't know how else they could make those claims. (CNN or not, which I don't watch much of because it sucks.)
@@apoloonce1712 so you genuinely think mathematics is dependent on what news outlet you consune? 🤣🤣🤣
@@thisaccountisdead168 It's called projection. They watch Fox and Newsmax and Breitbart nonstop and regurgitate their claims uncritically, so *obviously* anyone who disagrees with them must be doing the same thing but with different news sources.
Found the brainwashed sheep 👉@@apoloonce1712🤡
“Facts are stubborn things, but statistics are flexible.” Mark Twain.
72.7% of statistics are made up on the spot.
Or was that 53.2%?
Facts: Google-TH-cam pushed this to the top of people's feed to make sure everyone saw it.
Statistics are constant, the problem lies in the flexibility of being able to pull them from misleading base data, or display them in a misleading way.
@@elmerbeltshire7599 "Facts: Google-TH-cam pushed this to the top of people's feed to make sure everyone saw it.
"
A trumpist's last excuse if all others fall on deaf ears.
*random quote* Mark Twain
"Hopefully this is the end of my election videos." Oh, Matt...
Until the next election.
No wonder I can't find a copy of his book... he has them all!
Ahahahahaha
@Juan Suarez that's not a response, it's a defense
Maybe he leaves them in his flat to multiply - has to save the cost of printing!
@Vinny Mac Only in the minds of crayon eating illiterate goons. For people who understand mathematics, the claim that he makes that what he is doing isn't maths is the final proof that emperor is naked. Not only is it explicitly maths, he is making mistakes that you teach 17/18 year olds not to make...
@Vinny Mac You can believe that if you want, just watch out for magic bean salesmen, you're their ideal consumer
"To people who think we are completely off base, prove us wrong"
"I think I'll give it a go!"
Oh they 'bout to get the smoke
That bookcase gag was a thing of beauty.
"I've debunked a thing and I got forty-odd comments saying “Too easy, debunk this!”, so here I am!"
Oh no. Oh no, I've seen this before. Don't do it, Matt. You'll be spending way too much time chasing the shifting goalposts.
I'll trust you that this plot is fascinating (haven't watched til the end yet), but please just ignore those comments, for sanity's sake.
He will get add revenue for it so it's not all bad.
if the fleeing goalpost flees through an interesting plot why not chase it?
@@Royvan7 It encourages the fleeing and soon he'll wind up debunking uninteresting plots.
@@durnsidh6483 then don;t keep chasing when they arn't running through interesting plots. it not hard problem to solve keep your eyes on the prize. an excuse to talk about plots that people care about.
@@Royvan7 Fair point, I just worry that this will turn into a "Matt vs. The Conspiracy Theorists" slug fest. Though if he keeps his eye on the prize as you said, he'll be fine.
Thank you Matt, I'm far from a Biden supporter, but I just had an online discussion where this video was brought up and exactly conjectured that this "anomaly" would also show up in Biden's results. There was no explanation for why one would expect a horizontal line in those graphs and no comparison to voting data of Biden.
I think deep down we all know the truth.. it’s not who votes that counts.. it’s who counts the votes...
I think Matt mentioned something near it @ 15:09, where he says that any slope
@@joshh828 re: "minusing" -- I'd recommend "subtracting"; it's the verb that makes a difference!
@Benjamin Maiorella a clear red herring used to stoke their base. It's no coincidence every single suite they pushed was strongly put down for lack of evidence and may of those by republican judges, and the two lawfirms decided to bail because they got nothing to bring forward. But sadly their base does not care and necer will. That being said it's quite scarry even from Canada seeing the road the GOP is taking right now this is an unprecedented attack on democracy and in the US of all places.
Erm, Phil has put out normal data where the county are hand counted.... It should be unpredictable
This has got to be the most apolitical political video I’ve seen. Well done Parker.
yet some people in the comments still feel the need to bash America and/or an apposing political party...
@@dogol284 havent seen anyone bash the awful hyper-nationalistic education system so here I am
@@burnlootmurder8517 he doesn't claim that its a mystery though and he explains why it isn't
Congrats, you failed the test! Yes, Matt didn't claim it's a mystery, I claimed Matt's underhanded manipulation and misrepresentation of results is a "mystery" (but not really).
His explanation is just plain wrong, there's much simpler mathematical explanation why they look the same, and nowhere Matt even hinted at it. But I'll give you another hint it's because those graphs are strongly dependent, to the point they are bound to have the same slope, no matter what manipulations were used in counting.
But lets try again, If I were to plot 1-gop% against 1-trump% , I would get 180 degree rotation of Trump's graph, but what else? What that graph would show? Could that result somehow correlate with Biden's graph in two party system?
Since you've probably bailed, I'll tell you what it means, the plot would show not Trump votes vs not Republican votes, in two party system like US, vast majority of those vote for dems and Biden. Therefore Biden's graph is extremely close to 180 degree rotation of Trump's, THAT'S WHY THEY LOOK THE SAME. They would always look the same no matter what Russian hackers would do!
Now, why would supposedly apolitical math enthusiast keep silent about that fact and pretend as if it wasn't simple 180 degree rotation? This is the "mystery" I was talking about.
But since I have to spell everything out, it's because:
a) Matt made a politically motivated lie under guise of math
b) Matt doesn't understand simple math concepts like rotation transformation and what it does to linear graphs.
c) a and b.