So, I'm an engineer, not a statistician, nor am I a good Goat player, but I'd like to add my two cents. There's a few points I think should be noted in this analysis. You did say on multiple occasions that many decks have low sample size, but I'd like to add to that point and say that we could maybe benefit from calculating confidence intervals. Like I said, I'm no statistician, so my calculations are based on a few formulas I found on google, but we have to start somewhere. Using something called a "Binomial proportion confidence interval" and using a confidence level of 80%, I calculated that the best deck in the format is... Chaos Return with a whopping 100% chance to make top cut! Ok, ok, I'm joking a bit, a sample size of 1 deck is not enough to reach any meaningful statistical conclusion, but it does put in perspective how much more meaningful Chaos Turbo results are in comparison with Warriors (8,7% to 2,9%), and besides Chaos Return, Goat Control got a very impressive 23,9%, while something like Earth Aggro has only 4,7%, lower than Chaos Turbo. Those percentages roughly mean the chance the deck has to make top cut with a 80% confidence. That sort of analysis tends to be a lot better than just looking at conversion rates if you have enough data, which in this case is a little bit of a problem. This sort of statistical analysis really doesn't work well with low sample size data. It could be better if instead of looking only at worlds we looked at the whole tournament cycle leading to worlds as well, since there would be more data, or if we looked at win rates rather than top percentages. I'd like to note that it's usual in engineering to use 95% confidence level, but I reduced to 80% due to the small sample size. Maybe someone with a better background than me in statistics can give some better tools for dealing with low sample size data. With that said, I'd also like to add a few extra things. There were 76 players in the tournament, so only 21,1% of players would make top cut. I do believe that's a important point to note, because it's hard to tell how meaningful a 16,7% representation of decks like Chaos Turbo and Warriors is without it. That shows that those decks under-performed a little bit. But I'd also like to leave an asterisk in Chaos Turbo, as it was the only deck with more than 16 players. With 36 players, that means that even if all 16 top spots were from Chaos Turbo player, the deck would still have only a 44,4% conversion rate, which really makes you question the statistical significance of those 50% CR decks. Finally, popularity always plays a big role in tournaments like these. Is something like Chaos Return under-represented? Possibly. Is Panda Burn over-represented? Also possible. Those are questions that are difficult to answer. There's also the question of how we divide decks. You said that you don't consider Goat Control with RotA to be a different deck, but at which point we draw the line? Specially with the amount of overlap between Control and Chaos decks in the format. Split decks too much and you'll eventually find multiple 100% CR decks. Group decks too much and the data becomes meaningless. It's a fine line to balance. You could even go further and see at win rates of individual cards. For instance, how often was Return from a Different Dimension meaningful for Chaos Return in their games? Of course, with the manual nature of DB, that would require manual verification of every replay, and we probably don't have enough data to reach meaningful conclusions at the end of it anyway. I'd also like to reiterate that I have no property to talk about Goat format, nor statistics for that matter. Those are just some points I feel are important to bring up when we're talking about data and statistics, and hopefully this can lead to an even better understanding of the format in the future.
Zaborg might very well be the best card. Certainly among the best. Being able to go into ter to steal or eat a book then zaborg is while. Hitting face down is wild. Sniping chaos mons is wild. Never really a bad card.
It was a shame to see my beloved warriors under-perform. It'll be interesting to see what innovations can be discovered. Perhaps main decked Book of Moon or Asura Priest to be better prepared for the Control match in game 1.
Hey, Goat Duels. I'm newly subscribed to your channel. Do you happen to have any videos that talk about where/how to locate goat tournaments (or maybe a reaource link)? I live in the North East United States and it's rather slim pickings around where I live.
Control still has the highest ceiling in the meta. My theory is that it's better when it doesn't have to beat itself and can concentrate on the other matchup. Cards like airknight and asura priest never did anything against warrior and turbo. They can be cut now that mirror matches aren't the norm.
Hey, I was wondering if you could elaborate on something for me. Is there any reason that Apprentice Magician isn't more popular? It's a dark monster that summons a Magician of Faith (light), Old Vindictive Magician, or itself from deck face-down. While not super relevant you can also use its on-summon effect with Breaker.
Definitely agree Warrior needs to switch it up. I think many decks are getting caught up on only using the best of the best of the card pool. I think players may see some interesting results if they try some less than optimal card choices over what's standard. The gearfried+BWC combo is a pretty good example.
I was surprised by lack of Monarchs. Nice to see more strats as Shaggy was deff the hype duelist in the event. But the top cut was nothing but the highest level of duelists. Good job on everyone! Hopefully other attributes like Water sees the Earth treatment haha
I hope so. I'd love to see a fire or water deck doing well. Don't think Air stands a chance unfortunately lol but at least Earth made a showing in the top cut
@@RarecuisineGaming Water has seen a top or two at events thanks to cards like Abyss Soldier & Nightmare Penguin. Wind would be the next I'd think. Fire/Light are hard to see, but I hope one day some top duelists finds something!
Pretty misleading way to display the data. The way you've calculated these percentages, chaos turbo would have scored below 50% even if it had secured all 16 top slots. You should have calculated a figure using the % representation of the field and the % of top 16 occupied instead. Even by this measure, chaos turbo did indeed underperform but the percentages you've given are all but meaningless.
No, it's not misleading at all. If "turbo" secured all 16 spots, then it would've had the highest conversion rate in the tournament. This data gives you relative odds of making top cut given that you had played a deck. This has practical significance to anyone entering a tournament. The more popular a deck becomes, the harder it is to outperform everyone playing the same deck. Realize that displaying the data as % of top cut vs % of field is the exact same thing, except with a different scaling factor. If you outperform in one, you're guaranteed to outperform in the other.
What's misleading is that you're using tops divided by total players to describe relative performance of decks, which doesn't make any sense to do because some decks' percentages cap out at less than 50% while others can reach 100%. You're correct about the figures meaning something relative to each other, but only insofar as some of them are bigger than others. You cant actually compare the exact % figures. You may as well have just rank ordered them.
Meta may change, evolve and what not but Chaos Turbo is always the most balanced deck out there. With solemn, without, with RB or PWWB, however you play it it's always there. Techs and sides may change but at the end of the day they are destined to always be just temporary
Swap plays are nice with goat, sangan and sinister serpent. WIth spirit monster and recruiter it's already way riskier because of ring of destruction/phoenix wing blast/raigeki break. Against warrior you also risk a simple battle trap if you give recruiter.
Hello. We are a new goat group from Ecuador. Can I use this info for translation as an update on the Goat Format's general state? of course giving credits to the channel as source.
i play a hand control thestalos deck on dueling book and it can be super fun..but against really consistent decks i need to pull really really great or else i dont stand a chance :D
Gillasaurus isnt a bad card for monarchs either, cause when you special summon him. Your opponent usually will summon a monster and you can brain contol for a tribute. And with necrovalley its just a cyber dragon cause necrovalley prevent cards to be moved from grave
I feel like entering a tournament with steingate is a ballsy move lol. You might steal some wins but just do to the laws of probability it would be hard to win 2 duels for every three duels in order to win all your matches in a row
If a deck has good matchups, it has good matchups. Some degree of luck is needed to do well in any tournament with many rounds, regardless of what deck you're playing.
Not sure Goat control gonna be more popular that deck has high skill ceiling and well hate me all you want but goat format is still small and there are many non so skilled players and there are safe options. Oh and plus rep for calling it Standard Chaos not Chaos Turbo because it's just not a turbo deck :D .
It is pretty incredible how diverse the meta is. Chaos Turbo seems like a cop out at this point. Goat truly is a great format.
So, I'm an engineer, not a statistician, nor am I a good Goat player, but I'd like to add my two cents. There's a few points I think should be noted in this analysis.
You did say on multiple occasions that many decks have low sample size, but I'd like to add to that point and say that we could maybe benefit from calculating confidence intervals. Like I said, I'm no statistician, so my calculations are based on a few formulas I found on google, but we have to start somewhere. Using something called a "Binomial proportion confidence interval" and using a confidence level of 80%, I calculated that the best deck in the format is... Chaos Return with a whopping 100% chance to make top cut!
Ok, ok, I'm joking a bit, a sample size of 1 deck is not enough to reach any meaningful statistical conclusion, but it does put in perspective how much more meaningful Chaos Turbo results are in comparison with Warriors (8,7% to 2,9%), and besides Chaos Return, Goat Control got a very impressive 23,9%, while something like Earth Aggro has only 4,7%, lower than Chaos Turbo. Those percentages roughly mean the chance the deck has to make top cut with a 80% confidence. That sort of analysis tends to be a lot better than just looking at conversion rates if you have enough data, which in this case is a little bit of a problem.
This sort of statistical analysis really doesn't work well with low sample size data. It could be better if instead of looking only at worlds we looked at the whole tournament cycle leading to worlds as well, since there would be more data, or if we looked at win rates rather than top percentages. I'd like to note that it's usual in engineering to use 95% confidence level, but I reduced to 80% due to the small sample size. Maybe someone with a better background than me in statistics can give some better tools for dealing with low sample size data.
With that said, I'd also like to add a few extra things. There were 76 players in the tournament, so only 21,1% of players would make top cut. I do believe that's a important point to note, because it's hard to tell how meaningful a 16,7% representation of decks like Chaos Turbo and Warriors is without it. That shows that those decks under-performed a little bit. But I'd also like to leave an asterisk in Chaos Turbo, as it was the only deck with more than 16 players. With 36 players, that means that even if all 16 top spots were from Chaos Turbo player, the deck would still have only a 44,4% conversion rate, which really makes you question the statistical significance of those 50% CR decks.
Finally, popularity always plays a big role in tournaments like these. Is something like Chaos Return under-represented? Possibly. Is Panda Burn over-represented? Also possible. Those are questions that are difficult to answer. There's also the question of how we divide decks. You said that you don't consider Goat Control with RotA to be a different deck, but at which point we draw the line? Specially with the amount of overlap between Control and Chaos decks in the format. Split decks too much and you'll eventually find multiple 100% CR decks. Group decks too much and the data becomes meaningless. It's a fine line to balance.
You could even go further and see at win rates of individual cards. For instance, how often was Return from a Different Dimension meaningful for Chaos Return in their games? Of course, with the manual nature of DB, that would require manual verification of every replay, and we probably don't have enough data to reach meaningful conclusions at the end of it anyway. I'd also like to reiterate that I have no property to talk about Goat format, nor statistics for that matter. Those are just some points I feel are important to bring up when we're talking about data and statistics, and hopefully this can lead to an even better understanding of the format in the future.
i would love to see earth agro/bazoo return decks explored more in the format. very underrated and has alot of toolbox options and good matchups
I see agro decks coming back in full swing. I see sprit reaper making a return . Love seeing Zaborg in decks he’s my boy !
Zaborg might very well be the best card. Certainly among the best. Being able to go into ter to steal or eat a book then zaborg is while. Hitting face down is wild. Sniping chaos mons is wild. Never really a bad card.
If library ftk wasnt banned, how good do you think it would have been?
I love the fact this is still being run. I'd have loved to see a zombies list in there but I understand why it wasn't.
It was a shame to see my beloved warriors under-perform. It'll be interesting to see what innovations can be discovered. Perhaps main decked Book of Moon or Asura Priest to be better prepared for the Control match in game 1.
Love these type of stat videos man thanks for the upload. Incredible stuff
Hey, Goat Duels. I'm newly subscribed to your channel. Do you happen to have any videos that talk about where/how to locate goat tournaments (or maybe a reaource link)? I live in the North East United States and it's rather slim pickings around where I live.
Control still has the highest ceiling in the meta. My theory is that it's better when it doesn't have to beat itself and can concentrate on the other matchup. Cards like airknight and asura priest never did anything against warrior and turbo. They can be cut now that mirror matches aren't the norm.
Hey, I was wondering if you could elaborate on something for me. Is there any reason that Apprentice Magician isn't more popular? It's a dark monster that summons a Magician of Faith (light), Old Vindictive Magician, or itself from deck face-down. While not super relevant you can also use its on-summon effect with Breaker.
Hey. We made a video about Apprentice Magician (and other underplayed monsters) a couple years back: ua-cam.com/video/x4ht_Z3PlXw/v-deo.html
Definitely agree Warrior needs to switch it up. I think many decks are getting caught up on only using the best of the best of the card pool. I think players may see some interesting results if they try some less than optimal card choices over what's standard. The gearfried+BWC combo is a pretty good example.
Would definitely like to see some content around Earth Aggro, it looks interesting
I was surprised by lack of Monarchs. Nice to see more strats as Shaggy was deff the hype duelist in the event. But the top cut was nothing but the highest level of duelists. Good job on everyone! Hopefully other attributes like Water sees the Earth treatment haha
I hope so. I'd love to see a fire or water deck doing well. Don't think Air stands a chance unfortunately lol but at least Earth made a showing in the top cut
@@RarecuisineGaming Water has seen a top or two at events thanks to cards like Abyss Soldier & Nightmare Penguin. Wind would be the next I'd think. Fire/Light are hard to see, but I hope one day some top duelists finds something!
Definitely want to see some earth aggro playtesting. First time I've seen it
Pretty misleading way to display the data. The way you've calculated these percentages, chaos turbo would have scored below 50% even if it had secured all 16 top slots. You should have calculated a figure using the % representation of the field and the % of top 16 occupied instead. Even by this measure, chaos turbo did indeed underperform but the percentages you've given are all but meaningless.
No, it's not misleading at all. If "turbo" secured all 16 spots, then it would've had the highest conversion rate in the tournament. This data gives you relative odds of making top cut given that you had played a deck. This has practical significance to anyone entering a tournament. The more popular a deck becomes, the harder it is to outperform everyone playing the same deck.
Realize that displaying the data as % of top cut vs % of field is the exact same thing, except with a different scaling factor. If you outperform in one, you're guaranteed to outperform in the other.
What's misleading is that you're using tops divided by total players to describe relative performance of decks, which doesn't make any sense to do because some decks' percentages cap out at less than 50% while others can reach 100%. You're correct about the figures meaning something relative to each other, but only insofar as some of them are bigger than others. You cant actually compare the exact % figures. You may as well have just rank ordered them.
@@lsw3364 Dude, it's literally the same statistic that YOU SUGGESTED that I use, just scaled differently.
Meta may change, evolve and what not but Chaos Turbo is always the most balanced deck out there. With solemn, without, with RB or PWWB, however you play it it's always there. Techs and sides may change but at the end of the day they are destined to always be just temporary
Earth aggro looks so powerful and so underdeveloped at the same time lol. It looks perfect for creature swap but im not sure why they dont play it
Swap plays are nice with goat, sangan and sinister serpent. WIth spirit monster and recruiter it's already way riskier because of ring of destruction/phoenix wing blast/raigeki break. Against warrior you also risk a simple battle trap if you give recruiter.
interested in seeing more vids on earth aggro for sure
Stein gate is the machine version with limiter right?
That is a type of Stein Gate deck, but Stein Gate is broader than that.
Hello. We are a new goat group from Ecuador. Can I use this info for translation as an update on the Goat Format's general state? of course giving credits to the channel as source.
Yeah it's cool
i do love monarchs just couldn't find many examples for monarch decks. so far I've been liking necrovalley monarchs
i play a hand control thestalos deck on dueling book and it can be super fun..but against really consistent decks i need to pull really really great or else i dont stand a chance :D
@@illoquenta3940 i took inspirations from soul control and tweaked it to my liking. Also been messing around with mefist the infernal general.
Gillasaurus isnt a bad card for monarchs either, cause when you special summon him. Your opponent usually will summon a monster and you can brain contol for a tribute. And with necrovalley its just a cyber dragon cause necrovalley prevent cards to be moved from grave
We profiled Monarchs in this recent video: ua-cam.com/video/M2bbfes2-rI/v-deo.html
@@GoatDuels i will give this monarch deck a try
I feel like entering a tournament with steingate is a ballsy move lol. You might steal some wins but just do to the laws of probability it would be hard to win 2 duels for every three duels in order to win all your matches in a row
If a deck has good matchups, it has good matchups. Some degree of luck is needed to do well in any tournament with many rounds, regardless of what deck you're playing.
🔥🔥🔥🔥
Not sure Goat control gonna be more popular that deck has high skill ceiling and well hate me all you want but goat format is still small and there are many non so skilled players and there are safe options.
Oh and plus rep for calling it Standard Chaos not Chaos Turbo because it's just not a turbo deck :D .