Sunday, March 22, 2009

A little progressive betting perspective here, and some thoughts on why robots can't ski, and gorillas can't putt...

_
In a situation where you have lost several bets in a row or losses outnumber wins so much that you are in a deep hole, you have two recovery options:

Increase your bet values in the hope that going forward, a smaller number of wins will undo the damage done by a greater number of losses, OR
Keep betting the same amount each time, and cross your fingers that soon, the number of wins will exceed the number of losses, resulting in a profit.


Advocates of card counting will tell you that the only winning method is to set bet values according to the high/low ratio of the cards remaining to be dealt in the deck or shoe, and to control the spread to avoid attracting attention.

Counting only applies in blackjack, of course, and even then has to overcome the fact that a "rich" deck is only fractionally better for the player than for the house, if at all.

I have always questioned the message that comes across in pretty much every book about gambling that was ever written and in all mathematical papers on the subject: "Sorry, suckers...you can't win."

The truth is that you can win, and win consistently. But not if you play like a fool.

Over the years I have used several analogies to help illustrate the unscientific idiocy of applying clumsy runaway sims to the nuanced art of gambling well, and two of my favorites are the golfing gorilla and the robot on skis.

The golfing gorilla putts the way he drives, so in spite of the pinpoint accuracy of his long shots, he can never win a game.

The robot on skis sometimes gets to the bottom of the slope in one piece, but more often he ends up in a crumpled heap, defeated by such subtleties as trees and rocks, moguls and other skiers.

The reality of gambling is that you are, in the end, certain to lose more often than you win.

And if you don't bet in such a way that the sum total of all your wins exceeds the sum total of your greater number of losses, you are perforce doomed to lose more money than you win.

What is required is a combination of progressive betting and a keen sense of self-preservation.

Mathematicians argue that "damage control" is a waste of time and effort because negative expectation applies equally to every bet, so running away from a losing trend is as likely to make matters worse as it is to improve them.

They reject my contention that prolonged negative patterns are a relative rarity in games of chance, even though the runaway sims that mythematicians depend upon confirm what is in any case plain common sense.

A runaway sim will show a betting strategy nosediving into ruin after a pattern of losses than no human being would tolerate.

Like the golfing gorilla and the skiing robot, sims blunder into certain failure, unable to shift balance, steer away from trouble, or modulate their "stroke."

What I am hoping readers will take away from this blog is the realization that the technique I advocate can be applied with skill and subtlety, adjusting to changing circumstances and modifying the rules to match resources.

As in golf and skiing, sometimes brute force is appropriate, and sometimes precision saves the day.

There is an alternative, of course. It's called losing.

I have posted an Excel RNG file to make it available for download. LTD+1 RNG.xls uses a simplified set of target betting rules to demonstrate that a clear house advantage can be consistently beaten. Hit the recalc (F9) key several times, and you will see far more frequent overall wins than losses. When the green line dips drastically south of the Zero line, go look for the pattern that did the damage and ask yourself if you would stick around for such a battering if you were making the decisions rather than a robot on skis or a gorilla with a 5 iron! If you don't believe that a move has a better chance of helping than hurting, freeze the RNG values above the killer series, then watch the problem disappear. Life's like that, too: bend over, and you will be badly beaten; take steps to protect yourself and you (probably) won't. The odds in your favor are better than 7 to 1 in this situation.

An important reminder: The only person likely to make money out of this blog is you, Dear Reader. There's nothing to buy, ever, and your soul is safe (from me, at least). Test my ideas and use them or don't. It's up to you.
_

No comments: