_
My "spring cleaning project" (tidying up my data files so anyone can understand them) is going well, and I should have more to say about that by the middle of next week.
One topic that the task has brought to the surface again is the extent to which runaway sims differ from what a living, breathing player is likely to encounter during real play in real time for real money in a real casino.
The question at the top of this post sums it up, I hope.
It is fashionable for academic mathematicians to respond to my criticism of RNG game simulations by saying that I don't like them because my betting method cannot beat them.
They are free to bolster their position on the arithmetic of gambling with as many wobbly pit-props as they can find, but the truth is that no one alive plays the way a runaway sim tackles a gigantic sample of random numbers.
I have just started reading Sam Harris's controversial polemic, "The End of Faith," and take great delight in the analogies and allegories he uses to demonstrate the weakness of positions that some very smart people have been taking for centuries.
Mr. Harris and I have different objectives, of course, because I do not deny the existence of the house advantage at games of chance: I simply reject the contention that it cannot be beaten.
I fully accept that because of the house edge, any gambler is certain over time to lose more bets than he wins, and I have always conceded that the only way to avoid that certainty is to stay away from casinos and never place a bet.
Mr. Harris relies on irrefutable logic to back his arguments, and so do I.
Who in his right mind would question the proposition that if you win more when you when than you lose when you lose, then losing more often than you win will not cost you money in the end?
The argument is so solid that I often forget to add the qualification that of course the product of the percentage by which your average win value exceeds your average loss value must be greater than the product of the house advantage applied to overall action.
In other words, a difference of pennies between the AWB and the ALB simply won't cut the mustard!
Example: A 2% house edge was confirmed at the end of 100 hands because you lost 51 bets and won 49; happily (but not by chance), your AWB at $11 exceeded your ALB at $10 by 10%, giving you an overall win of $29 (+2.8%) rather than an "expected" or "indicated" loss of $21. Had AWB/ALB summed +3% (say, $10.30/$10), you would have won $504.70 and lost $510 for a final outcome of -$5.30 (-0.52%), less than negative expectation but not "less enough."
We must first agree that it is possible to consistently achieve an AWB that tops our ALB sufficiently to offset the house edge that most people think is unbeatable.
If you do not think it can be done, you should be reading someone else's blog right now (or be planning your next losing trip to a casino!).
I mention the Harris book because its premise is that throughout history, countless millions of people around the world have been persuaded to adopt beliefs that make no sense.
Getting gamblers to accept that games of chance are so uniquely magical and mysterious that they have their own special laws of mathematics is no stretch at all in comparison with the tenets that Mr. Harris cites.
The truth is that consistent winning does not take a miracle...just discipline, commitment, and an abundance of cash.
Target betting's primary rule, NB=LTD+ in response to a mid-series win, is on its own enough to achieve the "win more" goal at blackjack, and even betting the field at craps, because both games pay back 100% of the bet on a win, and occasionally more than that.
Blackjack's naturals and the x2 or x2, x3 craps "bonuses" for 1,1 and 6,6 have no equivalents in baccarat, so additional ammunition is often needed to get ahead of the house edge in that game.
Baccarat ranks below blackjack for target betting, but above craps and roulette, which should always be abandoned as soon as a series LTD begins to look like real money.
Logic supports both the OL (opening loss) and MSL (mid-series loss) rules, which seek to exploit the high probability that a house win will be immediately followed by a player win.
The math could not be simpler: isolated wins are the most common outcomes on both sides of the table, which helps the MSL and OL, 2L and 3L rules but works against NB=LTD+.
Paired wins come next, followed as we would expect by triples, quads and so on.
The core rule of target betting invariably pays off well ahead of serious trouble because the NB value is always modestly more than the LTD value, and an eventual win will save the day.
Thanks to the MSL, OL, 2L and 3L tactics, more often than not a target betting player does not even have to wait for two consecutive wins ("twins") - a single right bet will turn the recovery series around.
Given that reality, casino table games are logically less hazardous for a player who can recoup his prior losses in two wins or less than for a fixed-sum or random bettor.
Mathematicians repeatedly state that trying to run away from a prolonged negative trend is pointless in the long run because the house advantage will always see to it that matters will get worse ("out of the frying pan and into the fire") just a little more often than they get better.
Possibly true for a haphazard player, but logically not so for player in search of an oh so slight offset to a threatening downturn.
There is, it is true, no way of knowing if what might be a long losing streak will reverse itself if we stay put and tough it out instead of bailing out.
So we have to look at the numbers and apply them to any threat situation.
If a downturn continues (and, again, it may not), serious harm might come to our bankroll.
If we choose to suspend play for a while and walk away from a potential threat, the odds are better than 10 to 1 in our favor that we will quickly encounter paired wins, or a well-timed single win.
That is because most of the time, the win-loss pattern in a table game bounces back and forth (-1, +1, -2, +1, -3, +2, -1, +2) with just a slight bias for the house.
Paired wins show up, on average, every five bets or so. Single wins are slightly less than a 50% probability.
Most players bet a very tight spread, pick their bet values more or less randomly, and bet either too much or too little during a prolonged downturn.
They also hope (but mostly do not expect) to win with very little risk, and so they come to the table with an inadequate bankroll that is quickly wiped out if a long losing streak sets in.
Time to drag out a little homily: If you can't afford to win, you shouldn't bet.
Against most players, fluctuations in the win-loss pattern (WLP) do not have to be very great to hand the house the "contribution" it figures it is due.
Against a target strategist, the ups and downs of a gambling game have to be seismic for the house edge to prevail.
The gambling industry has always been very effective in sending players a mixed message which says, "Hey, we're offering you a chance to make more money than you could hope to earn any other way...but chances are you will lose."
Steve Weinberg, credited as the creator of modern-day Las Vegas, was on "60 Minutes" a while back telling viewers that the only long-term winners are the folks who own casinos.
I hate to use the name most people know him by, because to me a casino operator changing his name to WYNN is like a doctor adopting the moniker KYLL: It is insensitive at best!
So I will just say this in response to his cheery You Can't Win message: "Steve, you know better."
It is, however, very important for Mr. Weinberg's chosen business that its customers believe that winning is impossible, and losing is fun.
Money alone is not a guarantee that a player will beat the house edge and boost his bankroll in spite of losing more often than he won.
Money and a plan - the right plan - will invariably get the job done.
Spread, spread and spread are the three keys to success at gambling, which is why a casino does all it can to limit spreads, exploiting the average player's inclination to stick to the same game and keep betting far longer than he should.
And of course a target bettor will face severe challenges from time to time, and that is when he will need his discipline, confidence, commitment and money the most.
His exposure is greatest when he has reached his maximum bet, whether it is self-imposed or mandated by the house rules.
The good news is that in order to push his bets that high, the current house edge will have to have reached a number far above the norm for the game.
The experts tell us that past outcomes can have no effect on future outcomes and smugly quote what is sometimes called the Law of Independence of Trials.
In the short term, it is a powerful principle. Long term, it is hogwash.
When the house edge for a given sample reaches, say, 40% in a 1% game like blackjack (and it does happen, occasionally), we cannot state with certainty that there will be a significant offset in the next 10 or 20 bets.
We can however be sure that a swing against the house must come eventually in order for the laws of real mathematics (as opposed to mythematics) to be complied with.
It may not be - probably will not be - a 100% swing in the opposite direction from the prolonged negative trend.
But it invariably will prove to be enough, as long as the player's cash resources are likewise.
Prospects are helped by the fact that the rules of target betting simply will not permit a player to stay in one place longer than it is safe to do so.
After starting a new series with, for example, a $5 bet (at craps or roulette or even 3-card poker if all the available blackjack tables have a $25 minimum), it is probably safe to spread as wide as 1-20 or $5 to $100 in a busy casino before moving on.
Generally, a 1-5 spread is as wide as you should go in one place, to avoid attracting the unwanted attention of pit personnel ("Thousand in play! Up from $100! Come get him, Boss!").
That means that in a prolonged downturn, you will be moving from layout to layout and even game to game quite a bit before EOS.
And each time you do that, there is better than a 90% probability that EOS is just a few bets away.
There is also a 90% probability that if you had stayed where you were, recovery would have happened in a hurry.
But since your NB would have had to have been inside the danger zone, who cares?
Runaway sims that depend on robot "players" wandering blindfold into a minefield require us to believe the preposterous without question.
Some people do that every day of their lives. You don't have to...
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.
_
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment