Sunday, May 24, 2009

Irrefutable logic supports target betting as the best way to win consistently at casino table games. So yet again, the conventional wisdom is wrong!

_
(The truth is that when the word conventional precedes "wisdom" it is more often than not a synonym for false or erroneous. Remember the flat earth theory? Or all the "expert" assumptions about house prices and Wall Street?)

This has never been an argument about whether or not the house advantage exists in games of chance. Of course it does, or the games themselves would not exist.

And I have never disputed that the house edge is the undoing of at least 99.9% of all gamblers: anyone who bets either relatively flat or random amounts is sure to lose eventually.

I am accused of being innumerate or (my favorite) discalulic, because I can prove the obvious over and over again.

And in the minds of many who consider themselves smarter than the rest of us, when the obvious contradicts the conventional wisdom, then the obvious is obviously wrong.

You would think that the following statement is too logical to run any risk of being disputed:

If the combined value of a smaller number of winning bets exceeds the combined value of a greater number of losing bets, then the negative expectation derived from (L-W)/N or losses minus wins divided by number of bets is not relevant to the end result.

In other words, "If you win more when you win than you lose when you lose, then losing more often than you win will not hurt you."

But by golly, stalwart defenders of gambling's status quo contradict the notion all the time.

There is, they claim, no way to prove the "win more" concept, because past outcomes from games of chance are always anecdotal or subjective and can never be used to predict the future.

Never mind that the target betting principle can be applied successfully to any sample of outcomes of almost any size from any honest and objective source.

Or that using just the "LTD+" rule and none of the camouflage against countless outcomes as yet unplayed will confirm that target betting - progressive betting - makes the house edge irrelevant.

My critics keep falling back on the argument that if you lose more bets than you win, than you must in the end lose more money than you win.

And that makes me wonder exactly who those critics are, since casino operators everywhere know full well that their games can be beaten and are constantly on the lookout for players who threaten the house's bottom line.

I take comfort in the fact that those who are skeptical of target betting contradict themselves and each other all the time.

For example, the hundreds of thousands of real-play outcomes that I offer as indicators of target betting's power over the house advantage are routinely dismissed as anecdotal and mathematically suspect.

But when I confirm that there were two "busts" among 59,901 recovered series (a win rate of 99.997%), the response I get is: "Aha! That proves that your ideas are nonsense!"

Target betting has so far achieved a "hold" of 8.1% against 84,000 rounds of blackjack with a collective house edge of between 0.74% and 5.7%, winning $306,530 when it "should have" lost at least $27,000.

It won about the same amount against 202,000 rounds of baccarat, in spite of an indicated loss (house edge of 1.19% x total action) of $101,000.

It "tanked" twice in an additional 114,000 rounds of baccarat, confirming that target betting is not a 100% winning proposition, and therefore should not be treated as such by the gambling industry.

I look at it this way: Odds of better than 29,500 to 1 in my favor are logically and irrefutably superior to odds of 49-51 against me.

The caveat that we all need to be aware of is that the application of any rigid set of rules to a very large sample of outcomes, retroactively or in advance of play, contravenes normal human behavior.

Likewise, the pretense of accuracy, honesty and relevance for a runaway sim that denies the "player" any response to conditions and circumstances is disingenuous.

The Wizard of Odds, for example, is a well-funded shill for the gambling industry. His website touts his $20,000 "systems challenge" and its reliance upon a billion-bet simulation to "prove" that the house advantage is always unbeatable.

But what does any sim - let alone one that claims to represent at least 10,000 years of continuous betting! - have to do with real play? Answer: Nothing.

I have been using variations of target betting for more than 30 years now, and my application of the flexible rules depends on the current level of two things: my confidence and my bankroll, which while connected do not always move up and down together.

There is not a player alive who would sit through the kind of punishment that is routinely dished out by simulations that eliminate not just the critical human element but cards, dice, tables, wheels, balls, chips and real time.

Sim proponents like the Wizard of Odds (a great moniker when it was first used by someone else on TV in the '70s) would have us believe that human behavior can have no long-term effect on the probabilities that apply to gambling games.

That makes about as much sense as a claim that a rider cannot control a horse, or that a speeding car will crash with or without a driver behind the wheel who knows the difference between the gas and the brake pedal.

For a flat or random bettor, it may be true that "damage control" will make matters worse about as often as it improves them.

But a target player can recover prior losses in a fraction of the time it took for him to get into trouble in the first place. And a win-loss pattern that will cause him serious problems is, at worst, about a 10% probability.

The truth about the random game outcomes that we call hands, rounds, spins or rolls is that while they are individually unpredictable, in concert they form patterns and trends that are reliable and cyclical.

We know, for example, that most wins for either side, player or house, are immediately followed by an opposite result, that half as many are paired, and half as many again constitute a three-bet "streak." It's called arithmetic.

And in spite of all the "expert" claims to the contrary, the rules of probability keep a tight rein on the pattern variations that can occur in the course of a game of chance like blackjack, baccarat, craps or roulette.

There are exceptions to every rule, but they are rare enough to be of almost no consequence.

Here are some results that the conventional wisdom says are impossible, or at best fraudulent:


The good news for you and me is that the results above (for baccarat and blackjack samples with the optimal target betting rules applied) are scrupulously accurate, and confirm a consistency that is mathematically predictable.

The more complex the target betting tactics become - with the objective of increasing win values while camouflaging the strategy from prying eyes - the wider the variation between one positive outcome and the next.

And since additional strategic switches necessarily increase overall action and, by extension, potential risk, it makes sense to add them gradually, as the bankroll grows stronger.

Adjusting a betting method intuitively to suit circumstances and/or available resources is not something that "sim" supporters like the Wizard of Odds will ever permit.

Their systems-busting weapon of choice will only work for them if the "player" at the heart of them behaves like a suicidal idiot.

Like hurricanes in The Hamptons, aberrational departures from probable win patterns hardly happen.

But they will occur, and the player response to them before they can prove fatal becomes absolutely critical, while being blithely ignored by the Wiz and his ilk.



As the caption above points out, casinos themselves are much less impressed by computer game simulations than they are by real-time research out on the floor.

No new game, or modification to the rules of an old one, will ever happen without reliable data "from the field" to support its predictable contribution to the house's bottom line.

The summary below confirms that when the non-essential (but very effective!) target betting switches are turned off, session outcomes fall within a much more narrow range.


One of the jobs of a blog like this is to state the obvious the way the "experts" do ad nauseam, so let me say that given an average per-session win of around $10,000, it makes very little sense to persist with a million-dollar "bust" limit.

I would urge readers to take a very close look at this summary, and pay particular attention to the numbers on the far right, which track the win per round for each session.

Those numbers are, you will see, mostly within a percentage point or two of one another.

They confirm the consistency and predictability that I felt confident must exist in win-loss patterns in games of chance when I began this quest in 1978.

The "sore thumb" above, of course, is the double-whammy red splat next to sessions 14 and 15 from the Rodriguez baccarat data set.

Damage control would not have been merely optional in those two instances, it would have been strictly enforced by the casino's own table limits.

That being so, would the huge losses indicated in the summary have been possible in real play?

All the other numbers in the summary answer that question: No way.

Another question that is begged, of course, is the effect on all the other results of spread limits enforced throughout exactly as they would have had to have been in the losing sessions.

The honest answer: little or none.

That is because prolonged negative trends are rare events, and the chances of one of them continuing from suspended play into the very next session are slim indeed.

In the next post, I will put up results from the baccarat+blackjack database with a $25,000 to $250,000 spread limit applied.

Ignore any experts who tell you that what I am doing here amounts to experimentation with anecdotal data until a winning strategy is found.

Right now, all we are concerned with is the application of a target betting rule that has been in the public domain for a dozen years, plus an end-of-series rounding up rule which is nothing more "experimental" than a nod at reality.

Given a winning previous bet (PB) of $5,000 and an updated loss to date (LTD) of $775, would would your next bet (NB) be?

If your answer is $775, you are not wrong, exactly, just a little out of touch with the demands of real play...

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: