To be fair, if you're a juror, normally, there isn't a "insufficient evidence" option, and "Not guilty" actually does not mean "Innocent". The verdict is used for when we think someone is innocent OR there isn't sufficient evidence to affirm the person is guilty. So if Geoguessr intended "Not guilty" to be synonymous with "Innocent", you could say it's an incorrect use of the term from a judicial point of view.
If having a completely normal guessing pattern equates to "insufficient evidence" then "not guilty" should be removed from the system entirely. I assume each option has a weight and I think his marking of "not guilty" was completely justified. The system is flawed though.
@@Mercure250 the intended use of it I think is if they're obviously googling but you couldnt tell if theyre also scripting. that's why I would primarily use it, anyway
The player saw the .sk domain and quickly clicked on the capital. You don't need a script to know that .sk is Slovakia and you don't need to google it either, he didn't even have time. Not guilty was 100% the right option. An insufficient evidence report would have been ridiculous for that round, it was the clearest non-cheating in the whole video.
the first person literally had it space plonked on the 5k by the script and then they moved away from it to be less sus, was pretty obvious scripting imo
@@zi8gzag man Oscar at least do insufficient evidence on that one😂 (and for others as well. You said "I'm gonna need more for that one" and put not guilty. If you need more evidence, you choose insufficient and only if you are 100% they are legit you can put "not guilty" )
@@martian_heidegger if you compare it to Csgo, it's understandable since there are only 2 categories, insufficient evidence and evidence beyond most reasonable doubt. I don't know if I like the addition of "not guilty" and "guilty" but we will see.
Hi Zigzag! I was actually the guy who played that campeche round a few weeks ago so I thought I'd give you a bit of context and also my thoughts on the round. This was confirmed as it was the only investigation i didn't get in my lot of investigations. I am a 1200 rated Australian player in Master 1 division that only plays intermittently, instead I mainly watch the rainbolt daily challenges and some videos from yourself on a regular basis. I pretty much only play moving and base my guesses on vibes/feel, architecture, having a good understanding of languages and am able to use information to determine where I am. I have very little interest in studying poles / bollards / camera generations as it's simply boring in my opinion and has very little to do with actual geographical knowledge. Of course I know the most obvious / unique examples such as Peru/ Malay signs or Senegal poles or Gen 2 South Africa but am less interested in learning this as a whole. My thoughts on this round: This round was played on a high multi and actually won me the game with my opponent guessing by the coast west of Mexico City, obviously my opponent was a bit tilted and decided to report lol. When coming to this sign, I read Campeche on it and remembered seeing this in a Rainbolt daily challenge from a few days earlier and knew that it was a state/province/city somewhere but I couldn't quite remember where. But, the road, vegetation and architecture looked relatively similar to stuff I've seen in Uruguay / Northern Argentina but upon scanning the map and having another look at the surroundings - this clearly wasn't right. Now, remember I am not very familiar with specific poles / camera generations so did not know Mexico poles or that Uruguay was Gen 3 only, but that's definitely good info to know for the future. So while reciting Campeche, Campeche, Campeche in my head trying to remember where it was I looked for more information. Towards the end of the round after my opponent has guessed, I found a Nissan truck and a Chevrolet which are brands that I assosciate with Mexico and also saw the short plates on the truck which is more of a Mexican thing. So with 6 seconds remaining, I initially zoom into the centre of the more green / tropical part of Mexico and fantastic I see the label for Campeche and guess right on the label with 3 seconds remaining and its pretty damn close. I hope that I've provided some insight into this round :)
Youre 1200 rated without knowing Mexican poles? That sounds very unlikely to me Im 900 rated and Ive known that meta from when I was waay lower. Its one of the most known meta's in the Americas. Regardless, not saying youre a cheater 😅
That's strange because I'm absolutely sure that's me. I checked the tower and assumed it was a company name. Checked the ambulance because a lot of them have town names. Saw Campeche on the sign (missed the town name) and hit a mental block about where it was and forgot to look at the poles. One was right there. It looked vaguely like Uruguay at first but I soon realized it wasn't. As I looked around my opponent guessed and I saw the pickups with short plates, an octo-pole and realized it was Mexico. I had maybe 3-4 seconds left and rushed to Mexico. My rating at the time was mid 700s to 800. I'd be better than that but my memory isn't what it was now that I'm 70. Now I would like to see more info as I've been defending this all over the place. For the record, I don't know what regions have which camera generations and can't tell gen 3 from gen 4. I do know some poles and bollards. I play the game a modest amount, usually just the 20 games a week in Gold but just went up to Master.
@wileym Haha that's pretty cool I think you likely had this round in a game as well as ~100000 rounds are in each map used for comp, but cycled through each month from a larger set of rounds. The thing is I remember[ed] when writing this comment the specifics of what I did in the round - so based on the gameplay I remembered zooming inti uruguay, moving in pretty much that manner, looking at the number plate and zooming into Mexico in that manner after my opponent had guessed. More importantly, this was the only investigation I didn't get. There has been an update to the investigations tool which shows the elo range of the player when playing the round so we can 100% confirm based on that elo range if someone posts a new video showing this round. It's fun to see your story as well :)
You need to be putting insufficient evidence if you’re unsure, not “not guilty”. Deciding they’re not guilty when you’re even saying you’re unsure is going to multiply the amount of cheaters that get away with it.
You need to be putting insufficient if you're unsure, not "not guilty". Deciding they're not guilty when you're even saying you're unsure is going to multiply the amount of cheaters that get away with it.
@@lukebradfordnah plonking the exact road without even zooming in on the map at all is very suspicious and extremely hard to do, even if you know it’s Vienna
The idea of other players checking reported players is probably ok (especially for catching script users), but the problem is that high ranking players seem to play the game differently. They think about poles, bollards, phone area codes, transformers, google car colours and antennas and other irrelevant information. Of course, it's relevant for high ranking competion players, but not for casual players. We read the signs and try to guess the language or we try to find something useful from it. We check the road markings and the side they are driving. We check climate, vegetation, architecture, people, clothing etc. I think high ranking player use meta and make guesses based on very different information much more than more casual players. I have watched two other videos about this new mode and all three show that high ranking players do not understand the way casual players (ranking 600-1000) play. One thing common for zi8gzag and other pros seem to be that they think that you have to zoom to correct country right after you know the country or place. I don't to that. If I find something I know, I keep looking around and moving trying to find more information. I open the map when: 1) I have to make a guess (other player guessed), 2) I know enough that I can guess something and don't want the other player to keep searching, or 3) I know something I have to check from the map (e.g. road sign which says Canberra 197, some other place 49, and I need to find that other place). I work with satellite images (resolution 200-5000 m) and I have used google street view to check the surface condition (land cover type, vegetation, small scale topography). I have also read a lot (fiction and fact) and I know lot's of strange places which may not be familiar for typical high ranking players. I have travelled quite a lot (for work and leisure), so I have been in many places. I have also lived abroad and I have relatives in other countries. I have been interested in maps and geography all my life. I have been browsing google maps and google earth just for fun. Of course I use meta when I know it (Vienna street signs, Nigerian police car, some Google cars, such as Ghana, Guatemala, Kirgistan, Kyrgyzstan), and I know some common places in the game (at least earlier, Monaco, Midway, Christmas Island), but my game is not based on meta. I check the scenery and try to find signs if possible. I do NOT google, but still I may use 10-15 seconds trying to decipher signs in foreign languages. I'm trying to find something I know. I think that some of my rounds may look suspicious for high ranking players because I look at things they think are irrelevant or stop to think too often.
Not everyone plays the game the same way. I am in that 600-1000 range but I absolutely use poles, bollards, and area codes and stuff when I know it. I also only play No Moving. I mean, the problem with this is thinking that everyone who isn't a pro plays like this or knows this but doesn't know that. I happen to know all 96 departments of France, for example. Most of the time it's not useful but if I see a sign with "Allier" written on it I know where to go. The problem is that below the pro level, every body is just going to be very, very different in terms of the things they know and the things they don't. I absolutely knew that one round was Mexico from the poles, but didn't know the state. Other people might know the states but not the poles. Everyone learns different things so it's just wrong to say "a 600 wouldn't know that" or "600s play like this." That is why it would be very wrong for GG to include the rating in these replays. You can tell they are googling when they stare at a sign for 40 seconds while googling in another tab and then zoom straight for the exact spot, but basing your decision on anything other than the gameplay would be wrong in my opinion. I was playing against someone who was much higher rated than me the other day and we got the Santo Domingo car with the roof rack and the tape running long across. I instantly clicked Santo Domingo while my opponent, again much higher rated, went Guatemala because they didn't know about the tape. You just don't know what other people know and what they don't.
this is decipherable though, if someone looks at a sign with a town name for awhile and then looks around for awhile and makes a bad guess, you can do not guilty/insufficient. if it's googling for alot of these it's quite clear, because they'll look at the sign for an extended period and then get a good guess. this is where knowing a players elo would also be imperitave
I've seen only two cases of obvious cheating. One we were in California and there was a sign with the town name on it. I scanned up and down the coast looking for the town but couldn't find it. My opponent stared at the sign perfectly still for 30 seconds then activated the map and zoomed straight into this small town. I made a close enough guess of LA or whatever but they got 4997 or something. Second time there was a pole with some advertisement on it in English. I think I went in India somewhere but my opponent stared at this random sign for like 45 seconds then activated the map and plonked in Bangladesh. I realized after the game that if you google the company slogan from the ad on the pole you get some company in Bangladesh. In both cases the players ELO is irrelevant, it is just the gameplay that matters. I reported both players and neither got banned fwiw, I think primarily because geoguesr is so overwhelmed with bogus reports like we see in this video.
the one thing I noticed was that he mentioned generations (gen3?) on the uruguay/mexico one which I mean it is a bit bold to assume that you have to learn camera generations to be able to make good guesses regardless of if that player may have ended up being a cheater
I paused and took a look at the report decision page, the not guilty should only be chosen if you’re confident that the person is not cheating, so I think a lot of “not guilty” decisions should maybe be insufficient evidence instead Cool video though very interesting system they’ve got
The New Zealand round at 13:05 was a team duel. It's not indicated anywhere, but when you said that we can see the opponents guess, that was actually the team mate. You can also see the team mate giving a hint. (Also the last round in North Queensland is a team duel.)
That should be made far clearer in the interface as it affects how you play enormously - both in that you will sharing information and perhaps showing each other things on your screen if you are sitting next to each other, and also in terms of how you switch locations as you decide how to hedge. I could certainly see how teamplay could be confused for cheating - if often show each other signs, especially if one of us knows a language or country better.
The weird movement one in Canada felt to me like someone Canadian who was streaming. Like the pause at the lake felt like someone saying out loud "I should know this. I should know this. Don't tell me chat, I should know this."
As a lower rated player, I have to say that the likelihood of 10:00 is not low. I don't care about poles. I don't know anything about Mexico's poles. Campeche on the other hand is written on the map. I've known about it for a long time because I watch GeoGuessr content and people always call out state names. One of my first 5ks in duels was in no moving at the Veracruz/Tabasco border. I'm yapping but the point is that knowing states happens infinitely earlier than knowing poles unless you're tryhard training to be an NMPZ main.
I know states of other countries from google maps and the Spanish states from Spanish class. That round made me concerned I would be accused of cheating.
@@frijoless22 That part is weird, but I'm specifically responding to ZigZag's statement that it's unlikely for a player to know Campeche but not know Mexican poles. As for zooming into the wrong country, I once thought Zacatecas was a city in Peru. Then after scanning Peru for a while I noticed that there were no Z's, prompting me to realize that it's Nahuatl and not Mayan. Misremembering and thinking you know that something is somewhere that it isn't is entirely possible. As for cheating, why would he zoom into the wrong country if he was cheating? That makes even less sense.
a big reason why rating should NOT be shown is because it discriminates against people who just know a lot about the world but might not necessarily be big geoguesser players. i think rating shouldnt be shown as it just biases the reviewer into thinking "well hes high rank obviously he will know immiedetaly", while in fact that person might still be cheating, and the same "oh he's low rank so its impossible he got 5k so fast" while that person might just be fluent in many languages, be well traveled, know the various cultures, plants, etc. therefore even without "geoguessr knowledge" like google car, street poles, etc. they might find something faster than a pro player could depending on the situation. or someone simply could've physically been to the location of a photo - i once watched a geoguesser video where some guy randomly got streetview by a gas station thats like 15 min away from my house, so i definitely would've gotten a 5k if i got that one when playing.
In addition to what you listed, it's also possible someone plays a ton of non-duels and then decides to play some duels and already knows the metas. Geoguessr forces you to rank up pretty slowly when starting out.
On average you will catch more cheaters if you are biased.(rating, where they come from, other rounds, does he know metas or not, etc.) If you say you dont want a single falseban then you will have to let 95% of cheaters run free.
@@mursmumies123 i mean you had to before anyway - your games used to be reviewed by devs instead of both players and devs. from what i understand, you won't be punished simply because a lot of players chose "guilty" on one of your guesses. it will simply bring your case more attention and be looked into further by the devs themselves. if you aren't cheating, don't worry about it. and if you can't help but worry about it, maybe just slightly move your screen around when you're reading signs so it's obvious you're not tabbed out.
I really think they need to show more than one round. A really lucky guess once can just be really lucky. Five really lucky guesses in a row, and it's not luck that the guesses are so good
I think he doesn't realise what 1 000 elo means, iirc it's like top 5% or something and we used to have A LOT of vienna rounds in multiplayers games at some point. Some friends started playing with me and after a week or two he could easily plonk Vienna most of the time.
This is the problem with having pro players be the judge, they have been so good at the game for so long that they have no idea what lower rated players know or don't know. I might be 600 rated but I recognize Vienna street signs, German bollards, Mexican octo-poles, etc etc
Capital should never be sus and honestly most reasonably big cities are just knowable. I'm sure I've stared at a sign just to think of what language it sounds like the most, or at a bollard trying to remember the country it belongs to. Basing it on 1 guess seems almost impossible unless it's extremely obvious scripting.
I'm good at geography so I know a lot of places around the world, but I definitely don't recognise poles and such. It's funny how you think that someone knowing the state of Campeche is "pretty low".
for some cases it could be also people that are decently knowledgeable in geoguessr, but dont really play any duels, so they have low elo. so if this cheater detecting thing gets an update where it shows the person’s elo, dont completely base it on that
At 12:15 note that Babin Zub is a famous mountain in Serbia. Like this player I also instantly would have got it… seems unfair to call it cheating just because you don’t know the region.
The problem is that the player didn't instantly get it. They paused for quite a bit on the road sign, which is likely what led to the conclusion that they were googling.
@@hamburger2430 Fair enough, I was just more referring to him saying “hotel… river… yeah this is not gettable”. It’s perfectly gettable if you’ve spent some time in southern Serbia. Though I do agree it was probably a google anyway.
Hint: In Canada, when they don't list the area code on a phone number, then it's usually a province with only one area code. This is only New Brunswick or Newfoundland/Labrador now. (Yes, the far north territories are area code 867... but you shouldn't need area codes to help you there)
I don't think you should go off rating. I am very low rated just because I don't play competitive very often, but I know a lot of the more common metas (such as the Vienna street sign in the first round).
I would just like to say, I'm only in the low to mid 700s for elo. I know Vienna signs. I don't know why being below 1000 would make someone a cheater for knowing that.
I agree with being able to see more rounds though. Usually if they cheat once, they'll cheat twice within the same duel. Seen obvious googling too many times from people in my games, "Oh wow, you just happen to 5k every time there's a sign written in the Latin alphabet, but you can't tell the difference between India and Thailand. Huh."
Im about 900 elo and I agree with you. Why im not higher elo is because im so inconstant, I go Aus on South africa etc but I know Vienna signs also. looking at one round says nothing, the whole overwatch systems seems very flawed.
@@McKennasFeverDream Yeah, a lot of the mid to upper-mid elos seem to mostly be people that will know some bits stuff to determine a country or city, but not others. Like, I can almost always tell if a country if Peru, Chile, or Argentina. I can't even explain why for Peru or Chile tbh, I just kinda have a "feel" for them. Bolivia though? Fuck me, I don't know shit about Bolivia, and basically never guess it and will lose because of it.
Vienna street signs is one of the first things I have learned (well, and I've been to Vienna too) and remembering what districts are on the other bank was not too difficult too. I am consistently below 1000 rated. So just as a small reminder that sometimes niche knowledge is quite common and legit.
Same, also, 9.9 times out of 10, if it's in any way Austria, it's Vienna. I get a LOT of close Vienna guesses, but I suck at large portions of Europe/Southeast Asia/South America
I think the biggest takeaway from this video is that way, way too many people are reporting people for cheating whenever their opponent makes a good guess, and all that noise is gonna make it harder to catch the actual cheaters. Some of these are just ridiculous, like it's an obvious Taiwan and they weren't even particularly close, opponent was probably tilted and just reporting their opponent for no reason
Definitely agree that we need more context for these investigations like the rating, country and maybe also the other rounds with the reported rounds highlighted.
Devs should definitely make it that you could watch the entire game, it would give much more context.. And also we need to remember that amazing plonks and repeats do happen. And by just watching a lot geoguessr content and rarely have time to play I'm very underrated so rating wouldn't give necessarily better picture. For example just like 2 days ago I 2 rounded my opponent by getting the exact location in Russia that was in Jake Lyons game with his sister and 5k'd then next round immediately noticed California yellow stripes and thought that this looks so much like Sacramento where I was 5 years ago. Immediately zoomed in and chose first place that lines up and 5k. And just minutes later I got Malta location that was just down the road where I used to live some years ago.
True, sometimes you have a total "Slumdog Millionaire" game going on. I remember in one game it plopped me down at the exit I would always take to come back into my hometown. My rating sucks, I don't know a lot of metas because there are just so many different types of poles, similar bollards, etc, and I just don't have the time or inclination to memorize them all. But sometimes get lucky, be it a lucky guess, or a lucky location you've been to. I can usually ID my home state in the US just by road condition alone.
This feature has been so fun! It's cool seeing a few repeat rounds from other people's vids on this, glad they're getting a large # of people to judge before making any banning decisions.
The one where you said “They ‘might’ be googling” and proceed to click guilty, you need to eliminate the MIGHT before clicking guilty. Might should be insufficient evidence, you should be confident about a guilty verdict. Much love!
As im watching this when you are talking about what you think new players might be thinking its very wrong. you think like a high level player focusing on stuff that a new person wouldnt really care about. for me language is the best - i see æ and ø, i think denmark, norway, etc. i see š and other slovak letters i know its slovakia (š appears in many other slavic alphabets but its pretty easy to figure out for me by for example saying the text in my head to "hear" how it sounds and that way i can even sometimes understand it (im polish and we can _kinda_ read slovakian) or at least get a good idea if its slovakian, czech, or serbo-croat) many people might have a lot of real life knowledge, but not geoguesser knowledge - for example u called out that guy for guessing mexico even tho he didnt recognise mexico poles - well lets be honest there were a lot of things screaming mexico in there, and probably the southern part due to the climate
Exactly. I think ziggiezaggie is so far into the Geoguesser hole he forgets that other people may have other type of knowledge, such as linguistic/toponymical knowledge which often helps to guess the country or the region, or just plain old geographic knowledge.
I often stop in the middle of a low pressure round, could be to chat on signal on my second screen, get a drink, etc. It's not very long but still. Sometimes I just feel like taking my time...
I saw on other channel that Replay was bugged too, he was checking it 3-4x to see what is going on, and last was reply was correct with a sign. Googling for sure
On the "Capeche" round i would say it is pretty plausible. Maybe he thought the name was from south america and then remembered where it is from. I play a lot of team duels with my friend (currently gold 1 for better reference) and while i know a lot of bollards, car meta and poles he knows little to none, in other hand he is way better at vibes, landscape and city/state names so i would say diferent people know where to guess from very different reasons. And thats why this game is so fun
Pretty sure Campeche is me. The initial vibe was S. America for me and briefly hovered over Uruguay (probably all the blue) but kept looking. I know Mexican poles but forgot to look and had a mental block about where Campeche was. Near the end I saw the American pickups and short plates followed by a pole and it all made sense. I had to hurry because my opponent had guessed.
And then everyone on Reddit says “nah they’re just good and you suck” whenever somebody complains. It’s obvious when somebody doesn’t speak a random African language but find the exact village that they are cheating.
17:53 this is one of the ones where you wish you could see where they are playing from or their practice history. I don't think a low level player who isn't Australian or hasn't practiced it a lot would ever get this
This is where Oscar should have tried googling some of the things they saw. Like the person just looked away then googled "TNQ nursery" I just did it and it came right up. That person was definitely cheating. This the same story on the adelaide round 11:00, if you google "Brewer valley Rd" it instantly takes you right outside of Adelaide
window.addEventListener('keydown', checkForAltTab); // Now all you need is a checkForAltTab function to handle the event, read the keystrokes, and note that the event occured.
Your assumption that someone who doesn’t know Mexico poles doesn’t know where Campeche is crazy. Many people know geography and maps but not geoguessr meta.
I have an issue with Mexico apparently. It always seems to look like somewhere else. I knew where Campeche was and I knew the poles but had a mental block. Couldn't recall where Campeche was and forgot to check the poles until the end. I'm currently mid to high 800s but was mid 700s when I played the round.
I mean I live in México and I didn't know that our poles were different, but when i saw Conagua (the institution in charge of water) in te video I recognized it was mexico so i agree the pole thing is weird
Finally geoguessr is doing something about these damn cheaters. Getting to Master was so hard, I felt like someone was googling literally every other game lol
Saying that someone is suspicious for not knowing Mexican poles which means they surely wont know Mexican state names is... questionable. You can be good at general geography and still be a great geoguessr player without any meta knowledge
I'm sure that was me. I know both but forgot to check the pole since I saw the sign and went there first. I saw the pole right at the end. Being Canadian I've been to Mexico a few times and also knew where Campeche was but for some reason my recall was temporarily broken. That happens a lot at my age.
I think this is very easy to abuse if someone is annoyed at their opponent. Also someone's knowledge won't necessarily match up with their country. Geoguessr thought I was from the wrong continent the last time I went there. And there's people who know a lot about geography but not meta (ex new players).
@@RetsamX i don't know which part you didn't understand. He said it would be sus for players below 1000 and I said that it absolutely wouldn't as it is easy to learn meta. Like pink taxi or red city district signs for Prague
Ya same. My highest elo was around 980 when seasons were still a thing, but since the new ranking system, I'm currently at gold 1 floating around that same elo range as you. And I know basic metas like Vienna as well
Beside what you or others in the comments mentioned: 1) If you play on your mobile phone via app some movements might look strange/less smooth 2) Weak internet connection or ping might also cause that it looks like a skript is in use In most cases it's not possible to determine if someone is cheating from my point of view with just one round. There are definitely more rounds and/or additional information needed - but that is a lot of effort. The question is also if everyone should be able to review everyone else. Can a beginner properly evaluate a pro game? Or vice versa, can pros properly evaluate a beginner game? I like the approach, but to have a fair system to ban cheater we might need a little bit more than just one replay.
Until we upgraded our internet I was almost ready to stop playing because the moving was so bad and it often took too long for the image to get viewable.
Interesting stuff, there's definately good content in this. Watching some other YTubers doing the same thing I spotted some dupes of the recordings you got, so either the sample set is small or Geoguessr are doing a first pass on calibration using a curated set of problems. I think you were supposed to pick the default insufficient in general, but I liked the fact you gave most people the benefit of the doubt ;)
i feel like just having 1 round does not showcase if someone is cheating or not on the less obvious cases as i feel like we've all had insanely lucky guesses where we get really close just by pure luck and obviously without cheats, so full games should be given in these cheater overviews for it to be more valid imo
I'd imagine they have people to conduct a more holistic review of the player based on these reviews, not that anyone gets banned automatically just based on a round. It's just a way for the community to exclude those obviously false reports.
Interesting… I would assume that cheaters aren’t cheating every single round. But if someone got reported a few different times, it would be good if you could see all of the suspicious rounds.
Cheating is actually really common in low ranked games (Gold 3 Teams ranked) we had so many people google up stuff it was crazy. They looked at a City sign for like 10-15 seconds and Right away they go into the map and exactly in that City.
I noticed you got a lot of the same rounds as wook's video, so there is a limited pool to review. Evaluating just a single round is tough for all the reasons you mentioned, would be much easier to have an entire game to review and see repeated suspicious moments, etc.
not guilty and insufficient evidence should be the same option. you can never rule out cheating entirely, and the presumption of innocence means they're not guilty unless you can prove it. also, yes. a single guess is almost always inconclusive.
I think showing the whole duel on replay would be preferable. You wouldn’t even need to necessarily watch all of it you could scrub through but having access to it would be very helpful. Also knowing where the opponent guessed is helpful as well because it can help you tell if it’s a high level lobby based on the opponents guesses.
I think it's really hard to tell as they might just be thinking about it or chatting with someone. The only obvious one would be a perfect plonk on the exact place without good enough clues
the tasmania round was weird when reviewing it i also had them stare at a field near the sign, rewinding fixed it and they were staring at the sign the whole time for like 3/4 of the round which is a bit suspicious some rounds would benefit from a flag and rank
Zigzag in the first 1:20 in, I have a 594 rating in geoguessr and would've guessed the Vienna house signs correctly simply because I watch your videos.
It's funny how neither zi8gzag nor GeoPeter never selected Insufficient Evidence To be fair, there should only be two options: *probably cheating* and *probably not cheating*
I stopped watching GeoPeter after I saw him and his moderators go into another Twitch stream to practice guessing based on each others' distances. He got bored enough to cheat years ago. One of them, jbuuki, still has a top 10 score last I checked. And that kid was BLATANTLY cheating.
if you read a sign with a town name for 30 seconds with the map closed, and then open the map and immediately zoom into it, then thats sus. if you read the sign and were scanning for the town name, or already knew the town name, you wouldnt need to stare at the sign for 30sec with the map closed
On the 2nd round I believe he was googling as he zoomed in on the sign moved without panning to not look suspicious and a lot of pauses whilst moving which could be an indicator of him pulling out his phone and searching the street name where the result after pressing enter is there in selfoss and the map was centred around selfoss and also clicking Reykjavik to be less suspicious and doesn’t click on selfoss but somewhere around it
It would be great if the Geoguessr team could, at the very least, provide country of player and rank, it could definitely help. The only downside I could see is a player who watches Geoguessr videos all the time and studies the metas and all that, and then one day decides to make an account and gets really good guesses, he may be reported for cheating... but that would be such a small percentage (Edit: also I like this type of video! Would definitely like to see you review more!)
I'm confused of what you perceive a lower than 1000 ELO player might be capable, that is master ELO. I am 800-900 have multiple times achieved double digit country streaks NM and I would assume that of most gold and some silver players. Getting lucky guesses or even good region guesses is nothing special for "lower" rated players. High rated players are just much more consistent and often better fine tuned.
I don't get why they only show one round, seems like very few people could be convicted based on one round. For example i've had repeat rounds in duels before and instantly zoomed in on the right area which is not reflective of my skill level. This might look like scripting if that was the only round that was shown. Feels like it should show a whole duel.
I think it would be an interesting format if you would do like a goeguessr "guess the elo" like you watch a whole duel from 2 people and afterwards you have to guess what rank they are or their elo. Was just thinking of this when I saw you thinking about what rank the people were
regarding the south australien switch up you claimed as guilty. i do that often times too, in a bunch of countries - usually the bigger ones. I am a high gold / very very low master player, so got some knowledge but still absolutely suck haha
The only improvements I'd like to see to this system are the ability to view other rounds of the same person to determine if they're cheating. The fact you yourself said you'd be swayed by their rating is justification to leave it out IMO, after all, you want it to be as fair as possible, but at the same time you do need more than one round to determine it.
Tips: When the player just gets the place without zooming on the map and then switches it so it's less obvious it's most likely a scripter (the first example) If he's staring at a sign for over 7 seconds without zooming on the map or going further then it's a googler
3:45 what you called "mile markers" are actually "kilometer markers." the philippines uses metric when it comes to distances however, seeing those markers does not mean you're anywhere near manila. if you're in the luzon island, yes, the number indicates your distance in kilometers from luneta, manila however, some islands also have their own marker system as well i thought that the distances were only in reference to manila too. i was once thrown in a round with one marker with a number less than 500 written on it i know that baguio is 250 (in burnham park) and pagudpud is 550-ish, so i thought somewhere in between (or bicol maybe) but the letters confused me. It was "O" on one side and "LJ" on the other I was actually in Misamis Occidental. The letters meant Oroquieta and Lopez Jaena, respectively. The numbering used was the Mindanao system. There are other cases in some islands as well like Negros and Cebu So the moral of the story here is that the markers don't necessarily mean you're near Manila. Also pay attention to the letters because they tell you the town you're approaching. You might say, "aren't the letters ambiguous?" Yes. If you're lost and you don't know geography and the only thing you have is a marker, yes. But this is geoguessr, you can just tell "oh this looks pangasinan. the marker says T on one side and SN on the other. this must be between Tayug and San Nicolas, then"
only while watching this video I realized I kept clicking "insufficient evidence" instead of "not guilty" on the not guilty ones when I was doing my verdicts.
They should make it a 2d plotting graph where the points are insufficient evidence in the middle, guilty and not guilty on opposites of the x axis, and scripting and googling on opposites of the y axis
I did this myself earlier and found it very fun and addicting. I had these exact rounds but in a different order. It's a decent system but with buggy replays it can be hard to tell what's going on sometimes. Sometimes it can be hard to tell if a plonk is just lucky or scripted in just one round, so seeing multiple rounds of one game would be nice. Overall I was probably more aggressive with the guilty verdict though...
there’s nothing that pisses me off more than cheaters as someone who’s more or less amateur level who occasionally gets really good vibe guesses. i’m actively trying to get better and have to deal with getting trashed by people just using whatever they can to win.
In regard to the first one would have known the first was Vienna straight away and my rating is around 750. Also my wifi isnt the best and my laptop is awful so i usually dont move at all unless im totally stumped. I dont think that guy would need to be Austrian or over 1000 for that to not be considered cheating.
Just a interesting thing the round when you got the cheater on serbia where it said "hotel babin zub" hotel stands for hotel but babin zub is an actual place
I've seen people look at a sign, retain the information and then zoom in or idle without information in view while googling - it's a pretty 'smart' way to try and hide that you're googling lol, but still obvious in most cases. I thought the New Brunswick guy might've done that but who knows.
![Can I help Daniel find his Grandparents? [GEO DETECTIVE #27]](http://i.ytimg.com/vi/wTy5FQo87RQ/mqdefault.jpg)
![ลองสุ่มไอโฟน 990 โกงมั้ย? [ โกงมั้ยครับ ep.101 ] | DOM](http://i.ytimg.com/vi/bfsWe0j492k/mqdefault.jpg)
zigzag is guilty of not clicking insufficient evidence when he's not sure
To be fair, if you're a juror, normally, there isn't a "insufficient evidence" option, and "Not guilty" actually does not mean "Innocent". The verdict is used for when we think someone is innocent OR there isn't sufficient evidence to affirm the person is guilty. So if Geoguessr intended "Not guilty" to be synonymous with "Innocent", you could say it's an incorrect use of the term from a judicial point of view.
If having a completely normal guessing pattern equates to "insufficient evidence" then "not guilty" should be removed from the system entirely. I assume each option has a weight and I think his marking of "not guilty" was completely justified.
The system is flawed though.
@@Mercure250 yes but there is a "insufficient evidence" button and by not clicking it Rainbolt is essentially saying they are 100% not cheating
@@Mercure250 the intended use of it I think is if they're obviously googling but you couldnt tell if theyre also scripting. that's why I would primarily use it, anyway
@@pojiggle lol did you forget which channel you were on
7:44 "That's definitely insufficient evidence" - proceeds to click not guilty for both options
He did it multiple times when he was not certain.
The player saw the .sk domain and quickly clicked on the capital. You don't need a script to know that .sk is Slovakia and you don't need to google it either, he didn't even have time. Not guilty was 100% the right option. An insufficient evidence report would have been ridiculous for that round, it was the clearest non-cheating in the whole video.
In fairness that is how courts work and the legal mind of some people will simply gravitate to that.
@@paveladamek3502 Except that there are 3 options here; not 2 like in court.
7:44 "That's definitely insufficient evidence" - proceeds to click not guilty for both options
the first person literally had it space plonked on the 5k by the script and then they moved away from it to be less sus, was pretty obvious scripting imo
in hindsight i do agree
yep that's a scripter
@@zi8gzag man Oscar at least do insufficient evidence on that one😂 (and for others as well. You said "I'm gonna need more for that one" and put not guilty. If you need more evidence, you choose insufficient and only if you are 100% they are legit you can put "not guilty" )
@@RetsamX It's clear that for the first few cases, he did not read the categories correctly.
@@martian_heidegger if you compare it to Csgo, it's understandable since there are only 2 categories, insufficient evidence and evidence beyond most reasonable doubt. I don't know if I like the addition of "not guilty" and "guilty" but we will see.
Hi Zigzag! I was actually the guy who played that campeche round a few weeks ago so I thought I'd give you a bit of context and also my thoughts on the round. This was confirmed as it was the only investigation i didn't get in my lot of investigations.
I am a 1200 rated Australian player in Master 1 division that only plays intermittently, instead I mainly watch the rainbolt daily challenges and some videos from yourself on a regular basis.
I pretty much only play moving and base my guesses on vibes/feel, architecture, having a good understanding of languages and am able to use information to determine where I am. I have very little interest in studying poles / bollards / camera generations as it's simply boring in my opinion and has very little to do with actual geographical knowledge. Of course I know the most obvious / unique examples such as Peru/ Malay signs or Senegal poles or Gen 2 South Africa but am less interested in learning this as a whole.
My thoughts on this round:
This round was played on a high multi and actually won me the game with my opponent guessing by the coast west of Mexico City, obviously my opponent was a bit tilted and decided to report lol.
When coming to this sign, I read Campeche on it and remembered seeing this in a Rainbolt daily challenge from a few days earlier and knew that it was a state/province/city somewhere but I couldn't quite remember where.
But, the road, vegetation and architecture looked relatively similar to stuff I've seen in Uruguay / Northern Argentina but upon scanning the map and having another look at the surroundings - this clearly wasn't right. Now, remember I am not very familiar with specific poles / camera generations so did not know Mexico poles or that Uruguay was Gen 3 only, but that's definitely good info to know for the future.
So while reciting Campeche, Campeche, Campeche in my head trying to remember where it was I looked for more information.
Towards the end of the round after my opponent has guessed, I found a Nissan truck and a Chevrolet which are brands that I assosciate with Mexico and also saw the short plates on the truck which is more of a Mexican thing.
So with 6 seconds remaining, I initially zoom into the centre of the more green / tropical part of Mexico and fantastic I see the label for Campeche and guess right on the label with 3 seconds remaining and its pretty damn close.
I hope that I've provided some insight into this round :)
Youre 1200 rated without knowing Mexican poles? That sounds very unlikely to me
Im 900 rated and Ive known that meta from when I was waay lower. Its one of the most known meta's in the Americas.
Regardless, not saying youre a cheater 😅
That's strange because I'm absolutely sure that's me.
I checked the tower and assumed it was a company name. Checked the ambulance because a lot of them have town names. Saw Campeche on the sign (missed the town name) and hit a mental block about where it was and forgot to look at the poles. One was right there. It looked vaguely like Uruguay at first but I soon realized it wasn't.
As I looked around my opponent guessed and I saw the pickups with short plates, an octo-pole and realized it was Mexico. I had maybe 3-4 seconds left and rushed to Mexico.
My rating at the time was mid 700s to 800. I'd be better than that but my memory isn't what it was now that I'm 70.
Now I would like to see more info as I've been defending this all over the place.
For the record, I don't know what regions have which camera generations and can't tell gen 3 from gen 4. I do know some poles and bollards. I play the game a modest amount, usually just the 20 games a week in Gold but just went up to Master.
@@jasper4331 I think it's me. My rating then was under 800.
@wileym Haha that's pretty cool I think you likely had this round in a game as well as ~100000 rounds are in each map used for comp, but cycled through each month from a larger set of rounds.
The thing is I remember[ed] when writing this comment the specifics of what I did in the round - so based on the gameplay I remembered zooming inti uruguay, moving in pretty much that manner, looking at the number plate and zooming into Mexico in that manner after my opponent had guessed.
More importantly, this was the only investigation I didn't get.
There has been an update to the investigations tool which shows the elo range of the player when playing the round so we can 100% confirm based on that elo range if someone posts a new video showing this round.
It's fun to see your story as well :)
im like you in way, going into geoguesser as a geography nerd not super interesting in learning bolards.
You need to be putting insufficient evidence if you’re unsure, not “not guilty”. Deciding they’re not guilty when you’re even saying you’re unsure is going to multiply the amount of cheaters that get away with it.
I think that's the exact reason why there's a difference between an acquittal and a not guilty verdict in court. Comes down to reasonable doubt IIRC.
So true annoying to watch
So you'd never select not guilty then, since there's always a possibility they're cheating?
You need to be putting insufficient if you're unsure, not "not guilty". Deciding they're not guilty when you're even saying you're unsure is going to multiply the amount of cheaters that get away with it.
The first guy was surely scripting the way he put his marker on the 5k without zooming in, then moving off of it to look less suspicious.
Good spot! That will be one of the only ways to detect a scripter
I don't know, I'd honestly say insufficient evidence. That's not enough for me.
@@lukebradford nah that was blatant scripting
@@lukebradfordnah plonking the exact road without even zooming in on the map at all is very suspicious and extremely hard to do, even if you know it’s Vienna
@@lukebradford Normally, yes, but that was clearly a 5k without even opening the map
The idea of other players checking reported players is probably ok (especially for catching script users), but the problem is that high ranking players seem to play the game differently. They think about poles, bollards, phone area codes, transformers, google car colours and antennas and other irrelevant information. Of course, it's relevant for high ranking competion players, but not for casual players. We read the signs and try to guess the language or we try to find something useful from it. We check the road markings and the side they are driving. We check climate, vegetation, architecture, people, clothing etc. I think high ranking player use meta and make guesses based on very different information much more than more casual players.
I have watched two other videos about this new mode and all three show that high ranking players do not understand the way casual players (ranking 600-1000) play. One thing common for zi8gzag and other pros seem to be that they think that you have to zoom to correct country right after you know the country or place. I don't to that. If I find something I know, I keep looking around and moving trying to find more information. I open the map when: 1) I have to make a guess (other player guessed), 2) I know enough that I can guess something and don't want the other player to keep searching, or 3) I know something I have to check from the map (e.g. road sign which says Canberra 197, some other place 49, and I need to find that other place).
I work with satellite images (resolution 200-5000 m) and I have used google street view to check the surface condition (land cover type, vegetation, small scale topography). I have also read a lot (fiction and fact) and I know lot's of strange places which may not be familiar for typical high ranking players. I have travelled quite a lot (for work and leisure), so I have been in many places. I have also lived abroad and I have relatives in other countries. I have been interested in maps and geography all my life. I have been browsing google maps and google earth just for fun.
Of course I use meta when I know it (Vienna street signs, Nigerian police car, some Google cars, such as Ghana, Guatemala, Kirgistan, Kyrgyzstan), and I know some common places in the game (at least earlier, Monaco, Midway, Christmas Island), but my game is not based on meta. I check the scenery and try to find signs if possible. I do NOT google, but still I may use 10-15 seconds trying to decipher signs in foreign languages. I'm trying to find something I know. I think that some of my rounds may look suspicious for high ranking players because I look at things they think are irrelevant or stop to think too often.
Not everyone plays the game the same way. I am in that 600-1000 range but I absolutely use poles, bollards, and area codes and stuff when I know it. I also only play No Moving. I mean, the problem with this is thinking that everyone who isn't a pro plays like this or knows this but doesn't know that. I happen to know all 96 departments of France, for example. Most of the time it's not useful but if I see a sign with "Allier" written on it I know where to go.
The problem is that below the pro level, every body is just going to be very, very different in terms of the things they know and the things they don't. I absolutely knew that one round was Mexico from the poles, but didn't know the state. Other people might know the states but not the poles. Everyone learns different things so it's just wrong to say "a 600 wouldn't know that" or "600s play like this." That is why it would be very wrong for GG to include the rating in these replays. You can tell they are googling when they stare at a sign for 40 seconds while googling in another tab and then zoom straight for the exact spot, but basing your decision on anything other than the gameplay would be wrong in my opinion.
I was playing against someone who was much higher rated than me the other day and we got the Santo Domingo car with the roof rack and the tape running long across. I instantly clicked Santo Domingo while my opponent, again much higher rated, went Guatemala because they didn't know about the tape. You just don't know what other people know and what they don't.
this is decipherable though, if someone looks at a sign with a town name for awhile and then looks around for awhile and makes a bad guess, you can do not guilty/insufficient. if it's googling for alot of these it's quite clear, because they'll look at the sign for an extended period and then get a good guess. this is where knowing a players elo would also be imperitave
I've seen only two cases of obvious cheating. One we were in California and there was a sign with the town name on it. I scanned up and down the coast looking for the town but couldn't find it. My opponent stared at the sign perfectly still for 30 seconds then activated the map and zoomed straight into this small town. I made a close enough guess of LA or whatever but they got 4997 or something.
Second time there was a pole with some advertisement on it in English. I think I went in India somewhere but my opponent stared at this random sign for like 45 seconds then activated the map and plonked in Bangladesh.
I realized after the game that if you google the company slogan from the ad on the pole you get some company in Bangladesh.
In both cases the players ELO is irrelevant, it is just the gameplay that matters.
I reported both players and neither got banned fwiw, I think primarily because geoguesr is so overwhelmed with bogus reports like we see in this video.
the one thing I noticed was that he mentioned generations (gen3?) on the uruguay/mexico one which I mean it is a bit bold to assume that you have to learn camera generations to be able to make good guesses regardless of if that player may have ended up being a cheater
I paused and took a look at the report decision page, the not guilty should only be chosen if you’re confident that the person is not cheating, so I think a lot of “not guilty” decisions should maybe be insufficient evidence instead
Cool video though very interesting system they’ve got
The New Zealand round at 13:05 was a team duel. It's not indicated anywhere, but when you said that we can see the opponents guess, that was actually the team mate. You can also see the team mate giving a hint. (Also the last round in North Queensland is a team duel.)
Would also make the long pauses make sense if they are chatting with eachother for the last round
I think it is indicated by "Game Mode: Duels"
@@1977akom needs to be way more specific than that
That should be made far clearer in the interface as it affects how you play enormously - both in that you will sharing information and perhaps showing each other things on your screen if you are sitting next to each other, and also in terms of how you switch locations as you decide how to hedge. I could certainly see how teamplay could be confused for cheating - if often show each other signs, especially if one of us knows a language or country better.
The weird movement one in Canada felt to me like someone Canadian who was streaming. Like the pause at the lake felt like someone saying out loud "I should know this. I should know this. Don't tell me chat, I should know this."
0:48 the most obvious scripter: space plonks the 5k
zi8gzag: not guilty!
As a lower rated player, I have to say that the likelihood of 10:00 is not low. I don't care about poles. I don't know anything about Mexico's poles. Campeche on the other hand is written on the map. I've known about it for a long time because I watch GeoGuessr content and people always call out state names. One of my first 5ks in duels was in no moving at the Veracruz/Tabasco border. I'm yapping but the point is that knowing states happens infinitely earlier than knowing poles unless you're tryhard training to be an NMPZ main.
I know states of other countries from google maps and the Spanish states from Spanish class. That round made me concerned I would be accused of cheating.
i think the suspicious thing is that they immediately zoomed into uruguay, it's clear that they didn't know mexican states because of that
@@frijoless22 That part is weird, but I'm specifically responding to ZigZag's statement that it's unlikely for a player to know Campeche but not know Mexican poles.
As for zooming into the wrong country, I once thought Zacatecas was a city in Peru. Then after scanning Peru for a while I noticed that there were no Z's, prompting me to realize that it's Nahuatl and not Mayan. Misremembering and thinking you know that something is somewhere that it isn't is entirely possible.
As for cheating, why would he zoom into the wrong country if he was cheating? That makes even less sense.
yeah definitely more people know that Campeche is a region of Mexico than how Mexico poles look like
Known about Campeche since Sid Meyer's Pirates, were it was on of the big Spanish settlements in the New World.
1:13 Wdym not guilty, he space plonked on the 5k and then switched it to the other side of the river so it's less obvious lmao
@@EthanSSBthe guy clicked on the 5k without even zooming in, and then slightly changed his guess
@@notmeowth oh yeah never mind you're right
18:13 “he’s either Australian or cheating”… why not both?
I'm sure it was an inclusive "or"
a big reason why rating should NOT be shown is because it discriminates against people who just know a lot about the world but might not necessarily be big geoguesser players. i think rating shouldnt be shown as it just biases the reviewer into thinking "well hes high rank obviously he will know immiedetaly", while in fact that person might still be cheating, and the same "oh he's low rank so its impossible he got 5k so fast" while that person might just be fluent in many languages, be well traveled, know the various cultures, plants, etc. therefore even without "geoguessr knowledge" like google car, street poles, etc. they might find something faster than a pro player could depending on the situation. or someone simply could've physically been to the location of a photo - i once watched a geoguesser video where some guy randomly got streetview by a gas station thats like 15 min away from my house, so i definitely would've gotten a 5k if i got that one when playing.
Wich is why we also have to see the other rounds imo, you can correlate the two and make up your mind way better.
Kinda makes me anxious that I now have to think about how I play to not seem sus by accident lmao
In addition to what you listed, it's also possible someone plays a ton of non-duels and then decides to play some duels and already knows the metas. Geoguessr forces you to rank up pretty slowly when starting out.
On average you will catch more cheaters if you are biased.(rating, where they come from, other rounds, does he know metas or not, etc.) If you say you dont want a single falseban then you will have to let 95% of cheaters run free.
@@mursmumies123 i mean you had to before anyway - your games used to be reviewed by devs instead of both players and devs.
from what i understand, you won't be punished simply because a lot of players chose "guilty" on one of your guesses. it will simply bring your case more attention and be looked into further by the devs themselves.
if you aren't cheating, don't worry about it. and if you can't help but worry about it, maybe just slightly move your screen around when you're reading signs so it's obvious you're not tabbed out.
I really think they need to show more than one round. A really lucky guess once can just be really lucky. Five really lucky guesses in a row, and it's not luck that the guesses are so good
1:18 you can just click insufficient evidence if unsure
See 7:14
@@Okurka. ya saw later
"that's definitely insufficient evidence" *picks not guilty*
this video is a great example of why this "overwatch" type anticheat system is awful
The thing that scares me about this is that people like zigzag think people under 1000 elo don't know Vienna street signs
I think he doesn't realise what 1 000 elo means, iirc it's like top 5% or something and we used to have A LOT of vienna rounds in multiplayers games at some point. Some friends started playing with me and after a week or two he could easily plonk Vienna most of the time.
I'm only ~850 rated but I definitely meta game hard and have studied countries like Australia. These reviews are a bit spooky imo
This is the problem with having pro players be the judge, they have been so good at the game for so long that they have no idea what lower rated players know or don't know. I might be 600 rated but I recognize Vienna street signs, German bollards, Mexican octo-poles, etc etc
@@M4ke4lexactly, up until the recent map changes Vienna came up all the freaking time. It's pretty much the first city you learn to recognize
@@mursmumies123 Same here. I'm about 1000 elo in Masters II, and we all have good understandings of meta's, european languages, poles, etc.
Capital should never be sus and honestly most reasonably big cities are just knowable.
I'm sure I've stared at a sign just to think of what language it sounds like the most, or at a bollard trying to remember the country it belongs to.
Basing it on 1 guess seems almost impossible unless it's extremely obvious scripting.
1st one was obvious scripting
Reporters should be required to explain why they got to their decision so it will be easier for people to appeal.
please make this into a series!! this was a fun watch! i just hope geoguessr fixes things up!
please make this into a series!! this was a fun watch! i just hope geoguessr fixes things up!
I'm good at geography so I know a lot of places around the world, but I definitely don't recognise poles and such. It's funny how you think that someone knowing the state of Campeche is "pretty low".
for some cases it could be also people that are decently knowledgeable in geoguessr, but dont really play any duels, so they have low elo. so if this cheater detecting thing gets an update where it shows the person’s elo, dont completely base it on that
At 12:15 note that Babin Zub is a famous mountain in Serbia. Like this player I also instantly would have got it… seems unfair to call it cheating just because you don’t know the region.
The problem is that the player didn't instantly get it. They paused for quite a bit on the road sign, which is likely what led to the conclusion that they were googling.
@@hamburger2430 Fair enough, I was just more referring to him saying “hotel… river… yeah this is not gettable”. It’s perfectly gettable if you’ve spent some time in southern Serbia. Though I do agree it was probably a google anyway.
Hint: In Canada, when they don't list the area code on a phone number, then it's usually a province with only one area code. This is only New Brunswick or Newfoundland/Labrador now.
(Yes, the far north territories are area code 867... but you shouldn't need area codes to help you there)
came here to say this!
I don't think you should go off rating. I am very low rated just because I don't play competitive very often, but I know a lot of the more common metas (such as the Vienna street sign in the first round).
I would just like to say, I'm only in the low to mid 700s for elo. I know Vienna signs. I don't know why being below 1000 would make someone a cheater for knowing that.
I agree with being able to see more rounds though. Usually if they cheat once, they'll cheat twice within the same duel. Seen obvious googling too many times from people in my games, "Oh wow, you just happen to 5k every time there's a sign written in the Latin alphabet, but you can't tell the difference between India and Thailand. Huh."
Agree as someone around same elo as you
Im about 900 elo and I agree with you. Why im not higher elo is because im so inconstant, I go Aus on South africa etc but I know Vienna signs also. looking at one round says nothing, the whole overwatch systems seems very flawed.
just like to mention that im 700 elo and have no idea what a vienna sign is
@@McKennasFeverDream Yeah, a lot of the mid to upper-mid elos seem to mostly be people that will know some bits stuff to determine a country or city, but not others. Like, I can almost always tell if a country if Peru, Chile, or Argentina. I can't even explain why for Peru or Chile tbh, I just kinda have a "feel" for them. Bolivia though? Fuck me, I don't know shit about Bolivia, and basically never guess it and will lose because of it.
Vienna street signs is one of the first things I have learned (well, and I've been to Vienna too) and remembering what districts are on the other bank was not too difficult too. I am consistently below 1000 rated. So just as a small reminder that sometimes niche knowledge is quite common and legit.
Same, also, 9.9 times out of 10, if it's in any way Austria, it's Vienna. I get a LOT of close Vienna guesses, but I suck at large portions of Europe/Southeast Asia/South America
You need to consider the fact that they arent pros like you guys.
I think the biggest takeaway from this video is that way, way too many people are reporting people for cheating whenever their opponent makes a good guess, and all that noise is gonna make it harder to catch the actual cheaters.
Some of these are just ridiculous, like it's an obvious Taiwan and they weren't even particularly close, opponent was probably tilted and just reporting their opponent for no reason
Instead of saying not guilty when you're unsure, I recommend just clicking insufficient evidence.
16:30 LOL never thought I'd see my own game featured in here
Definitely agree that we need more context for these investigations like the rating, country and maybe also the other rounds with the reported rounds highlighted.
Devs should definitely make it that you could watch the entire game, it would give much more context.. And also we need to remember that amazing plonks and repeats do happen. And by just watching a lot geoguessr content and rarely have time to play I'm very underrated so rating wouldn't give necessarily better picture. For example just like 2 days ago I 2 rounded my opponent by getting the exact location in Russia that was in Jake Lyons game with his sister and 5k'd then next round immediately noticed California yellow stripes and thought that this looks so much like Sacramento where I was 5 years ago. Immediately zoomed in and chose first place that lines up and 5k. And just minutes later I got Malta location that was just down the road where I used to live some years ago.
True, sometimes you have a total "Slumdog Millionaire" game going on. I remember in one game it plopped me down at the exit I would always take to come back into my hometown. My rating sucks, I don't know a lot of metas because there are just so many different types of poles, similar bollards, etc, and I just don't have the time or inclination to memorize them all. But sometimes get lucky, be it a lucky guess, or a lucky location you've been to. I can usually ID my home state in the US just by road condition alone.
This feature has been so fun!
It's cool seeing a few repeat rounds from other people's vids on this, glad they're getting a large # of people to judge before making any banning decisions.
7:52 "Sundhedshus" is a type of hospital and "amager" is the city :) sunhed = health hus = house
Amager is not the city. It's the island south east of Copenhagen.
The one where you said “They ‘might’ be googling” and proceed to click guilty, you need to eliminate the MIGHT before clicking guilty. Might should be insufficient evidence, you should be confident about a guilty verdict. Much love!
As im watching this when you are talking about what you think new players might be thinking its very wrong. you think like a high level player focusing on stuff that a new person wouldnt really care about.
for me language is the best - i see æ and ø, i think denmark, norway, etc. i see š and other slovak letters i know its slovakia (š appears in many other slavic alphabets but its pretty easy to figure out for me by for example saying the text in my head to "hear" how it sounds and that way i can even sometimes understand it (im polish and we can _kinda_ read slovakian) or at least get a good idea if its slovakian, czech, or serbo-croat)
many people might have a lot of real life knowledge, but not geoguesser knowledge - for example u called out that guy for guessing mexico even tho he didnt recognise mexico poles - well lets be honest there were a lot of things screaming mexico in there, and probably the southern part due to the climate
Exactly. I think ziggiezaggie is so far into the Geoguesser hole he forgets that other people may have other type of knowledge, such as linguistic/toponymical knowledge which often helps to guess the country or the region, or just plain old geographic knowledge.
I often stop in the middle of a low pressure round, could be to chat on signal on my second screen, get a drink, etc. It's not very long but still.
Sometimes I just feel like taking my time...
If the panning is off, a lotta innocent people are gonna get flagged as cheating imo. They gotta sort that out.
Interesting I had the case at @15:20 and it showed me the correct panning of him staring at a sign that said "Evansdale"
I saw on other channel that Replay was bugged too, he was checking it 3-4x to see what is going on, and last was reply was correct with a sign. Googling for sure
On the "Capeche" round i would say it is pretty plausible. Maybe he thought the name was from south america and then remembered where it is from.
I play a lot of team duels with my friend (currently gold 1 for better reference) and while i know a lot of bollards, car meta and poles he knows little to none, in other hand he is way better at vibes, landscape and city/state names so i would say diferent people know where to guess from very different reasons. And thats why this game is so fun
Pretty sure Campeche is me. The initial vibe was S. America for me and briefly hovered over Uruguay (probably all the blue) but kept looking. I know Mexican poles but forgot to look and had a mental block about where Campeche was. Near the end I saw the American pickups and short plates followed by a pole and it all made sense. I had to hurry because my opponent had guessed.
Glad they added this cuz holy crap the previous system was bad.
And then everyone on Reddit says “nah they’re just good and you suck” whenever somebody complains. It’s obvious when somebody doesn’t speak a random African language but find the exact village that they are cheating.
17:53 this is one of the ones where you wish you could see where they are playing from or their practice history. I don't think a low level player who isn't Australian or hasn't practiced it a lot would ever get this
This is where Oscar should have tried googling some of the things they saw. Like the person just looked away then googled "TNQ nursery" I just did it and it came right up. That person was definitely cheating. This the same story on the adelaide round 11:00, if you google "Brewer valley Rd" it instantly takes you right outside of Adelaide
They should add a way for you to see when the player tabs out of the GeoGuessr window, which is something they should be able to track.
window.addEventListener('keydown', checkForAltTab); // Now all you need is a checkForAltTab function to handle the event, read the keystrokes, and note that the event occured.
They should share ppl rating/country with you, it’d make it easier to judge
Your assumption that someone who doesn’t know Mexico poles doesn’t know where Campeche is crazy. Many people know geography and maps but not geoguessr meta.
I have an issue with Mexico apparently. It always seems to look like somewhere else. I knew where Campeche was and I knew the poles but had a mental block. Couldn't recall where Campeche was and forgot to check the poles until the end.
I'm currently mid to high 800s but was mid 700s when I played the round.
I mean I live in México and I didn't know that our poles were different, but when i saw Conagua (the institution in charge of water) in te video I recognized it was mexico so i agree the pole thing is weird
Finally geoguessr is doing something about these damn cheaters. Getting to Master was so hard, I felt like someone was googling literally every other game lol
Saying that someone is suspicious for not knowing Mexican poles which means they surely wont know Mexican state names is... questionable. You can be good at general geography and still be a great geoguessr player without any meta knowledge
I'm sure that was me. I know both but forgot to check the pole since I saw the sign and went there first. I saw the pole right at the end. Being Canadian I've been to Mexico a few times and also knew where Campeche was but for some reason my recall was temporarily broken. That happens a lot at my age.
I think this is very easy to abuse if someone is annoyed at their opponent. Also someone's knowledge won't necessarily match up with their country. Geoguessr thought I was from the wrong continent the last time I went there. And there's people who know a lot about geography but not meta (ex new players).
This was a really fun watch, would love to see another!
1:20 below a thousand? Vienna meta is one of the few first i learned. And I have a 600-700 range rating.
600 is below a thousand
@@RetsamX i don't know which part you didn't understand. He said it would be sus for players below 1000 and I said that it absolutely wouldn't as it is easy to learn meta. Like pink taxi or red city district signs for Prague
@@arekzawistowski2609 well I was watching the video and didn't stop to skip to your timestamp so it seemed weird.
@@RetsamX okay
Ya same. My highest elo was around 980 when seasons were still a thing, but since the new ranking system, I'm currently at gold 1 floating around that same elo range as you. And I know basic metas like Vienna as well
13:53 "I think there is insufficent evidence [...]"
Clicks on not guilty 😂
Beside what you or others in the comments mentioned:
1) If you play on your mobile phone via app some movements might look strange/less smooth
2) Weak internet connection or ping might also cause that it looks like a skript is in use
In most cases it's not possible to determine if someone is cheating from my point of view with just one round. There are definitely more rounds and/or additional information needed - but that is a lot of effort.
The question is also if everyone should be able to review everyone else. Can a beginner properly evaluate a pro game? Or vice versa, can pros properly evaluate a beginner game?
I like the approach, but to have a fair system to ban cheater we might need a little bit more than just one replay.
Until we upgraded our internet I was almost ready to stop playing because the moving was so bad and it often took too long for the image to get viewable.
Definitely do more of this, this is actually so fun to watch
Interesting stuff, there's definately good content in this. Watching some other YTubers doing the same thing I spotted some dupes of the recordings you got, so either the sample set is small or Geoguessr are doing a first pass on calibration using a curated set of problems. I think you were supposed to pick the default insufficient in general, but I liked the fact you gave most people the benefit of the doubt ;)
i feel like just having 1 round does not showcase if someone is cheating or not on the less obvious cases as i feel like we've all had insanely lucky guesses where we get really close just by pure luck and obviously without cheats, so full games should be given in these cheater overviews for it to be more valid imo
I'd imagine they have people to conduct a more holistic review of the player based on these reviews, not that anyone gets banned automatically just based on a round. It's just a way for the community to exclude those obviously false reports.
0:29 valve removing overwatch was the worst decision they ever made
I just reviewed someone who looked at this sign in nevada, stood still for a good 30 seconds then zoomed into the 5k lmaooo
Interesting… I would assume that cheaters aren’t cheating every single round. But if someone got reported a few different times, it would be good if you could see all of the suspicious rounds.
They should just track the mouse pointer
If it goes up past the screen or two where a second monitor is that’s a lot of evidence for Google
Cheating is actually really common in low ranked games (Gold 3 Teams ranked) we had so many people google up stuff it was crazy. They looked at a City sign for like 10-15 seconds and Right away they go into the map and exactly in that City.
I noticed you got a lot of the same rounds as wook's video, so there is a limited pool to review. Evaluating just a single round is tough for all the reasons you mentioned, would be much easier to have an entire game to review and see repeated suspicious moments, etc.
OR they are testing people with a set number of rounds to see how close they are
There are not a lot of rounds to review. I went through the whole queue in like half an hour or so
@@NoThankUBeQuiet It would make sense considering they're only letting high rated players judge right now.
I got the same rounds too
not guilty and insufficient evidence should be the same option. you can never rule out cheating entirely, and the presumption of innocence means they're not guilty unless you can prove it.
also, yes. a single guess is almost always inconclusive.
5:42 I think that new brunswick round was me lol either that or it was a repeat and I did pretty much the same things somehow
What's crazy is I thought this was my game too. I've definitely had that exact round but I don't remember where I or my opponent guessed for it.
Some people set wrong countries to their profile specifically not to give away these information to the opponent.
I think showing the whole duel on replay would be preferable. You wouldn’t even need to necessarily watch all of it you could scrub through but having access to it would be very helpful. Also knowing where the opponent guessed is helpful as well because it can help you tell if it’s a high level lobby based on the opponents guesses.
I think it's really hard to tell as they might just be thinking about it or chatting with someone. The only obvious one would be a perfect plonk on the exact place without good enough clues
This was fun to watch .. would love to see more!
the tasmania round was weird when reviewing it i also had them stare at a field near the sign, rewinding fixed it and they were staring at the sign the whole time for like 3/4 of the round which is a bit suspicious
some rounds would benefit from a flag and rank
Zigzag in the first 1:20 in, I have a 594 rating in geoguessr and would've guessed the Vienna house signs correctly simply because I watch your videos.
It's funny how neither zi8gzag nor GeoPeter never selected Insufficient Evidence
To be fair, there should only be two options: *probably cheating* and *probably not cheating*
I stopped watching GeoPeter after I saw him and his moderators go into another Twitch stream to practice guessing based on each others' distances. He got bored enough to cheat years ago. One of them, jbuuki, still has a top 10 score last I checked. And that kid was BLATANTLY cheating.
i think you should have the whole match to analyze, not only one round
I'm now scared to play duels. I like to read signs. So if I was reading a sign and knew the answer and get reported, what then
if you read a sign with a town name for 30 seconds with the map closed, and then open the map and immediately zoom into it, then thats sus. if you read the sign and were scanning for the town name, or already knew the town name, you wouldnt need to stare at the sign for 30sec with the map closed
Hmmm... What if someone reports me? Oh, I'm clicking Africa when it's Brazil so no :)
11:25 no, not a scripter i would've honestly done the same thing
script just gets the 5k and then the player switches it so it's less obvious
100% agreed, if I were that player and I got banned for getting a lucky rural guess then I'd be so angry.
On the 2nd round I believe he was googling as he zoomed in on the sign moved without panning to not look suspicious and a lot of pauses whilst moving which could be an indicator of him pulling out his phone and searching the street name where the result after pressing enter is there in selfoss and the map was centred around selfoss and also clicking Reykjavik to be less suspicious and doesn’t click on selfoss but somewhere around it
Fun video idea would watch more of these for sure!
you need to turn this into a series (now that you also know about the insufficient data option lol)
It would be great if the Geoguessr team could, at the very least, provide country of player and rank, it could definitely help. The only downside I could see is a player who watches Geoguessr videos all the time and studies the metas and all that, and then one day decides to make an account and gets really good guesses, he may be reported for cheating... but that would be such a small percentage
(Edit: also I like this type of video! Would definitely like to see you review more!)
I'm confused of what you perceive a lower than 1000 ELO player might be capable, that is master ELO. I am 800-900 have multiple times achieved double digit country streaks NM and I would assume that of most gold and some silver players. Getting lucky guesses or even good region guesses is nothing special for "lower" rated players. High rated players are just much more consistent and often better fine tuned.
I don't get why they only show one round, seems like very few people could be convicted based on one round. For example i've had repeat rounds in duels before and instantly zoomed in on the right area which is not reflective of my skill level. This might look like scripting if that was the only round that was shown. Feels like it should show a whole duel.
I think it would be an interesting format if you would do like a goeguessr "guess the elo" like you watch a whole duel from 2 people and afterwards you have to guess what rank they are or their elo. Was just thinking of this when I saw you thinking about what rank the people were
He did a series like that before!
regarding the south australien switch up you claimed as guilty. i do that often times too, in a bunch of countries - usually the bigger ones. I am a high gold / very very low master player, so got some knowledge but still absolutely suck haha
usually i move away from the right guess though hahaahha
The only improvements I'd like to see to this system are the ability to view other rounds of the same person to determine if they're cheating.
The fact you yourself said you'd be swayed by their rating is justification to leave it out IMO, after all, you want it to be as fair as possible, but at the same time you do need more than one round to determine it.
Tips:
When the player just gets the place without zooming on the map and then switches it so it's less obvious it's most likely a scripter (the first example)
If he's staring at a sign for over 7 seconds without zooming on the map or going further then it's a googler
3:45 what you called "mile markers" are actually "kilometer markers." the philippines uses metric when it comes to distances
however, seeing those markers does not mean you're anywhere near manila.
if you're in the luzon island, yes, the number indicates your distance in kilometers from luneta, manila
however, some islands also have their own marker system as well
i thought that the distances were only in reference to manila too. i was once thrown in a round with one marker with a number less than 500 written on it
i know that baguio is 250 (in burnham park) and pagudpud is 550-ish, so i thought somewhere in between (or bicol maybe)
but the letters confused me. It was "O" on one side and "LJ" on the other
I was actually in Misamis Occidental. The letters meant Oroquieta and Lopez Jaena, respectively. The numbering used was the Mindanao system.
There are other cases in some islands as well like Negros and Cebu
So the moral of the story here is that the markers don't necessarily mean you're near Manila. Also pay attention to the letters because they tell you the town you're approaching.
You might say, "aren't the letters ambiguous?" Yes. If you're lost and you don't know geography and the only thing you have is a marker, yes.
But this is geoguessr, you can just tell "oh this looks pangasinan. the marker says T on one side and SN on the other. this must be between Tayug and San Nicolas, then"
only while watching this video I realized I kept clicking "insufficient evidence" instead of "not guilty" on the not guilty ones when I was doing my verdicts.
yessss i was so hoping you’d do one of these videos
They should make it a 2d plotting graph where the points are insufficient evidence in the middle, guilty and not guilty on opposites of the x axis, and scripting and googling on opposites of the y axis
I did this myself earlier and found it very fun and addicting. I had these exact rounds but in a different order. It's a decent system but with buggy replays it can be hard to tell what's going on sometimes. Sometimes it can be hard to tell if a plonk is just lucky or scripted in just one round, so seeing multiple rounds of one game would be nice. Overall I was probably more aggressive with the guilty verdict though...
there’s nothing that pisses me off more than cheaters as someone who’s more or less amateur level who occasionally gets really good vibe guesses. i’m actively trying to get better and have to deal with getting trashed by people just using whatever they can to win.
i love this! please make it a series
In regard to the first one would have known the first was Vienna straight away and my rating is around 750. Also my wifi isnt the best and my laptop is awful so i usually dont move at all unless im totally stumped. I dont think that guy would need to be Austrian or over 1000 for that to not be considered cheating.
bruv, "insufficient evidence" is an option for a reason
Just a interesting thing the round when you got the cheater on serbia where it said "hotel babin zub" hotel stands for hotel but babin zub is an actual place
I've seen people look at a sign, retain the information and then zoom in or idle without information in view while googling - it's a pretty 'smart' way to try and hide that you're googling lol, but still obvious in most cases. I thought the New Brunswick guy might've done that but who knows.