I wrote a bot in 2002 that won, and at a pretty fair rate. If you remember TheCount, he developed poki bot and that started the bot wars. Poki bot was definitely better than 95% of lhe players at the time.

RGP poster Abdul Jalib, aka Andy Prock is probably the most notorious and talented. He wrote poker stove. Now that I think of it, both these guys went to MIT at some point I think, might be CalTech, but same difference.

My bot was much more primitive, but it doesn't take much skill to win at micro stakes, only a ton of a patience...a computer has that. It's biggest flaw was running out of memory, sloppy VB coding did that.

And no, I didn't profit from it as I made much more in HS games. I would usually just blow off steam by having it shove instantly all in for a while in the low stakes games to piss people off. Then I got told to stop, so it ended there. It won easily when I let it play for real.

Playing HS you have a small player pool and that helps when you actually know who it is you are playing more than anything else. You can tell quickly when it is not the same person, and figuring out new accounts of people is usually pretty easy. It's only a good gimmick for a couple hundred hands at most.

Bots aren't the issue, if a bot is so good that it can beat you, then you don't have to play it. A lot of sites use bots to seed games, UB, Absolute, Paradise, PokerRoom, 24h, and Party did at one time. Party used to restrict them to the beginners games and they were impossible to lose to, but you had to have a new account, the other sites used them as props in the low limit games, and they were easy to beat...

But now the bots are really good, iPoker's lower limits, especially for limit games are entirely bots. I can't honestly beat them, and I am pretty damned accomplished at lhe. I tried playing all the bots at .5-1 for fun one day and I got massacred.

Finally, in Romania there are a couple firms that specialize in botting and bonus whoring. They pay people to stick to a chart and/or mind the bots. The services of these companies have been retained by real money sites. I know this incidentally from those who have started sites, I was never involved.

The bots winning now are purely doing it by way of volume and weak competition. Any competent players can destroy a bot program in the long run. Good luck with programming a bot that can beat a competent HU NL player because it hasn't been done nor will it any time soon. The only truly big edge they have is the collusion aspect of programs sharing information, but even with that they are exploitable.

You are speaking out of your arse...

Bots are playing weak competition at lower stakes because that is the most profitable way to do it. It doesn't mean bots can't beat the best players at HUNL.

I've agreed with everything you've said except that. If bots could beat the best players at HUNL they could easily take several million out of the high stakes HUNL economy in a couple months.

I know it's theoretically possible to write a bot that could beat the best players at HUNL and it might not be long before it happens, but it hasn't happened yet and the reason isn't a lack of financial motivation.

isn't there a good chance that the bot gets caught and funds confiscated if someone was to release a bot at 50/100 or w/e? there more chance of a hsnl reg figuring out who they are playing than say a 1/2 or 2/4 reg, right?

also how do we know that there isn't bots beating hu games right now? these programmers who write the bots are obviously v tech savvy, whats to say that they have figured a way of going under the rader on sites such as pokerstars?

also there is a ton of variance in hunl cash, the guys who own these bots might not want to swing 6 figures on a regular basis. (assuming there is a bot out there that could beat hsnl hu)

I'm surprised it has not happened faster (not that I'm for it at all) Look at the % of daily volume these days that quants and HFT funds do in financial markets.

http://www.mitpokerbots.com/

The 6.912 MIT Pokerbots Competition is a computerized poker tournament, where teams have 1 month to program a completely autonomous "pokerbot" to compete against other teams. Poker has become a cultural phenomenon as as an inherently complex game of incomplete information, brings many interesting problems and challenges to players. Competitors must learn and apply concepts in economics, mathematics, and computer science not normally developed together in academic settings to conquer their opponents and emerge victorious.

Over $30,000 in prizes and attention from some of the most prestigious tech and trading firms out there await the best competitors. This IAP, channel your programming ability and strategic quantitative thinking skills to become the inaugural pokerbots champion!

Here is how PTR detects bots ...

Bot detection is a 4-step process:
1. PTR collects hand histories from the major poker sites (over 8 billion and counting). The hands are stored in our massive database cluster.
2. We use a mathematical grouping algorithm based upon select stats that are highly indicative of a particular play style. This process requires intensive processing in order to compare large numbers of players to one another over long periods of time. Each player is analyzed and placed in a group. During the process, very few groups grow beyond a single player. Some groups, however, contain many players which is quite abnormal even when analyzing huge numbers of individuals. We call these "suspect groups".
3. Each suspect group is analyzed based upon a wider range of stats than originally used to group them to verify that the suspect play is consistent over all statistics we record. Any players whose play is shown to be different by this analysis are removed from the suspect group.
4. We observe individual hand histories of these accounts in order to find abnormal tendencies that are consistent across the group of accounts. This provides additional proof that the suspect accounts are being operated by the same computer program.

Nanonoko has a bot score of 76 on ptr. He clearly not a bot so the PTR bot score is not worth much.

That is 76 out of 100.

A player rated 0 has entirely unpredictable play habits and a player rated 100 would have extremely consistent, almost impossibly long sessions with extremely low variance in length or schedule.

Bot score is not the same thing than their bot detection. Bot score just tells players' consistensy and lenght of sessions, it doesn't really have anything to do with detecting bots.

Yes.

Similar terms. Different processes.

Interestingly, the described "Bot detection" method implies that if a bot is not duplicated or used twice, it will remain a "single player group" and thereby avoid detection.

Bots that are sold, shared, or copied increase their chances of detection.

Bots will and have always been there if your coming from a video game backround you would know how hard it is to fight them. All you can do is play on Stars, hope for future regulation and much longer account reviews when you want to cash out. I dont care if i have to wait 1 week or longer as long as more botters get caught.

Botting itself can never be stopped but cashing out the cheated money can be stopped.

That's funny because the bots PTR detected on pokerstars had a bot score around 60.

Nanonoko is more of a bot than the bots!

PTR never detected any bots on pokerstars. It was PS itself or 2p2

Bot Ring Discovered On Poker Stars

It was the info on their site that helped the original poster and they provided additional information.

Yea it was the info on the site not PTR itself. stars didnt care at all about the info PTR provided they have much better methods themselves. Some Bots get banned on stars i would guess that happens every day, but no big PR for ptr lol.
Too bad i cant find the post from PSsteve atm but PTR had no influence at all there they just like to let people think they had.

I'm bored and reading the thread now in between stuff... First investigation from PS cleared the suspected players. I'm guess that something else changed since then.

I know everyone hates PTR but I think there are some good aspects and it looks like Stars didn't act until after PTR did it's detailed analysis. Not saying PTR's analysis influenced Stars decision but at the very least it kept players motivated to keep pressure on Stars to do something.

I don't think that this is correct.

Timeline is here: http://forumserver.twoplustwo.com/sh...&postcount=997

gg online poker

Minor fact correction:

Nope. Andrew Prock is not Abdul Jalib. Back to the discussion.

Regards, Lee

I hate to tell you, but if it's the bots that are like three to a table, those are ****ing terrible. I'm not going to list their leaks publicly, but it's basically like the worst tagfish in those games decided to make 10 copies of himself. They win slightly at .25/.5 because those games are lol. They get crushed at even .5/1.

yeah baby plo time. there is no human on earth who can create a bot that is profitable at plo200++