If the average hand is 8BB, 100,000BB would be 12,500 hands. If the per hand variance is 5, the per hand SD is sqrt(5)=2.24 . The SD at 12,500 hands would be 8BB x 2.24 x sqrt(12,500) = 2000BB . A 1% edge would be 8BB/100 hands or 1000BB/12,500 hands. So that's like 67% luck and 33% skill at that point.
Radio show interview of poker pro making a living off the bot in Vegas:
pokerfuse.com/poker-podcasts/filthy-limper-poker-radio/1741/#
--- In vpFREE@yahoogroups.com, "kcace1024" <cy4873@...> wrote:
>
> Iguana,
>
> Your rock-paper-scissors analogy is right on target. You can see the bots starting cards after the hand is over, so you would think you could easily get a handle on the bot's playing strategy/personality. At 2+2 someone said the bot has several personalities, but I saw only the LAG or maniac personality in my 3 playing sessions.
>
> With a small variance I was hoping I could determine my edge or lack of an edge within a reasonable timeframe. The variance for limit holdem should be much smaller than no limit and twice the variance of blackjack sounds reasonable.
>
> For example, after 100,000BB of action what should my standard deviation look like. If the house edge is even 1% I expect I will be losing most sessions and won't need to play very long to see that I am not good enough to play this game.
>
> The best VP left is only 99.0%
>
> Chris
>
> --- In vpFREE@yahoogroups.com, "nightoftheiguana2000" <nightoftheiguana2000@> wrote:
> >
> > In a lot of ways, poker is like a game of rock-paper-scissors, except it looks more like this:
> >
> > only playing top hands (TAG/ROCK/SHARK) beats Sir Calls-A-Lot (calling station, fish) (most poker rooms operate at this level)
> >
> > "raise or fold" bully beats the TAG (TAG folds too much to aggression, doesn't like to play with bullies, prefers fish)
> >
> > the maniac (LAG, Stu Ungar, WSOP, bots) plays any hand raising and calling based on the position of the second hand on his swiss watch or whatever (often they don't even check their hole cards, they are "playing their opponent", Stu Ungar liked to check just a bit of the edges like baccarat players do and then guess at his hole cards, of course if you don't know your hole cards can you possibly be giving off reliable tells?) beats the bully (bully gets in over his head then forced to fold something he should have called, all bullies will turn and run to some level of return aggression)
> >
> > the calling station beats the maniac and sanity returns, if only for a short while till the TAG comes back to the table
> >
> >
> > Extra links:
> > Ed's website: notedpokerauthority.com
> > google.com/search?q=squeezing+baccarat+cards
> > google.com/search?q=stu+ungar+movie
> >
> >
> >
> >
> >
> >
> >
> >
> > --- In vpFREE@yahoogroups.com, "kcace1024" <cy4873@> wrote:
> > >
> > > Thanks to everyone for their replies. I have looked at the thread on 2+2, but sometimes the jargon is hard for me to decipher.
> > >
> > > I have only played it a few times, but I did use the strategy suggested by Iguana to call a lot and only raise on very strong hands to limit the pot size. I did well, but I know luck is a big factor over the short term.
> > >
> > > Playing a bot heads up is a completly different game from playing a full table in the poker room. No need to read players and no need for patience because you will play so many hands.
> > >
> > > The button stealing is not a problem. I put several hundred dollars in the machine and played for a long time and you can cash out after you have the button.
> > >
> > > Question for Ed: If you are the small blind and have a negative EV starting hand is there any reason to bet. Shouldn't you just fold every time?
> > >
> > > Thanks again,
> > >
> > > Chris
> > >
> > > --- In vpFREE@yahoogroups.com, "nightoftheiguana2000" <nightoftheiguana2000@> wrote:
> > > >
> > > > LOL. Have you ever tried it? I don't think it can be that good, because if that was the case nobody would play it, at least not for long, the machines would sit empty. Sklansky plays it, but I guess he would qualify as one of those top but somewhat unknown players and maybe he's just got a gig as a shill. The early revisions had some serious flaws that caused a number of casinos to pull the machines, I assume the really bad flaws have been fixed by now. One flaw had to do with its fold frequency when you got all in from being short stacked to start. One advantage it has is that it likes to steal the button on any rebuy or cashout or freeplay or handpay or taxable and so on which a lot of players I've seen don't seem to notice. I guess it's an argument whether or not the button has the advantage in this game, but my guess is that it does.
> > > >
> > > >
> > > >
> > > > --- In vpFREE@yahoogroups.com, Ed Miller <ed.miller@> wrote:
> > > > >
> > > > > I have thoughts about the IGT bot. If you aren't a proven, successful
> > > > > high-stakes heads-up limit hold'em player, I expect you to get slaughtered.
> > > > >
> > > > > Ed
> > > > >
> > > > > On Wed, Mar 27, 2013 at 1:40 PM, nightoftheiguana2000 <
> > > > > nightoftheiguana2000@> wrote:
> > > > >
> > > > > > **
> > > > > >
> > > > > >
> > > > > > The variance of basic blackjack is something like 2, seems like hold'em
> > > > > > should be some small multiple of that, I dunno, maybe 4 or 5, I think that
> > > > > > kind of figure has been floated before. Sklansky's twoplustwo.com might
> > > > > > have some info. Bob should get Sklansky on his radio show. Once you have
> > > > > > the variance, the SD can be approximated as half the average pot times
> > > > > > sqrt(variance x hands). Hold'em edges are usually expressed in terms of
> > > > > > BigBlinds per hour, like 1-2BB/hour. Also, in hold'em, how you play effects
> > > > > > the variance, there are plays that increase the variance and there are
> > > > > > plays that decrease it, like "pot control". The igt bot is mostly a
> > > > > > hyper-LAG and the best response to that is to call a lot but not raise a
> > > > > > lot, that might give you some ideas of how to play. Hyper-LAG itself is a
> > > > > > high variance strategy. Ed Miller, the "noted poker authority", used to be
> > > > > > on this forum, maybe he has some ideas of how to play the igt bot.
> > > > > >
> > > > > > Strategy discussion:
> > > > > > http://www.lvrevealed.com/articles/research/selby.html
> > > > > > http://forumserver.twoplustwo.com/109/heads-up-limit/
> > > > > >
> > > > > >
> > > > > > --- In vpFREE@yahoogroups.com, "kcace1024" <cy4873@> wrote:
> > > > > > >
> > > > > > > The best EV game left at one of my local casinos may be the Head Up
> > > > > > Texas Hold'em game. How many hands will I need to play to have a good
> > > > > > estimate of the house or player edge? I guess the average pot size may be
> > > > > > required to figure the variance?
> > > > > > >
> > > > > > > If I assume no edge (breakeven play) what is the equation to get the
> > > > > > standard deviation for profit or loss after N hands. Can I use average pot
> > > > > > size and treat it like N coin flips or does the variation in pot size
> > > > > > affect the result. If I can play this game at breakeven then I can earn
> > > > > > freeplay and comps.
> > > > > > >
> > > > > > > I found a robot to play Heads Up on the Kongregate web site. Does anyone
> > > > > > know of other robots I can play online for practice? Is there a basic
> > > > > > strategy (like in blackjack) for playing robots heads up?
> > > > > > >
> > > > > > > Thanks,
> > > > > > >
> > > > > > > Chris
> > > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > > [Non-text portions of this message have been removed]
> > > > >
> > > >
> > >
> >
>
| Reply via web post | Reply to sender | Reply to group | Start a New Topic | Messages in this topic (13) |
