Sunday, December 16, 2007

To Bot or Not to Bot, That is the Question

One of the more interesting forum-discussion threads, for those who enjoy reading about the cat-and-mouse game regarding poker-bot detection by sites, has to be a thread entitled " Fulltilt froze my account with 47 grand in it," which is now nearing 100 pages of posts.

The Cliff Notes version is that a player on Full Tilt named 'Sillysal' had his or her account frozen, some $47,000 worth, due to the account either being a bot or being affiliated with botting. While Sillysal adamantly maintained that this account was not involved in botting, other evidence came to light that suggested something bot-related was occurring, and as such, Full Tilt did indeed confiscate the funds after a lengthy investigation.

Where Sillysal has been less than forthcoming, however, has been in the exact nature of his/her relationship to other accounts that were also banned for bot activity, accounts generally playing at lower levels. All of these accounts were playing heads-up limit poker, by the way, which would by its nature be one of the most 'bot'-able forms of the game.

When this thread first came out I sent a private PM to Michael Josem, who is one of many recognizable names who has posted in the thread. Josem, for those who don't know, played a key role in uncovering the Absolute Poker scandal. After reading the majority of the thread, I wondered as to this:

What if Sillysal was the owner/operator of a small network of bots playing heads-up at lower levels? In this scenario, Sillysal would be perfecting his/her game by playing manually at the higher levels, then using the knowledge gained, slowly honing his/her playing skills, and improving the bot's software patterns over time. While Sillysal has been shown to be a mild loser at the game on that account, one would guess that the same skill level would be quite profitable at the lower levels, where the bots that were detected may have operated. They profit there, and Sillysal would use the funds to improve his/her own game, which in turn improves the bots... the cycle keeps on running.

One hand would feed the other, in this scenario, with both the bots and the human master improving over time. While it's almost a certainty that the bots would be running from separate ISP addresses, it's also likely that some other link exists, and that's the part of the story that has not, to date, been publicly released. If the Sillysal account is indeed clean, as its owner asserts, then what is the evidence Full Tilt possesses? Cash transfers between Sillysal and the other bot accounts? In-common bank or address records?

[Edit: In reviewing Michael Josem's comment and a segment of the thread itself, a lot of the indirect references in the thread released via 2p2 moderator Mike Haven point at 'timing' indicators as having tipped off FTP to the existence of the bots. Could be... and the connection to another, similar account named 'pokergirl z' is also damning, but it doesn't prove or disprove this scenario here. And the one question never answered in the thread, as far as I can see, is why would a bot account continue to play only at stakes where it was a proven loser, if other players' stats on sillysal are correct? That makes no sense. There's no reason to play a bot at the highest stakes possible and thereby lose money, when there's money to be made elsewhere. Even with rakeback considerations included, it just does not add up.]

It has to be something. Nothing in 98 pages of posts has dissuaded me from my original beliefs, though I check back every now and then to read the latest. What are your thoughts on the issue? I have to admit, I'm curious how this one will end....


Anonymous said...

While your theory is, of course, possible, there seems to be little evidence that supports it.

guest said...

First, Hi, Michael!

Second, I haven't seen anything that would disprove it, either; that's the whole point of it being a theory. Unless I've missed something in there of late? Lord knows there's enough garbage in that thread.

That's the fun part about idle speculation and conjecture. It's just a "what if" -- an idea being stretched to see where it might lead.

And of course, privacy concerns aside, if the evidence in the case were publicly released, then everyone would know exactly what is was. Not likely to happen, though, not that we can't have a 'bot' of fun with the tale.