Very interesting video. There will always be people who sit games drunk, who don’t study, who just want to play for fun or who regularly tilt. Low stakes poker is still soft as shit and probably always will be. It doesn’t matter what the latest software shows us or what we learn about optimal strategy. Always remember 95% of people who play poker do not beat rake. For a lot of people it’s just a hobby or a bit of fun. There are still guys who make $100,000 a year playing push/fold SNG satellites. How is this even possible in 2020? It’s possible because there are still tonnes of fish who just play for fun
Playing against another player heads up or same players at a short-handed table isn't enough. The bot has to play in full games with players coming and going and win consistently to fully convince me.
The best years of poker were from 2003 to about 2010 That’s when simply being a rake back “NIT” could make a six figures sitting on his couch. 😍 Those were the days...
Pluribus lost something around 70k to the humans. They use AIVAT to claim without variance, the humans would've lost. Not here to debate or speculate either way, but Pluribus actually lost to the humans. It seems very odd that they claim victorious, even know they lost. Very similar attitude when Claudico lost vs the humans, and the Claudico developers declared a statistical tie. The humans defeated Claudico at 9bb/100. Libratus did in fact defeat the humans, in the 2nd brains vs AI challenge.
i like your comment man ..where? can i find infos about AI playing 6 max .. .. i also havent seen prove thet AI is superior yet ..except libratus in no limit holdem heads up im playin pot limit omaha & i ve a good win rate .. im studying solvers but their game is not applicable in human invoroment & sometimes im thinkin that none of this softaware is tested like libratus .. still ..there is a lot of value in studyin w solvers , cause u r forced thinkin about the game tree whille studyin ..
@@RodneyjSalinas i can t find articles that confirm it .. but i still believe u ..cause if ai was a winner they d advertising for ever .. i think they need to sell it thats why nobody talks from this point of view
since the same party in a match only gets the winning hand a small amount of time winning by mathematics on it self is not possible,since essential information like the other hands are not known.
So when the pros win by 700,000 plus after 80k hands, that's statistically insignificant?? The computer was down over 36 buy ins...seems like a pretty clear win by poker standards. When the computer prevails by 1.7mil after 120000 hands though, somehow that is deemed a clear victory? Seems legit lol
About 2.5x winnings in 1.5x hands. might make the result statistically significant. Statistical significance just means the result is less than 5% explained by chance alone. So if the player win is 7% explained by chance it's not statistically significant. If the AI win is 4% explained by chance it is statistically significant. People overthink what statistical significance means. And I'm oversimplifying it ofc but you get the idea.
nice ..video ..first of all the competitions r very few .. im also curious how linous perform acainst AI .. last & more important is that AI it self can not threat - cheat humans .. humans can use AI for cheatin & humans use many things f cheatin ... AI can help humans t improve .
