Who is the Best Classic Tetris Player in the World?

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
the 2022 classic Tetris World Championships are happening live in Portland Oregon on October 14th to 16th to determine the best Tetris player in the world so the big question is who's gonna win here's a hint it's almost guaranteed to be one of these six players these are the big six of Tetris but who's the best out of the big six in this video I'm going to use a ton of never before seen statistics and data to try and answer that question and make the best prediction possible [Music] so first I need to establish why I'm narrowing it down to a debate between these six players because on the surface it seems like there will be a lot of contenders to win the World Championships the main bracket for the tournament has 48 spots available and will feature tons of players from both the United States and all around the world as a new player fundraising initiative was a smashing success this year crowdfunding a trip to Portland for every player on the fundraising site but there's something that sets six particular players apart in the classic Tech Tetris monthly Masters event the most prestigious monthly tournament in the scene that features the top 16 qualifying players in the world going head-to-head every month credit by the way goes to Grand designs for first noticing this in the Discord servers in the past five months of the Masters event the players Eric icx dog playing Tetris Frac the 161 huffaluffagus Alex T and pixel Andy have gone undefeated versus everyone else in the world and only lost to each other as a result the semifinals of each month has always consisted of four of these six players this is a truly incredible streak of dominance so given how these players have mopped the floor with the rest of the scene the question I wondered was how do these players perform against each other when the pressure is at the highest for my last video which was about if Tetris matches are getting too long nowadays I gathered statistics on over 1 000 individual games from the Masters event over the past year because previously there had only been a record of who won in you lost this allowed me to track how much longer the tournaments as a whole were getting but since I also kept track of each game by player I can pull out every single game the big six played against each other and compare their performances across several key statistics to start let's give these six players each about a 15 second introduction Eric icx is best known for holding the current overall score world record in ntsc NES Tetris with over 6 million points which he actually achieved live in the August 2022 Masters event he won the New York World Championships qualifier this year and his best performance over the past five months in the Masters event was making the finals in June dog playing Tetris is best known for being the two-time tetris world champion winning each of the past two years in 2020 and 2021 he won the Waco Texas World Championships qualifier this year and his best performance in the Masters event in the past five months was making the finals in July fractal 161 is best known for being the first person to reach the levels with glitched colors in palinius tattoo actress he won the Lone Star qualifier this year and his best performance over the past five months was winning the Masters event in July and August Alex T is best known for being the greatest hyper Tapper in the world before recently moving to the rolling playstyle full time and was the last of the big six to make the switch he won the SoCal qualifier earlier this year and his best performance in the past five months was making the finals in the Masters event in September huffaluffagus is best known for finishing second in the 2021 World Championships he is the only player to have won two qualifiers this year winning both the Vegas qualifier and the Ohio qualifier and his best performance in the past five months was winning the Masters event in September and finally pixel Andy dog's older brother is best known for finishing second in the world championships in 2020. he is the only one of the big six not to win a qualifier this year but came close making the finals of the Lone Star qual and aside from fractal is the only other person to win The Masters event twice in the past five months taking the title home in man June our right on to the stats of the big six versus each other let's start with the simplest one winning percentage for this I counted each individual game rather than just match victories each match is a best of five game series so a sweep would count as three wins and zero losses and here's the results fractal and pixel Andy take the top two spots no surprise there as they both have two recent titles but surprisingly the two-time World Champ dog playing Tetris ranks dead last with a win rate of only 31.8 percent he is seemingly really struggled against his peers in recent months but is this an accurate ranking of the top six well wins are only part of the story the next stat I looked at was the median level 29 transition score the reason this is important is that for top players the levels before 29 are typically where almost all of the points are scored since the speed is much slower therefore there's a focus on playing efficiently with as many tetrises and as few singles and doubles as possible so who has done the best here are the results pixel Andy comes out on top as the most efficient player but Hoff who is in the middle of the pack and win rate actually comes in last however the difference between all these players is actually not that big it's only about one to two tetrises depending on what level all the tetrises are scored on the next step I looked at was the percentage of games that reached level 29 in the first place this measures the consistency of how often the players are able to get to the level 29 transition which is really important because topping out early against one of these other players is basically handing them an automatic win however there's a problem here in that typically when a player's opponent tops out very early they will Top out on purpose as well as soon as they're ahead in score because obviously they don't need to play anymore to win counting these games isn't really fair because we have no idea how long they would have survived Beyond this so I didn't include any time a player didn't get to 29 because their opponent topped out early and here's the results fractal takes the number one spot with a remarkable 85 again not surprising and pixel Andy unexpectedly comes in last tied with Huff but again all these players are pretty darn close to each other we need to find some stats where we can really see the difference between these players so the next thing I looked at was median lines past level 29. this is significant because after level 29 in competition strategy changes a lot there's now unlimited Runway since level 29 is the last speed increase in the game but tetrises are now much more difficult players have to balance playing safe and getting singles and doubles as long as they can with playing aggressive and trying to set up for the occasional Tetris whenever possible without making the board too risky so pure amount of lines of survival is one important half of the equation now I'm gonna be looking at several stats that involve playing past level 29 so it's worth addressing a certain complication up front there are roughly three types of ways a game can end here a natural top out an intentional top out and an aggressive top out a natural top out is when the player tops out with without trying to every game a player loses is a natural top out of course but games where the player wins is tricky sometimes a player keeps playing as long as possible to try for a high score this is known as mullening in the tetris community and is bad for tournament schedules if players do this every game to the point where Mullins are actually banned in the world championships this year but ironically it's good for statistics if players do mow in every game because that way we know how much they really could have scored because again sometimes after level 29 players will Top out on purpose after they've won which is an intentional top out and the third top out type is the aggressive top out this is where a player who's already won by playing somewhat safe starts suddenly playing Super aggressively and going for risky tetrises basically guaranteeing they're going to top out as soon as they encounter a drought this is seen as kind of a compromise between the player and the tournament organizer between moaning and intentionally topping out so what's the best way to measure a player here should I only include the natural top outs and take out the aggressive of an intentional ones unfortunately I don't think it's that simple because with the aggressive winning top outs a lot of them actually resemble natural losing top outs where one player gets really far behind and feels forced to start going super aggressive and with the intentional top outs some of them happen really far along into games for instance fractal's longest game during this stretch which got to level 60 and was a competition world record at the time is technically an intentional top out unlike a very early intentional top out on level 18 which is obviously a useless game to measure it would be incredibly unfair to discard this game as an unusable data point so how do I judge a game like this fairly this isn't a problem unique to Esports many traditional Sports such as baseball have an entire field of study known as saber metrics that uses fancy math to try and create context-independent statistics from heavily context-based data points but given that I have a short deadline for this video As the World Championships is happening in just a few days I will leave it to the tetris sabermetrics folks to figure figure out a more sophisticated equation in the future for measuring post 29 play and today I'm just going to include all the data points from every game that caught past level 29 and let the large sample size hopefully sorted out that's also the reason I'm using the median rather than the average unlike an average the median ignores the outliers on the top and bottom and just focuses on what a player can get at least 50 percent of the time so let's get to the results already Huff as it turns out is the survival champ lasting an astounding 71 lines past level 29 half the time with fractal not far behind but suddenly these stat methods are seemingly immediately put on trial with Eric all the way down at the bottom at 26 lines your reaction is probably really the current world record holder against top players gets less than 26 lines half the time surely he's got to have a bunch of early intentional top outs tanking the stats that was my concern as well so it's time for a stat audit I went back through all 21 of Eric's applicable games here to see what kind of top outs they were and well outlier games can be deceiving the vast majority of Eric's games were in fact natural top outs Eric's world record game had him surviving a ridiculous 1178 lines and scoring over 5 million points past level 29 but he also had games like this where he had a bad board going into level 29 and topped out right away and on the extreme end of the Spectrum this one game where he got zero lines and points past level 29 but won the game anyway overall if we look at where the median score is at 26 lines there are three intentional or semi-intentional top-outs above the mark and three below if we took those games out his median lines and score would be exactly the same the next stat I looked at was median score past level 29. this is important because when you compare with median lines it shows how much aggression players typically put into their post 29 runs and if it paid off and here's the results foreign the two-time World Champ dog takes the number one spot with a hundred and sixty two thousand just edging out Alex t as the most aggressive player on level 29 which is kind of crazy as it's been Alex T who most has the reputation for being an aggressive player after 29 and huff who was first in lines comes fourth in score out of the six players Huff's strategy leans the most towards pure lining out the next Sally looked at was median overall lines this basically amounts to measuring a combination of a player's consistency reaching 29 and how far they get on 29 afterwards and here's the results since everybody has a pretty similar consistency reaching 29 Huff's advantage in lines post 29 nabs him the top spot 50 of his games reach at least level 33. meanwhile at the bottom Eric again suffers from his low median line count past 29. the next step I looked at was the median overall score this basically amounts to a combination of the median 29 transition score and post 29 score whose strategy before and after 29 works out best together and here's the results Alex T ends up taking the top spot with his median heavily strengthened by the fact that some of his lowest 29 transition scores were followed up by some of his highest post-29 scores this is the stat I find the most interesting because at the end of the day it's all about scoring the highest and Alex T and dog who finished fourth and sixth in winning percentage here come in first and third suggesting that they both just had really unfortunate luck with a lot of their best games coming when their opponents also had their best games and finally just for fun since it's the high scores that most people remember let's take a look at everyone's high scores here's the results no surprise here Eric takes the top spot by a country mile the most astounding aspect of this stat is actually that fractal and dog the two spots at the bottom had both of their best scores happen in the same game but again here's the thing about high scores how well do they reflect that someone can win any particular game pulling back the chart of everyone's game's combined measured by scoring lines I have the outlier games on this chart clipping at the edges so you can see where the typical distribution is in the center what happens if I show this full chart for what it really looks like there's Eric's game all the way in the top right it's an unbelievable outlier so ultimately how do these six players compare here's a compilation of how they rank across every category they all rank number one in something and you can make a unique case for any of them like fractals the most consistent Andy is the most efficient Huff is the longest lasting dog is the most aggressive Alex T is the highest scoring and Eric has the highest potential however when I look at this compilation if I take out the win loss right record in high score and only focus on the rate base stats Alex T is the winner here he ranks in the top three in everything closely followed by fractal who only just falls short of the same achievement by ranking fourth in Total Lines fittingly enough Alex T and fractal are also currently the top two in the world on the ELO charts which ranks players based on their PB and who they've won and lost against so I find these stats a nice affirmation that these two really are most likely currently the best in the world so if I have to pick someone my final prediction for the 2022 classic Tetris World Championships is that the finals are between fractal and Alex T and Alex T barely wins but again it's worth pointing out that it is a close contest between these six all of them get to level 29 at least 80 percent of the time and all of them have a median score of over 1.1 million in a single elimination bracket I wouldn't be surprised if any of them end up taking the title and I'm not counting the other players completely out just yet either it's quite possible that the big six's undefeated streak gets broken this weekend by a pick up sets as at least one seems to happen every year so if you'd like to watch the world championships this weekend the basic schedule is that Friday will be qualifying day where each player has two hours to rack up as many Max outs as possible to determine seating many of the qualifiers will be live streamed on the classic Tetris twitch Channel and you can check out the qualifier signup page to see when your favorite players will be qualifying Saturday will feature a few final calls in the morning and then have the top 48 gold bracket start along with a 32 player silver bracket for those who didn't make the top 48 and in a change of plans it's been confirmed that every match in the gold bracket will be best of five unlike the best of three which had tentatively been announced and I thought would be one of the biggest time saving measures in my last video about tournament length but it looks like the tournament should be okay as they have a lot more stations and streams for matches this year Sunday will feature the final showdown of the top players from the silver and gold brackets and if you're not there in person you can still experience being part of the crowd by joining an online watch party on the ctwc Discord servers which I'll link in the description it should be an amazing spectacle I'll be at the event so if you see me there feel free to say hi thanks for watching and I'll see you guys next time [Music]
Info
Channel: aGameScout
Views: 72,286
Rating: undefined out of 5
Keywords: agamescout, nes tetris, classic tetris, agamescout tetris, classic tetris world championship, ctwc, best tetris player, best tetris player in the world
Id: G9uINGUdBP4
Channel Id: undefined
Length: 16min 7sec (967 seconds)
Published: Wed Oct 12 2022
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.