dude literally yday at 2 am on gg 5/10$ theres 18 tbls running with one casual at each table hows that not good lol.

ACR did this to some tiny extent (quite unintentionally I'm sure) by making ~10% of the pots bomb pots (5bb ante from all players).

I wonder how many of those 18 tables would be running two minutes after the casual left?

more than you might believe because of leader board point system. maybe see the ggpoker rake before jumping to 'professional are scumbags'

so people should be forced to play in games they have no edge in. Personally never bumhunted or ran from a table in my life but if im at a table with a couple players who i can consider as good as me or better and a couple marks. I dont think it would make a jerk to leave if the 2 donkeys left. Or am i obligated to stick around and bleed money ?

why else would they play poker if not to make money lol, theres no money when the house takes rake in regular games.

the amount of accounts which were playing up to 100 hands here and there of micro stakes over a few months at start of the year and now suddenly playing 1knl/2knl on stars is getting a lil sus.

TBH I assumed some of the 3 handed reg table lineups ive seen lately / random names at highstakes was more a result of people up a lot of money from crypto going for a punt , maybe a mixture of both

are they winning?

So just choose the 2nd or 3rd best choice x% of the time. Easy game.

Why do sites not just offer bomb pot only tables?

Most chess cheaters are pretty dumb (even very strong player).
My experience is that even GMs don't know how to use a chess engine besides just clicking "go" and looking at the first choice printed. Many probably don't know the engines can be run in multiPV mode (so they show top n lines, for example 6) and for some engines this doesn't even weaken them (MCTS engines like Leela).

I am sure a lot of things can be done to detect either:

1)Very strong but not very bright players cheating
2)Very weak players cheating

It's an uphill battle against smart determined cheaters though and poker sites are facing a lot of those.

This actually happens to some extent. I have played decent volume through the years vs the more known cheaters and usually they have some small suboptimal strategies.

I.e OBORRA played a lot lower cbet freq than was/is considered GTO, he was using a lot of 50% cbets on boards where it was both a trend and clear from population that it was better to play small cbets. I would almost argue that as far as sizing goes, the bots almost had worse sizings from a theoretical stand point. Which obviously doesn't matter at all as long as you play the strategy correctly.

And i have def seen bots/cheaters just ignore the solutions and go with their own judgement plenty of times. The combination of this makes it really hard to catch cheaters, but some sites do a very good job despite this.

Nice to see people in the thread who know what they are talking about speak to the difficulty of catching people who use this technology and are clever about it.

To address those talking about the games still being good and who cares. If someone posted here that they were playing in a live game where they knew players were cheating but they thought they still had an edge so they keep playing I think most people would call them idiots.

Personally, I don't want to compete in games where I have no idea if players I am against are cheating or not and that is the current state of online poker. I play as a recreational these days and that isn't an appealing or fun thing to think about doing with my time.

Which of course is exactly why collusion works. The dummy who makes a play that hurts you and themselves while helping the third guy, is no longer a dummy.

There is though a simple way to stop almost all of this. Allow anyone who requests it to to see everyone's hands any time after the hand is over. Not just the site. That has obvious downsides but it is really the only complete solution. So it may come to that.

Or perhaps make it so every folded hand is exposed at the moment it folds. That adds the downside of slowing down the game. But it has the upside of thwarting computers unless they are even faster than I thought.

In any case I like the idea of changing the game to one where there are many more multi way pots. But not so much because of bots but more so because games where folding is less often the correct option would be more fun for recreational players.

These solutions have downsides.

Sites need to pay for more customer support dealing with false reports, and a bit of the magic of poker is gone (I imagine recs will leave faster with revealed holecards post-hand etc).

Also this thing about regs not moving from NL to PLO... the reason regs play is recs. All new poker players start with NL, and basically vast majority from new to crusher have stronger preference for NL to PLO.

It's hard to move players to games with antes/more multiplayer pots games as a site with that product will be outcompeted by one with just plain old NL.

Pre 2017 stuff is very different to 2021. Landscape is worse now in the sense that cheating is more prevalent- you will more likely be facing bots (due to lower barrier to entry), but they also get caught more, and faster. The trend is not good for poker in reg on reg action (and it trickles down) obviously.

Btw identifying bots/cheaters deviating on purpose is inf harder than identifying bots/cheaters. As a player you only see showdowns and there are potentially many other reasons why they made such and such plays.

I have seen bots deviating on purpose but only because we managed to get a working copy and sometimes instruction manuals on how to avoid detection. No way I could tell as a player it wasn't a misclick/malfunction/other.

I kinda agree with you and it's partly why I got a job instead of playing. It takes a while to get caught and at highstakes many ppl don't intend on playing poker for long. But at lower stakes it's not as bad- still takes time for them to get caught obv.

It is what it is, meh.

@Tuti88

What do you exactly mean by "fraudelent mouse movement"??
As long as you are not moving your mouse like a bot, how can it be fraudelent?

I myself am tanking in a lot of spots while not moving my mouse.... so am i a cheater now because of your logic?
Or what exactly do you mean by saying it can be detected by looking at mouse movement.
There could be wrong detections i guess...

I kind of like the thought of showing all hole cards at the end of the hand. It will crush players with weak mental games. It takes away from the game but adds a little to it as well. The problem is going to the information the trackers get and how players will be able to use that with solvers and node locking.

Change poker format weekly. Add extra blinds, antes, straddles. Change blind sizes. Change max, min buy ins weekly.

I know this would be bad for regulars too buy it would help recreationals and will make cheaters work harder.

99% sure it was intentional to mess with bots/RTA users.

This shoudn't be hard.

Neural Networks on time to move, blunders made per hand, mouse movement in the screen, how the player interact with the software should be a good set of variables to make an easy tool to know if a player is a bot, human or some cheater using rta.

And how do you get the training data for this? There aren't tens of thousands of known rta setups that you can use in your dataset

Changing game rules is also not a great solution in theory, most formats can be done pretty quickly (just not with database of solutions approach). Punters multiway postflop game idea might work. There has been some new work on a game where 1 human vs rest ai beat the human, but 1 ai vs humans performed poorly. Pluribus might have just done well in 6max no limit because of how quickly most spots get headsup.

Currently talking about this on the S4Y podcast. Starts about 20 minutes into the podcast.

Poker's Cheating Problem-- The RISE of Real-Time Assistance w/KL Cleeton