Monday, May 4, 2009

Experimentation with past outcomes is a fruitless waste of time. But observation is the key to long-term success.

_
In the 30 years since I first figured out that a 2-win progression should be less risky and more effective than a 1-win Martingale, not much about the target betting concept has changed.

After my Eureka moment on the dusty road out of Las Vegas in 1978, I had to wait quite a while for personal computers to catch up with my need for a speedy but reliable way to evaluate my idea.

I posted a version of the strategy on the Internet in 1997.

Since then, I have simplified it while protecting the core concept that a mid-recovery win should always trigger an attempt to recover the entire loss to date (LTD) for the current series of bets.

A wide betting spread (sometimes as wide as the casino will permit) is the key to eliminating the threat of the house advantage, which exists in all table games and ranges from 1% or less for blackjack to 5.26% for American roulette.

Table limits do not prohibit target betting, although they do slow it down somewhat - not always a bad thing.

House limits matter very much, because if the maximum bet the strategy requires cannot be placed, all the arithmetic in the world will not get a losing series out of the red and into profit.

For more than a dozen years, I have had a parade of self-styled experts wagging their fingers at me and warning me that experimenting with a sample of outcomes - however large - to come up with a betting method that will nullify the house edge is mathematically insupportable.

It does not matter to my critics that I have never offered a strategy based on a fixed data set.

What would be the point?

I do not agree with those who claim that the patterns of wins and losses created by games of chance are somehow magically and mysteriously unique, never to be repeated.

That is an argument that works wonderfully well for the gambling industry and is swallowed whole by most of its customers, but has little or no basis in reality.

From the start, I have viewed the challenge of beating house games as two-fold.

Firstly, a long-term strategy must be able to handle the worst losing streaks that blackjack, baccarat, craps or roulette are ever likely to come up with.

In other words, it has to actually work.

Secondly, a successful method must be able to slip under a casino's radar so that paranoid pit personnel do not quickly spot it and come up with ways to interfere with its use.

The notorious Martingale or double-up method fails because it sticks out like a Harlem Globetrotter at a pygmies' picnic.

Sensibly applied, double-up can be very successful for a very long time...but I still do not recommend it.

(Against the Jones baccarat sample, SM achieved a 6.63% win!).

Target betting has the untouchable "NB=LTD+" rule at its core, and that alone enables it to make the house advantage irrelevant.

All the other rules serve a dual role, part camouflage and part profit enhancement.

The purpose of this post is to present spread data from the more than 200,000 outcomes in the Lee Jones baccarat sample, but rules variations for target betting are always relevant.

The opening loss (OL) rule at least doubles the bet after an opening loss in a new series, simply because most house wins (or player losses) are followed by an opposite outcome.

The same applies to the 2L and 3L rules, which together make the first four losing bets in a new series a little like a Martingale. But not suspiciously so, we hope.

The win progression (WP) rule exists because any strategy that does not exploit winning streaks is, at best, wimpy and wishy-washy.

The WP basically doubles the bet (5, 10, 25, 50) until $100, adds $100 each round for as long as the streak continues, and ends the series if a loss is $500 or greater.

If a streak-busting loss is less than $500, its value becomes the LTD and NB=MIN(PBx2,PB+$100).

The mid-series loss (MSL) rule kicks in only if a potential EOS bet of $500 or less goes south, repeating the LTD+ bet in the hope of a probable opposite result, probable being a word with special meaning when it comes to gambling games.

None of these "extra" rules need to be defended here. They do a great job of both boosting the bottom line and bamboozling the other side.

I mention them only because the results below included their use.


As you can see, the spread analysis summary for the 202,214 Jones baccarat outcomes very closely follows those for baccarat and blackjack in the previous post.

That brings us to almost 400,000 hands or rounds, enough to keep a full time punter busy for at least four years.

They together demonstrate that the pretense that win-loss patterns in gambling games never repeat themselves amounts to disingenuous disinformation.

As before, losing tactics (those applying much tighter or narrower bet spreads) required far more action, and therefore greater risk, than the spread I recommend for target betting.

Also as before, by far the greatest number of bets valued at $1,000 or more was required by the spread with a maximum bet of $1,000 than by the recommended maximum of $25,000.

The models for this test always assume a "bust" limit of 40x the maximum permitted bet, so 1-5 busts at -$2,000, 1-50 at -$20,000, and so on.

Target betting's 1-5,000 spread busts at a cool million, but it didn't come close to that in this baccarat sample.

By the time an $855,000 slump came along (one that would have been easily avoided in non-robotic play), the strategy was more than $500,000 ahead, reducing the exposure of the initial bankroll to barely 35%.

By the end of the Jones baccarat test, target betting had paid down its seed money and was sitting pretty with almost $3,000,000 in funny-money profits.

Once again, the arithmetic of games of chance is telling us that tight spreads will kill you!

In this very large sample, spreads of 1-500 and under failed, although they at least managed to dent the house edge.

That alone is "impossible" according to the house-trained experts who have tried to talk me out of my mathematical heresy.

The same was true for the 80,000 blackjack rounds, although a 1-5 spread (about average for most weekend punters) actually increased the house edge.

The Rodriguez sample, which he assures me is taken from Zumma Publishing's baccarat "tester" books, was the toughest to beat, and only target betting's 1-5,000 spread came out ahead.

Something to keep in mind is that any prolonged test like these must deal with deadly downturns that would be far more rare if the "player" had been permitted the option to bail out whenever he felt the need.

In real play, table limits would make huge dips like those shown here pretty much impossible.

The "experts" argue that when a player bails out of what he perceives to be a difficult and potentially prolonged downturn, his chances of finding improved conditions elsewhere are no better than 50-50.

That may be true for a random or flat bettor.

But a target bettor needs only two successive wins and more often than not just one isolated win to recover his losses.

For him, the odds of a safe escape from a possible threat to his bankroll are better than 10 to 1 in his favor.

In any event, table limits require the target strategist to keep on the move, because betting too wide in the same location will blow his cover.

This whole challenge comes down to observation and experience.

Contrary to the conventional wisdom, win-loss patterns are remarkably similar not just from one representative sample to another but from different games.

You cannot accurately predict the result of the very next bet. But you can learn from patterns from the past and be confident that both good and bad will be endlessly recycled in the future.

You can then bet in such a way that you optimally exploit them.

The gambling industry does not want you to know that, of course.

Casinos need you to bet a relatively tight spread, to select your bet values more or less at random, and above all to think of losing as fun.

If you plan to win consistently, you need to learn how to maintain a low profile.

Target betting teaches you how.

The most important endorsement of the strategy comes, as it always must, from outcomes as yet unplayed.

That is why I tell readers of this blog not to take my word for anything. Learn the rules, then use them - that is the one true test of a betting method.

Which brings me to the latest BST results...




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: