Did you see the recent Twitch video of the WPN bot that took 2nd in an MTT? I doubt anyone was impressed with its play HU, but when you look at its total winnings you will see how much money these things are capable of taking out of the games. That one bot had 35k in profit. That's one bot.

The fact that some bots seem to play oddly or appear to be exploitable is not that relevant. The truth of the matter is that these things are taking huge sums of money out of the games, regardless of how good you might think they are.



https://www.twitch.tv/videos/407900431

These bots are obviously owned and controlled by WPN. I play a bit of low stakes PLO, and the tables are generally 3-4 eastern European players 1 -2 USA players and myself. It is ridiculous to think that these players are just playing on US facing sites because the games are oh so good. They are playing there because the site is the bots. It is ridiculous to assume anything else.

No, it really isn't.

by what i saw so far , if it is the same bot as the others , it is not possible to exploit, the only thing you can exploit is that they do not see stats or anything like that, so , if you play very nit , they will still 3bet you with the same 9~16% depending on position, they will still 4bet etc ... , and even sometimes , they do some 4bets that they get commited , and they will go all in with A6s, you can only exploit this preflop , and sometimes make overbets and overshove with the nuts .

If a bot is going to fold to min raises too often, which is possibly what I was seeing, do you think that's exploitable?

I don't think the bots are run by WPN...allowed yes but not controlled by WPN. They do play on US sites with US players because the games are much softer. Due to black friday and US legislation players within the US just haven't evolved and advanced as much as the rest of the world.

you gotta try , but , the bots that i was against , at nl25 50 100 200 , they dont fold , and the "best" way to exploit then is to never bluff basically , because they play defending too much .

Last bot I played on WPN folded about 30bb in a row 3-handed FT. DDoS that bot by firing shots left and right.

I play lower, and my best "read" was preflop play. I don't have enough hands to determine all of the other tendencies. I did notice that they would float the flop quite a bit in position. Maybe they are programmed to float flop, and then check back turn, bluff river if checked to again? I picked off a bunch of these bad bluffs in 3 bet pots and was winning decent pots with King high type hands.

that "DuraShura" seems to be a bot , the other one , you wrote his name wrong , but i saw the guy you were talking about , he doesnt seem to be the same bot , could be another one, idk , the only thing i know about float , is that if you check twice , they are very likely to bet with anything , unless , they have a mid hand that cant really extract value and is a safe board , besides that , these bots are soo unpredictable that they can check behind turn straight flush draw , no human would do that ever ...

Ok I was going to type something similar.
Not long ago I log in at 6am EST and I see Alviola sitting alone at 3-4 tables and I played him heads up for a while. The next time I logged in at that time, I don't see Alviola at any table but I see Durashura sitting at 3 tables alone. I sit down at 2 tables and start using the strategy I used vs. Alveola but the play was very different. Durashura rarely folded to 3 bets (he 4 bet more than alviola) and was bluffing less on the river. His bet sizes seemed randomized between 1/3 and 2/3 pot and was checking lots of value hands, giving me some cheap showdowns.

The bot is different but they play at different times. Maybe the same person testing 2 different bots? Also I wrote his name wrong because I'm a science teacher and I was thinking of "alveoli" in the lungs.

If you look on statsname, Alviola’s small 6max sample has very similar PF stats as the default 29/22/9 bot. But Alviola tends to bemore customized. In one hand, it 5 bet AJo 140bb deep vs a cold 4bet + call. Really baffled me. Alviola also plays 4bet pots differently.

I don’t think Alviola is stronger than the default bot though.

Also, Penaple suspected bot.

Penaple joined April 2019 and has much more sophisticated game tree. Will minraise turn and rivers a ton. Not sure with what range.

Sitting with Alviola, I asked him if he was a bot. He did not respond but another player chatted that he wasn't. He definitely is one of the top winners at $10 nl where I play. He doesn't play 24/7 FWIW, but he's from Russia, LOL

DuraShura has superuser like stats. I've got him +48 over 3K. 22/18. We are ragging him now.

I spent quite some time pulling this data by hand, and as I click on the Internet Poker forum to create a new thread I see that near the top is a post on the same topic. I didn't read through the thread which was created in November, and only looked at the OP's initial Database picture (for ants), but it didn't seem all that conclusive. Even if it was, I think my post can still provide value by linking 'new' bots at 25nl.

#All of the numbers and stats that I will reference will be data collected by StatName.net. I don’t assume that their data is perfectly accurate, but I do expect it to be a decent estimation of reality.#

After checking the winrates and stats of the aggressive regs in ACR’s 25nl pool, and cross referencing with my sample on these regs, it brought me to the conclusion that the entire stake is being dominated by what I can only assume is a bot ring.

# https://imgur.com/a/4yTzbdC : complete imgur album that should do a good job of supporting my thesis, without any words. I'll also link individual points of reference using imgur, in case you choose to read along with my analysis.#

There are 8 accounts, none of which are older than 8 months, winning at a combined rate of 5.9bb/100 at 25nl over a sample of 1million hands, roughly 600 buyins or $15,000. Despite crushing this stake, as well as two accounts individually winning 120 and 150 buyins, none of these 8 accounts have played a single poker hand outside of 25nl on ACR. Imagine winning at 9bb/100 over a 170k sample and not playing a single hand at a higher stake.

These 8 accounts play identical aggressive pre flop strategies in 6max, Full Ring, and HU.

# (https://imgur.com/a/n62GEcS) #

Also, general postflop stats(WTSD, WonSD, WWSF, AFQ) are nearly identical as well. All 8 accounts use absurd levels of aggression in 6Max and Full Ring. They all follow the same trend: High Flop AFQ 39%+, higher turn AFQ 42%+, and passive river AFQ of 24-30%. Their HU strat is more normalized.

# (https://imgur.com/a/3h9LTFg) #

Another red flag, is their general strategy when facing CBets across each street. I used 6max as an example.

# (https://imgur.com/a/JRzwDrO) #

Folding to turn CBets only 33% of the time, and that’s the highest of the group?!

It’s the same story for Full Ring and HU too, under folding flops and turns then folding reasonably often on rivers. I have the stats to support it along with a bunch of other relevant and debatably irrelevant data scattered across different sheets and workbooks. But I’m tapping out.

My motivations for this write up originated from an article ACR recently posted mentioning transparency about bots moving forward. I don’t believe for a second that ACR is unaware of these exact accounts. These accounts win at 5.9bb/100, and rake >8bb/100. The massive rake (upwards of $20k over this sample) doesn’t even account for the additional rake they generate for ACR simply by starting games. Fun players don’t open sit, but they’ll happily play short-handed and unknowingly donate to a ‘bot’, that game then fills up and ACR earns their max hourly rake. These ‘bots’ are massively beneficial to ACR and that is why they continue to exist.

I know this problem isn’t exclusive to 25nl either. Shout out to ObeTwice, Gelllemar, and kr0na who have managed to win 580 buyins in half as many hands at 10plo , playing identical preflop and postflop strategies to each other, all the while maintaining the discipline to not play a single hand at a higher stake. # (https://imgur.com/a/ygjFwTV) #

There’s also a group at 50nl, from a similar part of the world as most of the 8 accounts at 25nl and the 3 accounts at 10plo, playing similarly aggressive and strange strats. And still winning, albeit at a more modest rate. “DonKalmar”, “workaholic”, “silentasasin”, and “unfathomable”, all fold to TurnCBets between 30%-34% (6max), have collectively won nearly 10k in less than the 8 months that their accounts have existed. None of these 4 accounts have played a single hand of poker outside of 50nl on ACR. #(https://imgur.com/a/2CHNACS) #

My purpose for this post is simply to deter potential customers from depositing on ACR. I’d like for anyone reading to comment on this for the sake of exposure.

I put in a lot of work on the exact same topic before knowing this thread existed. Bots at 25nl are still thriving. I compare eight 25nl regs' stats, preflop and postflop. They are nearly identical in every spot, and have some incredibly non standard tendencies. Here's my thread I just made on the subject: https://forumserver.twoplustwo.com/s...&postcount=166

Thank you OP for shining a light on this subject .
Idk about etiquette when it comes to shilling your own thread on someone else's, but I don't think you'll mind.

The reason I think he is a bot is because he exhibits some common bot tendencies(stabs hyperaggressively, always play HU, LAG preflop, etc.). But I agree, whoever programmed him is really good at poker.

Didn’t actually type out 25nl bots, only linked:

“Jitterbugg, sonantgiggles, deftswimmer, wickedhag, frontman11, skyjacker, igotnoroof, Nebuchadnezzar”

It is pretty sad to see that nothing has been done about these holdem bots despite conclusive evidence that you have provided here.

https://www.youtube.com/watch?v=-q7Npnz0ZS0

solid evidence fancypighulk! up until ~2 years ago I was a top 100 cash game raker on WPN but stopped playing there because of this. since I've been following closely as this is a really important subject and WPN is pretty much ground zero among regs for where the industry is in relation to bots in the games (lol, solid rep Mr. Nagy). hopefully they see how, year on year, the industry continues to stagnate and shrink. they either need to attack this cancer aggressively or shrivel and die.

thank you bro , you really did a good job with those images , i think its about time and this network will only have bots and few recreational players playing , and then , it will go broke, i mean , i really like the network but... playing only against those bots is unbeatable .