Don’t let the title fool you, for I know quite well that in all casino games the house has a probabilistic advantage for any one play of a game. But an interesting mathematical phenomenon, sometimes named Gambler’s Ruin, suggests that precisely 50:50 odds should still make them money.
Let me sum it up in slightly more of a layman’s terms. Assume the house has infinite money (compared to an individual player, this is essentially true). If the player never quits, he’ll eventually get a down swing in his luck and lose all his money to the house. If, on the other hand, the house gets a down swing, then it will give the player some more money, just to take it back when it hits its up swing.
Mathematically, on equal odds, the player with more money has the odds to take the rest. The house has more money, so it takes the money. Believe it or not, you can’t take advantage of that by “knowing when to quit.” For every time you double your money, a fellow just like you has gone bankrupt with nowhere to go but home. And that is the story of gambling: you can’t make yourself win with “money management”1. If you can’t win with everything you’ve got, you shouldn’t be playing at all2.
1Not to be confused with “chip management” in a Poker tournament. That will in fact get you much closer to winning—but you still have to know how to play the hands.
2Unless you have some ulterior motive, which many people do. If you go to Vegas to have a good time, then you’re paying some fraction of your bankroll to have that good time. In that case, you should play with the minimum amount of money possible that will still allow you to have fun.
GLIC has been stagnant for two months, weighing in at 2000 lines of Perl, with at least 4000 lines to go. It’s taken me these two months for the dust to settle, and for me to get hit by the cold, relieving sledgehammer of reality.
Why a new language? There’s already a language that I love to program in, has been subject to years of debugging, has support up the wazoo, and I’m trying to program a new language. Perl—Perl 5—is what I want.
So, change of direction. GLIC is going to be a set of modules for Perl that make explicit the idioms of game programming. There are a lot of low level modules out there: OpenGL, SDL, ODE (that I’ll write soon), but none of them brings everything together to make a game development framework. That is what GLIC is.
And what of my OSCON presentation, expecting a new language? I don’t know, I hope it will go over well. I think I can wow them with the 30 line Pac Man, and then they won’t care that it’s a familiar language.
Oh, and I definitely need a new name. GLIC is becoming so legacy that none of it makes sense anymore. Ideas?
I posted a new paper today about Implied Odds. Give it a read if you’re interested in the mathematics behind no limit poker.
I’ve begun reading Free Culture by Lawrence Lessig. It’s a great book; it has a lot of interesting ideas about social dynamics, and a lot of insight about how ideas spread. It’s not just a collection of essays, it’s definitely going somewhere. It wants me to do something :-). I strongy suggest you give it a read.
I lost the $50 tournament. I came in something like 15th of 22.
I’m really happy I did it. I played really well (a bad beat sent me out: straight beat by a full house). I was bluffing agressively and getting my way most of the time. But better, I had the greatest improvement in my game since learning to fold pre-flop. I’m getting reads. Lots of ‘em. Not the reads that I used to get (“he’s thinking something”), but important ones like “he likes his hand.”
It all comes from a little word of advice Eric Shapiro gave me. Don’t look at your cards until it’s your turn to act. He’d told me this before, but I figured that I wasn’t missing much by taking that split-second to see my cards. And in fact I wasn’t. But I was missing all the reads because I was busy thinking about what I was going to do. Now I can figure whether someone is just limping with crap (or raising with crap), and if so, I can steal it out without even seeing my cards. I did that a few times in the tournament, and was very proud of myself.
We played a little mini-tournament tonight, and I was sent out by ignoring one of my reads, thus proving how powerful they are. Paul limps in with me after a long stare at his chips. This is his tell, and I knew he had something very nice right then. The flop comes, Q high, and I have a pair of 7s. I had gotten so little action, and I figured he didn’t have the queen, so I pushed in. He called without a wink, and showed down not one, but two queens. And I knew it. Even if he didn’t have trips, I knew he had a big pocket pair, better than my 7s. Grr, stupid Luke.
I’m overall very happy with the direction my game is heading. Now I need to get a job to fund the game.
That $50 buy-in poker tournament I was talking about happens tomorrow. First place gets $600 and a shiny case of 1500 poker chips (so I’m told). We start with 650 chips and blinds start at 5/10. Blinds go up every time four people go out of the tournament, but I’m hoping to negotiate a fixed-time round scheme, 30 or 45 minutes.
I’m working on my bluffing, and in order to do that, I’ve got to get respect at the table. So, I went out and spent $30 on a sharp corduroy outfit with a baret that I can use to hide my eyes. If I think I really need to throw them off, I have a chinese bamboo hat (can’t seem to find what these are called) which hides my whole face, and better yet, looks really weird.
My plan is to essentially sit out the first two rounds of the button and figure out what I’m up against. I’m hoping it’s the usual low-stakes assortment: the calling-stations, the pushovers, and the over-agressors. I’ll have to watch for good players, because the way I play them is very different from how I play the other types. It seems that we have plenty of time, so I can play tight and I won’t have to gamble a lot. I hope I can just sit back and go trapping, but that depends on how tight the competition is.
This is much bigger money than what I’ve ever played before, so it should be a good experience anyhow.