Low Limit Expectation

[Vulpine]

When I run turbo Texas Hold’em for 5 million iterations for a 4-8 structured limit using the default toke and rake and using the default palyers, everybody comes out a loser. I didn’t set a jackpot rake because I assumed that in the long run you’ll get back everything that you put into the jackpot which may or may not be true. Of course, when I set the jackpot rake to 1 dollar per hand and assume that you’ll never get any of it back the results are even worse. If I set the simulation so that there’s no toke and no jackpot rake, then the best player makes about 75 cents per hour assuming 30 hands per hour. Poki seem to be the best AI based software player and it can average about .1 small bets per hand when there’s no rake and no toke. If you do the math for the rake and toke with 4-8 you'll find that Poki is not going to have a high standard of living. Although Turbo uses a formula based approach instead trying to estimate expectation Poki uses a more sophisticated mathematical approach and Poki can learn from other players using Bayesian statistics, neural networks, and genetic programming. 10-20 does seem to be a winnable game with both Turbo and Poki but are lower limits like 4-8 fundamentally a negative expectation game unless played online with a lower rake and no toke?

[DesertCat]

I've run similar scenarios with TTH and got similar results. But it doesn't necessarily mean low limit B&M games aren't impossible to beat, the loose B&M players are supposedly much looser than TTH's "loose" bots are. But the results were discouraging enough that I've decided to stick to 6/12 as my minimum at B&M's since the rake/tip costs are a much lower percentage of your wins.

[Ed Miller]

Real, live $4-$8 B&M games are easily beatable. TTH simulations do not accurately reflect the huge edge that good players can get over these games.

[mauisupaman]

Listen to Majorkong (well, Ed Miller now). I used to think live LL games weren't beatable when I first started. I'd try and blame the high rake and what not, but really I just wasn't good enough at first. I used the advice of Majorkong, Dynasty, Bob T, Clarkmeister, and the rest to built my bankroll up to play in live 8-16, then 10-20, and now I'm starting 20-40.

I actually started with $50 as my initial bankroll and learned playing with friends, got lucky, and never looked back.

But in any case, yes, all those simulations and what not are great, but the games are still beatable as long as you have good table selection and play well. I'm admittingly not even that great of a poker player, but I stay on the plus side. [img]/images/graemlins/cool.gif[/img]
A hui hou,
Adam

[Bob T.]

The difference between the bad players, and the good players in the simulation, isn't as much as it is in real life. The games are usually beatable, and for multiple bets per hour.

[BarkingMad]

I ran the same simulation when I first bought TTH, and based on results similar to yours I quit playing B&M, thinking the 4-8 was virtually unbeatable (My game is not ready for 10-20 or the no jackpot 15-30). I’m in Seattle, and the local casinos I started playing at rake $1.00 for every $10.00 in the pot to a max of $4.00. Since I won’t play 5 million hands in my life, I factored the jackpot drop as loss, as well as a conservative tipping estimate of $1.00 per pot won.

At 30 hands per hour, w/ a conservative average pot estimate of $30.00, I figured my local B&M (Tulalip, Diamond Lil’s) game extracted at least $150.00 per hour from the table. My share of that at a full table would be $15.00 per hour. So for me to just break even I would have to make almost 2 BB/hr, and the remaining players would have each have to lose $16.66 /hr (their $15.00 share of $150.00, plus my share divided amongst them). For me to profit 1BB /hr the remaining 9 players would have to lose $17.55 / hr. The math gets even more lopsided if you figure to have one or more other players at the table who play well who you'll have to share profits with.

Someone please correct me If my assumptions or calculations are in error.

At Party Poker the rake is about 5%, half of B&M’s 10%, plus no jackpot drop and no tipping. If you do the math it’s a pretty significant drop in the cost of playing.

SSHE is my current poker bible, so if Ed Miller says the B&M 4-8 is easily beatable if you play well, I’ll take it on faith. But to me, the real question is, is it looser and more beatable to the extent that it makes up for the increased playing cost, when compared to online play. I honestly don’t know the answer, and I’d be interested to hear from someone who does.

I know part of the expanation could be that online players are better than B&M players. This is an idea I'm still not convinced of though, in fact, in my experience (here in Seattle cardrooms) the opposite is true. I continually see Party 1-2 & 2-4 players turn over some crazy holdings that they cold called with, 9 5 offsuit etc. I haven't seen such craziness in my local B&M. I hear California is different though.

[Blarg]

California definitely can be different. The loose games that Ed describes are often essentially the California casino poker that I've seen.

It can be somewhat table and level dependent though, but you rarely have a table where there aren't at least a couple of people playing poorly, or at least much worse than the rest of the table.

[Ed Miller]

But to me, the real question is, is it looser and more beatable to the extent that it makes up for the increased playing cost, when compared to online play.

No. If you play well, an online $3-$6 will be far more profitable than a B&M $3-$6 or $4-$8 game. But rest assured, those B&M low-limit games are definitely beatable.