gamble

so anyway, if you want to gamble, there's one thing you have to know. no matter what you put your money into, you can't change your luck. (boots will be seething, but really, you can't.) what you can do is put more money in when the odds favour you, and less when they don't. if they never will, as they never do in games like craps or roulette, you can never win, and should not gamble. leave that to the suckers.

what does it mean, to take the best of it? one of two things. sometimes you have an edge but the amount you can bet is limited for some reason. in this case, you want to repeat the bet as often as you can. (a good example is a sitngo poker tourney. say you have an edge in a $10 sng. obv. you can only wager $10 at a time. so to get more money on, you play more than one table at the same time.)

the other case usually occurs in games in which either there are many opportunities to bet, but only some favour you (horseracing is a good example of this: you should not bet every race, because not every race will offer a good bet), or games in which you must bet at every opportunity to stay in, but only some of the time will you have an edge (a good example is blackjack, where most of the time the house has an edge, but sometimes you do, and if you count cards, you can know when that is).


what makes a successful gambler is information. in the first type of good betting, you have to recognise where you have an edge; in other words, you have to know where your skill is sufficient to beat others out of their money. it's no good playing ten tables of $20 sngs if you cannot beat one table of it; nor is it any good doing that if your edge degrades so much over ten tables that you no longer win. of course -- and this is central to all gambling -- it is not always simple to figure out your edge; sometimes, you cannot be sure that you even have one. i probably worry too much about this.

in the second type of good betting, you need to know when you have an edge and when you don't. in blackjack, as i noted, you figure this out by cardcounting. the margins are pretty slim in bj. if you learn basic strategy -- and i'm almost there with it -- you can cut the house's edge over you to .5%, which is very close to even. That's losing $5 in $1000. But the game changes over time, sometimes against you more than that, sometimes in your favour. you have to make up the bad times by whacking the money on when you're favoured.

of course, you do not win all time you're favoured. this is gambling, after all. but you win in the long term.

the long term is like a magic realm for gamblers. it's when everything works itself out. you cannot fix your luck. some days it goes your ways; some days it doesn't. it's tough to learn that, and tougher still to trust that you will win out if you stick with doing the right thing.

where things get interesting is horseracing. here, it's obvious that information is all important. you know you cannot beat the horses unless you understand them. but it's not so obvious that you must take the best of it, just the same as in bj or poker.

i'll explain, but first i'll digress.

when i was in my early twenties, i lived at home. i was totally unemployed and pretty much unemployable. but i was happy. i spent nearly every day, all day, in the bookies. i knew horses inside and out. i would double my dole money every week by betting the horses. i wasn't betting optimally, but i had a huge information advantage over most bettors. i was watching every race every day. i knew how that 2yo maiden had run at Yarmouth, because i'd seen it run.

but here's a thing. it's not how many winners you pick. look at this. say you check out one hundred races, and pick out 100 horses that are in your view evenmoney shots. for the purposes of this discussion, let's say that your information is good, you bet $1 a race, you are correct about the horses' chances, and also that you pay no tax. let's say your horses go off at 4/5. so you pick 50 winners and make $90. iow, you lose $10.

so you look at another 100 races and pick out horses that you think are 9/1. You put your $1 and each goes off at 10/1 (the astute reader will realise that these can be averages, and most handicappers will have some evenmoney shots go off at 5/4 and some at 4/6 and so on, although in fact we are looking to always beat the line we ourselves set). So 11 win, and you make $110.

see how you win more -- in fact, you win at all -- correctly backing the longer-priced horses than you would by backing more winners but at the wrong prices? this is horse handicapping in a nutshell. when you have sufficient information to handicap horses correctly, you can pick good bets at the bookies. it's not about which horse is most likely to win any given race but whether the set of horses you are betting offer better odds than they ought to. it's the long term. a horse that is 100 to 1 against winning but is offered at 200 to 1 is hugely profitable.

yeah, it sucks that you must risk 99 losers to get that profitable winner, but here's the thing: gamblers do not mind losing so long as their bet had a positive expectation. it's something i have to learn in poker. i can lose a lot but so long as i am betting with the best of it, it doesn't matter. give it time and the money will return with interest.

so i aim to play these three games. i play sngs okay, beating the $10 level, and i could expect to win a bit higher. i don't know how high. i am nearly there with basic strategy, and i am willing to put in the many hours of drilling i will need to count cards. it's going well though, and i think i can do that. and i plan to learn the British horses again.

in the last of those, i know that perfect information is impossible (although some elements of perfect info are very possible, such as knowing that a trainer will pull a horse, or knowing that a greyhound has a virus and will not run to his or her best), and even imperfect information is quite hard to come by, so i'll be looking for wide spreads. but if you correctly analyse 10/1 chances and they pay 20/1, you gain a ton by betting them. the same is true of evenmoney shots at 2/1, which is findable in markets in which information is fairly limited but most people in the market either lack it or can't be bothered learning it. maiden auctions come to mind.

you cannot change your luck. sometimes your horse falls at the first. you double down and hit an ace or deuce and dealer hits 17. you get it in with aces and the other guy flops his set. but everyone has the same chance of luck, if not the same luck. and you can ride the waves of luck, making sure that when you hit a peak, you have as much money as you can in play, and when you hit trough, you have as little as you can in play. learning how is not easy, but the difficulty lies not in knowing you have to surf, but learning how to recognise where you are on the wave.

well, it's no a science. i'm not relying on it, or risking what i don't have. but ir a year from now...

posted by Dr Zen @ 10:01 pm