For all of the math people, You can think of the probability of drawing the land in two tries as 100% minus the probability of not drawing a land twice. The first draw the probability of not drawing a land is 64/92 and the second try it is 64/91. If you multiply those 2 numbers together, you get that the probability of not drawing a land in 2 draws is 49%. Thus the probability of drawing at least 1 land is 51%. Please correct me if I am wrong.
The only thing I thought of was if you had partner commanders the ratio would be slightly higher given the extra card in hand. But that math looked good to me!
On your second try you should assume you hit a nonland on the first draw, making it 63/91, but that shouldn't make a big difference in the final result.
34:46 the math comes out to roughly 50/50 that Cam will hit ATLEAST one land in the first 2 draws. The way you figure this out is by figuring out the inverse probability (chances of hitting no lands) then subtracting that from 1. So considering 28/91 is roughly 30% of the deck is lands you would find out what’s the probability of hitting the 70% chance twice. So .7x.7=.49 or 49% which means Cam has roughly a 51% chance to hit land by turn 2 If you’d like to look at it the opposite way you can figure out the chances of all 3 possible successes (draw a land both turn 1&2, draw land turn 1 non land turn 2, draw non land turn 1 land turn 2) which would be .3x.3=.09(9%); .3x.7=.21(21%); .7x.3=.21(21%); add them all up and it’s 51%.
1/3 twice computes as 1 - (2/3*2/3) (as 1 minus the accumulated likelyhood of the oppose event happening) which is just about 55%. That excludes any fast mana though. I probably don't keep that hand either. I do generally agree with all mulligan choices on JT. Also I didn't see Madga in the Mardu mulligans so I assume Dylan still doesn't play Magda in JT and I swore to comment on that until Dylan plays Madga in JT.
for anyone wondering its about 5/9 chance to draw a land at 35:00 mark, which approximates to about 60% but you cant just add the two 30% chances together because then that would be mean by turn 4 your have a 120% chance of getting a land which just isnt true, you would multiply the chances of of drawing/not drawing al and and by eachother twice, twice as two instances of drawing and then add those totals together, which is 1/9 of drawing two lands, 2 times 2/9 chances of drawning one alnd and 4/9 of drawing no land, just fyi :)
Gotta say it. I love ya guys. The friendship between the three of you is infectious. Feels like I can actually sit down there and hang out with you all.
The way I’ve always looked at cEDH vs EDH: cEDH- mulligans and turn 1 are early game. You essentially need to be done developing your mana and hopefully have your draw engine in play. EDH- mulligan is hoping for 2-3 lands, a piece of ramp and a draw spell. Early game is turns 1-3
Yeah same for me on EDH. 2-3 lands, stuff to do early. Don't want to be passing for 4-5 turns playing lands and only ramping. I see way too many people keep hands of 3 lands, 1 ramp spell, 3 twelve drops.
I love Fridays because this is when my cEDH play group gets together and we always talk about the new pod cast. You guys have really changed how I look at the game and how I deck build for cEDH. Yall are the three musketeers of cEDH!!
My other favorite channel is cEDH TV and they always show/explain their mulligans. I love this aspect as it teaches me a lot. I would love to see yall explain mulligans in games.
Not sure if you guys have done this before/would be interested, but I'd enjoy the heck out of a commentary episode of your own games to see where your thought processes were during each step. I love seeing the opening hand conversations and made me realize I'd enjoy seeing a turn 2/3/4 and how the adaptations are played.
34:11 For what it's worth you would multiply the probabilities together of you NOT drawing ANY land during these two draws. With 92 cards left in the deck, 29 of them being lands you have a ~68.48% chance of not drawing a land. Assuming you don't draw a land, your subsequent draw has ~68.13% chance of not being a land. Multiply these together, and you have ~46.65% of NOT drawing a land, which means ~53.35% of drawing at least one land during your first two draws.
Although I'm not a fan of the Ent, if you're looking to fill the role of a massive beater that nets card advantage in green, have you thought about Elder Gargoroth? It doesn't land cycle, but it has trample, vigilance, and reach in addition to passively gaining life, drawing a card, or making a 3/3 body whenever it attacks or blocks. Idk if it really works for Kinan, but the card has been quite helpful in my mono green list against Kraum when Blue Farm has to fall back onto their beats plan. Between that, Boseiju, and Endurance, green has some relatively solid responses to some of the best combos in the format.
"And that is... what it is" ---- Cam 😂😂😂😂😂😂😂😂 Omg I love you guys so much, literally cracked TF up at 2am in bed watching this, made my wife mad for waking her up, but it was totally worth it 🤣🤣🤣. This is my favorite TH-cam channel bar none.
You guys should do a series where you help cedh and commander players edit their decks with their budgets and their stock. It's quick paced and you help multiple people with their decks while on the phone. It would be huge and could all be done with moxfield.
Lots of experience and knowledge from you guys as always, I have a bad habit of keeping slightly risky hands, like only has a fish turn 1 and rocks to get it, but then my pod decides that's the ONE game in a hundred they'll be responsible lolololl
I do think that last hand would be better to play the dork. Guarantee of 3 mana on turn two is a lot better than risking only having 1 mana to then play a dork that can't tap, especially when your commander is a tutor that requires a lot of mana. Dockside was also legal at this point, so as long as you could hit 5 mana by turn 3, then you're probably set in this pod. Because Rocco is somewhat unique in that as long as you have enough mana, you can get whatever you want, I do think that prioritizing guaranteeing that you hit 3+ mana ASAP is better than seeing two extra cards on turn 2. Since the dockside ban, I can definitely see a stronger argument for turn 1 Sylvan Library, especially since land counts have generally gone up a bit, so you're pretty much guaranteed to hit one with the library. Episodes like this are also fascinating to watch because a lot of that decision is player preference, and studying social psyc, it's really interesting seeing the thought processes of more turbo players like Dylan explained.
28:12 „the eagles look really great here“ … guess what, them being a land would have been a ton better here. And you can’t convince me, that a 3/3 flyer with that etb really is going to win you the game (or stops you from losing the game), when you would draw it late. I’d rather have a bicycle land tbh
As an exact value, I'm fairly sure the probability of drawing at least one land is the sum probabilities of (1st draw is land, 2nd isn't), (1st draw is land, so is second), (1st draw isn't, 2nd is) which is ((28/92)*(64/91))+((28/92)*(27/91))+(64/92*28/91)) which is about 51.8%. Stats is complex and I'm not a statistician so I could be wrong though. This does not account for fast mana either obv. And if you specifically want one land from the two draws, just remove the middle option and it calculates to about 42.8%
You need to be looking at the odds of missing a land in 2 draws thats whats tripping you up. Its like a 63/92% chance to miss a land then a 62/91% to miss it a 2nd time. Averages out to slightly better than coinflip odds to hit a land within two draws
You actually have to take the chances you don’t hit your lands and that leads you with the inverse being the chance you do hit. 64 non-land cards left in deck out of 92 cards total on the first draw then 63 non-lands out of 91. 64/92 *63/91=.4816 1-.4816=.5184 So you actually would have had a ~52% chance that you hit a lane by turn 2.
Also, in this case, I would say the math definitely says you should keep. Sometimes you get bad variance but as a professional poker player I’d take the extra 2% every time.
The probability to draw a land goes like this. Let's assume for the sanity of the commender that 1/3 is the probability to draw a land each time and 2/3 to not draw a land each time So let's calculate the event that you will draw at least one land over two draws. That is 1 minus the probability of drawing zero lands over 2 turns. So let's calculate that. The probability of drawing zero lands over two turns is (2/3)*(2/3) = 4/9. That means that over half of the time, 5/9 you will be drawing a land until turn two. Yeah it is an approximation I didn't want to do it on a calculator because I am on my phone. However if you want the exact match change the 2/3 with the ratio of land and cards in deck in the first 2 turns and repeat my math and you will have it. In conclusion. A flip of an almost fair coin is a good chance in my head and I would have taken in casual table. I don't know.if I would in a tournament which I paid to enter. I would consider my results up to then and if it is public info the results of my opponents as well. It is a close call
Ok, very very surprising take to begin the podcast : do y'all take your hands AFTER finding out turn order ? That's new and a bit broken to me. I've always played it "select your hand, THEN roll for place", and while I understand how you discuss strategy on your premise, I also believe my way of doing it is way more balanced precisely for that.
You might already know by now, but the rules state that you draw your starting hand after you reveal your commander and determine turn order. So, for practicing mulligans, you should also take that information into account.
You cant just add 30% + 30% = 60% to hit a land. Instead think of the probability you whiff, about 70%, so the probability u whiff twice is 70%*70%=49%, so you have about 51% chance to hit at least one land.
Guys, I love your stuff but you could be even better players if you learned just the basics of probability theory. Knowing how to calculate the overall odds of consecutive events is just usefull in... Life. 😅
The odds of drawing at least one land in the next 2 turns is the cumulative chance of drawing 1 and drawing 2. 29/92 is really close to 1/3 as you correctly pointed out so since you do this in your head at the table it's good to stick with this heuristic. What you want to do is to figure out what the odds are of NOT drawing a land and then reverse them. 70% you don't draw one on the first draw and almost similar odds on the second one. You multiply these to know what the chance is of NOT drawing a land. 0.7*0.7=~0.5 (again this is in your head) So your odds of not drawing a land in the next 2 draws is roughly 50% and in turn the same for drawing 1 or 2 lands, meaning "decent" both ways is exactly right 😂
Cannot stress enough how enjoyable it is when y'all have Tyler on the podcast!
I hate it! I hope they increase my suffering.
For all of the math people, You can think of the probability of drawing the land in two tries as 100% minus the probability of not drawing a land twice. The first draw the probability of not drawing a land is 64/92 and the second try it is 64/91. If you multiply those 2 numbers together, you get that the probability of not drawing a land in 2 draws is 49%. Thus the probability of drawing at least 1 land is 51%.
Please correct me if I am wrong.
The only thing I thought of was if you had partner commanders the ratio would be slightly higher given the extra card in hand. But that math looked good to me!
On your second try you should assume you hit a nonland on the first draw, making it 63/91, but that shouldn't make a big difference in the final result.
@@alexanderbuttler3902 Also a good catch.
@@alexanderbuttler3902 this is what I came here to say. It ends up at like 52% you hit your land.
That is the faster way of doing it. I just added up the odds of hitting land land, land nonland, nonland land.
34:46 the math comes out to roughly 50/50 that Cam will hit ATLEAST one land in the first 2 draws. The way you figure this out is by figuring out the inverse probability (chances of hitting no lands) then subtracting that from 1. So considering 28/91 is roughly 30% of the deck is lands you would find out what’s the probability of hitting the 70% chance twice. So .7x.7=.49 or 49% which means Cam has roughly a 51% chance to hit land by turn 2
If you’d like to look at it the opposite way you can figure out the chances of all 3 possible successes (draw a land both turn 1&2, draw land turn 1 non land turn 2, draw non land turn 1 land turn 2) which would be .3x.3=.09(9%); .3x.7=.21(21%); .7x.3=.21(21%); add them all up and it’s 51%.
I do this math by thinking about a probability tree and adding up the braches that come out to the outcome I wanted
Dylan: Pronounces fyndhorn elf correctly
Also Dylan: Corrects himself to say flinthorn elf
36:50
He incorrected himself
1/3 twice computes as 1 - (2/3*2/3) (as 1 minus the accumulated likelyhood of the oppose event happening) which is just about 55%. That excludes any fast mana though. I probably don't keep that hand either. I do generally agree with all mulligan choices on JT. Also I didn't see Madga in the Mardu mulligans so I assume Dylan still doesn't play Magda in JT and I swore to comment on that until Dylan plays Madga in JT.
Always enjoy when Tyler can make the cast 🙏🏼
for anyone wondering its about 5/9 chance to draw a land at 35:00 mark, which approximates to about 60% but you cant just add the two 30% chances together because then that would be mean by turn 4 your have a 120% chance of getting a land which just isnt true, you would multiply the chances of of drawing/not drawing al and and by eachother twice, twice as two instances of drawing and then add those totals together, which is 1/9 of drawing two lands, 2 times 2/9 chances of drawning one alnd and 4/9 of drawing no land, just fyi :)
Gotta say it. I love ya guys. The friendship between the three of you is infectious. Feels like I can actually sit down there and hang out with you all.
The way I’ve always looked at cEDH vs EDH:
cEDH- mulligans and turn 1 are early game. You essentially need to be done developing your mana and hopefully have your draw engine in play.
EDH- mulligan is hoping for 2-3 lands, a piece of ramp and a draw spell. Early game is turns 1-3
Yeah same for me on EDH. 2-3 lands, stuff to do early. Don't want to be passing for 4-5 turns playing lands and only ramping. I see way too many people keep hands of 3 lands, 1 ramp spell, 3 twelve drops.
Man, this podcast actually helped clear some of the fog in my brain for mulligan logic. Thanks a bunch guys!
The discussion between 7 is perfect. The form thought process that goes on is invaluable.
I love Fridays because this is when my cEDH play group gets together and we always talk about the new pod cast. You guys have really changed how I look at the game and how I deck build for cEDH. Yall are the three musketeers of cEDH!!
34:00 painful. Odds of getting at least 1 land in 2 draws = 1 - (change of not drawing a land twice) = approximate 1 - (0.66)^2 = approximate 56%
My other favorite channel is cEDH TV and they always show/explain their mulligans. I love this aspect as it teaches me a lot. I would love to see yall explain mulligans in games.
This is one of my favorite topics for cEDH. Thanks for the video!
Not sure if you guys have done this before/would be interested, but I'd enjoy the heck out of a commentary episode of your own games to see where your thought processes were during each step. I love seeing the opening hand conversations and made me realize I'd enjoy seeing a turn 2/3/4 and how the adaptations are played.
This is by far my favourite MTG YT channel!
34:11 For what it's worth you would multiply the probabilities together of you NOT drawing ANY land during these two draws. With 92 cards left in the deck, 29 of them being lands you have a ~68.48% chance of not drawing a land. Assuming you don't draw a land, your subsequent draw has ~68.13% chance of not being a land. Multiply these together, and you have ~46.65% of NOT drawing a land, which means ~53.35% of drawing at least one land during your first two draws.
Although I'm not a fan of the Ent, if you're looking to fill the role of a massive beater that nets card advantage in green, have you thought about Elder Gargoroth? It doesn't land cycle, but it has trample, vigilance, and reach in addition to passively gaining life, drawing a card, or making a 3/3 body whenever it attacks or blocks. Idk if it really works for Kinan, but the card has been quite helpful in my mono green list against Kraum when Blue Farm has to fall back onto their beats plan. Between that, Boseiju, and Endurance, green has some relatively solid responses to some of the best combos in the format.
More of these mulligan videos with tyler, please! Or other guests, discussing other commanders.
Love this ep guys. One of my new favorite channels! Learned a lot and had fun doing it.
"And that is... what it is"
---- Cam
😂😂😂😂😂😂😂😂 Omg I love you guys so much, literally cracked TF up at 2am in bed watching this, made my wife mad for waking her up, but it was totally worth it 🤣🤣🤣. This is my favorite TH-cam channel bar none.
Im so glad Tyler took stats in school.
You guys should do a series where you help cedh and commander players edit their decks with their budgets and their stock. It's quick paced and you help multiple people with their decks while on the phone. It would be huge and could all be done with moxfield.
Lots of experience and knowledge from you guys as always, I have a bad habit of keeping slightly risky hands, like only has a fish turn 1 and rocks to get it, but then my pod decides that's the ONE game in a hundred they'll be responsible lolololl
Very helpful video thus far. New cedh player here
35:52 so the math to not draw a land is what are the two odds multiplied. So it’s .66*.66 so you have a 56% to draw a land
THALROG for Thalia and The Gitrog Monster so you’re never confused about which card you’re talking about and have a name much easier to say
27:09 is t0 gemstone into t1 jeska's will (or ripping a positive manasource of the top) and then one ring not good enough?
8:50 most important part of the video for real.
very important skill to practice! mulligans are the single most important meal of the game
Yay Tyler is in the podcast 😊
I do think that last hand would be better to play the dork. Guarantee of 3 mana on turn two is a lot better than risking only having 1 mana to then play a dork that can't tap, especially when your commander is a tutor that requires a lot of mana. Dockside was also legal at this point, so as long as you could hit 5 mana by turn 3, then you're probably set in this pod. Because Rocco is somewhat unique in that as long as you have enough mana, you can get whatever you want, I do think that prioritizing guaranteeing that you hit 3+ mana ASAP is better than seeing two extra cards on turn 2.
Since the dockside ban, I can definitely see a stronger argument for turn 1 Sylvan Library, especially since land counts have generally gone up a bit, so you're pretty much guaranteed to hit one with the library. Episodes like this are also fascinating to watch because a lot of that decision is player preference, and studying social psyc, it's really interesting seeing the thought processes of more turbo players like Dylan explained.
hi i am new to cedh how does the muligan work u get the first for free and u do it in turn order is that right
I for some reason thought you guys were going to sit there and shuffle actual decks to do this😂. I forgot moxfield has this feature.
Cam trying to keep his hyperactive younger brothers in line a bit in this one.
28:12 „the eagles look really great here“ … guess what, them being a land would have been a ton better here. And you can’t convince me, that a 3/3 flyer with that etb really is going to win you the game (or stops you from losing the game), when you would draw it late. I’d rather have a bicycle land tbh
27:02 turn 1 Jwill into t1 One Ring?
Also lmao Dylan with the throat goat tech
It would be cool to know since you guys did mulligan down 3-4 times some times. Which hands were keeps if you had to tuck 2-3 cards
Nice, a video about doing hand stuff
As an exact value, I'm fairly sure the probability of drawing at least one land is the sum probabilities of (1st draw is land, 2nd isn't), (1st draw is land, so is second), (1st draw isn't, 2nd is) which is ((28/92)*(64/91))+((28/92)*(27/91))+(64/92*28/91)) which is about 51.8%. Stats is complex and I'm not a statistician so I could be wrong though. This does not account for fast mana either obv. And if you specifically want one land from the two draws, just remove the middle option and it calculates to about 42.8%
Funnily enough, this also means Cam was right when he said decent chance either way, it's basically a 50/50
If you have a 50% chance and you try it twice .... It's 100%, right?
I didn't mulligan with Yisan last night and stomped everyone. Now it's Dylan's turn.
Watching you guys do Tayam mulligans would have been peak comedy, maybe try that next time 😅
The probability of drawing at least one land with 30% lands in the deck is 1-(0.7*0.7) =0.51
It really throws me off hearing y’all talk about your lands; I get HEAVILY made fun of for running 30-32 lol
Love the discussion as always 😁
16:07
.
.
.
What sounds doth grace mine ears?
You need to be looking at the odds of missing a land in 2 draws thats whats tripping you up. Its like a 63/92% chance to miss a land then a 62/91% to miss it a 2nd time. Averages out to slightly better than coinflip odds to hit a land within two draws
man that rocco deck is so different then mine, i have a turbo i wouldnt keep the first hand cause i cant keep felidar in it
Glad to see that even cEDH players struggle with math.
I built dylans Winota list. Lets Do Winota Test Hands!!!
Why is Gemstone Cavern not good in your starting hand when you’re going last?
Still waiting on that cadaverous bloom list lol but alas, a tasty podcast will do 😂
Make another video like this!
Coulda watched already if you were on the Patreon😉*hint hint*
"No" comedy genius
Do nothing dan Is an amazing turn one play, and should be kept as often as possible.
You actually have to take the chances you don’t hit your lands and that leads you with the inverse being the chance you do hit.
64 non-land cards left in deck out of 92 cards total on the first draw then 63 non-lands out of 91.
64/92 *63/91=.4816
1-.4816=.5184
So you actually would have had a ~52% chance that you hit a lane by turn 2.
This is assuming 29 total land count in the 99. If it was 30 total lands like cam mentioned it would’ve been
1 - (63/92*62/91) = .533 or 53.3%
Also, in this case, I would say the math definitely says you should keep. Sometimes you get bad variance but as a professional poker player I’d take the extra 2% every time.
i would keep all the hands because all hands are beautiful :^)
The probability to draw a land goes like this.
Let's assume for the sanity of the commender that 1/3 is the probability to draw a land each time and 2/3 to not draw a land each time
So let's calculate the event that you will draw at least one land over two draws.
That is 1 minus the probability of drawing zero lands over 2 turns. So let's calculate that.
The probability of drawing zero lands over two turns is (2/3)*(2/3) = 4/9.
That means that over half of the time, 5/9 you will be drawing a land until turn two. Yeah it is an approximation I didn't want to do it on a calculator because I am on my phone. However if you want the exact match change the 2/3 with the ratio of land and cards in deck in the first 2 turns and repeat my math and you will have it.
In conclusion. A flip of an almost fair coin is a good chance in my head and I would have taken in casual table. I don't know.if I would in a tournament which I paid to enter. I would consider my results up to then and if it is public info the results of my opponents as well. It is a close call
Ok, very very surprising take to begin the podcast :
do y'all take your hands AFTER finding out turn order ? That's new and a bit broken to me. I've always played it "select your hand, THEN roll for place", and while I understand how you discuss strategy on your premise, I also believe my way of doing it is way more balanced precisely for that.
You might already know by now, but the rules state that you draw your starting hand after you reveal your commander and determine turn order. So, for practicing mulligans, you should also take that information into account.
Isn't it forty five point five percent when you count 2 cards out of 91 cards
Mmmm Tyler on the podcast really makes my neurons activate with *knowledge*
Never peak at the top when you mull, the bias is terrible for your decisions.
You cant just add 30% + 30% = 60% to hit a land. Instead think of the probability you whiff, about 70%, so the probability u whiff twice is 70%*70%=49%, so you have about 51% chance to hit at least one land.
Messing up the math to drive engagement
The numbers don’t lie, and they spell disaster for you!
30 percent of the time
You draw it
Every time
this is my kinda probability
The play to win struggling at math Podcast
I love when they do math
Dylan is totally big head, big mouth kinda guy 💖
Love this video but the math on drawing the lands hurt my soul aha
Well thats not how the math work,
29/92+63/92*29/91 = 53.3% chance to draw at least 1 land in your next to draw
Number in turkish is “Sayı” now u know Tyler
Infinite food would be nice if you and urza out, you just don't in kinnan xD
Yall need to learn to just plug this shit into a hypergoemetric calculator
"Alright, now that we're all done being assholes, let's move on"
To what? What is the next evolution of asshole?
It’s 30% with each draw fellas 😂😂😂 not 60%. That would mean after the 3rd draw, your chances of hitting a lane are 120%. Which we know is wrong.
❤❤❤
listening to these goobers try and do statistical analysis is giving me life
Many times, I find myself just trying see Dylan’s tattoos and not listening to what he said
❤🎉
Pog
Always remember, math is for blockers
Guys, I love your stuff but you could be even better players if you learned just the basics of probability theory.
Knowing how to calculate the overall odds of consecutive events is just usefull in... Life. 😅
The odds of drawing at least one land in the next 2 turns is the cumulative chance of drawing 1 and drawing 2.
29/92 is really close to 1/3 as you correctly pointed out so since you do this in your head at the table it's good to stick with this heuristic.
What you want to do is to figure out what the odds are of NOT drawing a land and then reverse them.
70% you don't draw one on the first draw and almost similar odds on the second one.
You multiply these to know what the chance is of NOT drawing a land.
0.7*0.7=~0.5 (again this is in your head)
So your odds of not drawing a land in the next 2 draws is roughly 50% and in turn the same for drawing 1 or 2 lands, meaning "decent" both ways is exactly right 😂
One, two, three
Unos, dos, tres
Une, deux, trois
Ein, zwei, drei
Ich, ni, san