These 3 were confirmed bots. Same stats. Theres dozen more of them on that site and they keep recreating accounts with a couple tweaks to hide. They crush all stakes on all sites. HU bots are even better as they are disguised as fish.
Even bumhunters don't get those winrates unless they 1-2 table.

Still - why only three then ? Presumably the programmers would wanna run their bots on as many entries as possible (probably with slight tweaks each entry, to delay detection) ...?!

You seem to be knowledgeable about the issue ... any estimate how their (or "its", as it seems to be just one bot then ?) strength might compare to Claudico ?

The online bots are analyzing millions of hands played by other bots, themselves and online players. They do not do anything spastic and they do not balance the same way claudico does, its a lot more similar to humans with good blockers.They have probably seen so many hands that they know how to replicate human bluffing tendencies very well. They are adapting to every situation. As much as they are aggressive, when you trap they play passive. They also have betsizing/timing tells on every player who puts any kind of decent volume.

Thought this several months ago when people were championing for Ike to be a guest on JRE. Doug's interpersonal skills would have made for a much more engaging interview. The 2p2 community has been very forgiving about how dull the interview with Ike was - probably sealed the deal that Rogan won't invite him back and probably won't bring on any other pro poker players.

I think the view count on his interview contradicts your statement. Ike was on ep 526 and that has more ustream views than the next 11 podcasts.

That is godmode.

how would anyone know that it knows that?

I have seeen a fold from Dong Kim on river (AQ on AQT J x fd) vs shove where he thinked about what Bot should have there but Brains please consider this:

thinking about what bot can have on a river spot vs shove can mislead you because the bot has to have balanced (gto like) ranges as engineer told us (even if it's ranges are far from optimal) so the only thing you have to consider that your card on that board is in the percentage of your possible cards compared the pot odds the bot offering.

I'm sure these players need proper coaching on equity and pot odds.

Dong is in a tough spot on every hand lol

+1,good questions.

Another good question for the podcast.

I apologize if this was already covered, but anybody know if Claudico adjusts his strat after each day depending on opponent? Or does it analyze the data from all 4 players and makes 1 same adjustment to use vs all 4 guys?

They are done:
Dong +40k
Doug -50k
Bjorn -95k
Jason -20k

something like that for the session I think

FWIW, Claudico had AA this hand. In the reverse match bjorn 4bet pre, too, checked back flop, kinda hated turn but bet iirc, checked back river (whereas claudico shoved). Another hand Dong tank/folded river with QJ on J2838, Claudico had 6c7c. Bjorn 3bet pre, chk/raised flop, bet turn shoved river that hand.

Bot with the 67 vs QJ was nasty. Bot 3bet , bet call, check check, check bet overshove jam fold.

I am not saying that fold was a bad or a good one (I don't know Dong Kim ranges against the bot on this exact spot, and probably I am not such a good player like the Brains are)

I am just saying the thinking method ("What should my opponent had") can mislead a player against a balanced range where and when if the bot shoves it means it has to have least as much bluff percentage as the pot odds what human player has.

I have seen Cheet is reading this topic (I liked your game Cheet, congrats! Anyways are you planning changes on your pre ranges based what you have seen from Claudico till now?), and I hope least he will consider this basic concept because I am afraid Claudico will crush the guys on river soon if they don't. (Claudico will know from HHs that the Brains fold or call suboptimally and it will win more on that spots, for example it keeps pots low more times and overbets on river a lot)

I still think Brains will win though because they use way better preflop ranges and flop strategy. Go Brains go!

anyone know if all this content will be loaded to youtube? assuming its legal and twitch doesn't hoard its content.

Would actually be so cool if all of this will be brought out by the organizers as .avi files ... just in case you CMU guys are reading this

You can find all sessions in the "Past Broadcasts" section on the profile pages of each stream. Even the ongoing session can be viewed/rewinded there.

Example: http://www.twitch.tv/claudico_vs_jas...ast_broadcasts

For downloading .flv files: https://twitchtools.com/video-download

Any programmers know how difficult/possible this ability would be to add to a bot?

I am aware of at least one publicly available timing tell software (which I have never used, so I can't say how good or not good it is), but the bot programmers can certainly get inspiration from that one, I'd say.

Wasp... in all honesty do you really think these players haven't been considering what you just wrote?