Monday, March 14, 2016

It's an international obsession, so let's get this straight: Oscar's Grind can help you beat negative expectation, but Target 3-Play can make you richer faster, with less risk. Just forget about betting strategies that are not progressive!

Now I've restated that important truth about gambling, I'd like to put on record that I am happy to see daily blog traffic hitting about 3x the numbers from this time last year, but I am still surprised that first-time visitors don't search for the latest post, then work back in time!

I say that because Target (formerly Turnaround) has always been, and was meant to be, a work in progress, and the nature of evolution is that the latest version of anything under the sun can usually be assumed to be better than whatever went before it.

I don't pay as much attention to this blog as I used to, but this week, I took my own little journey back in time and was reminded that my post about Oscar's Grind long, long ago remains the most read of all the 250-plus articles I have published here.

I learned about OG from a book by Tom Ainslie, and for a while accepted his claim that he had used the strategy many times himself and had won enough to pay for several gambling trips to exotic casinos in the Caribbean.

I loved his story about an ancient geezer named Oscar who spent his life playing craps, scribbling the outcome of every roll in a beaten-up notebook with an equally clich├ęd stubby pencil.

I even gave the Grind a go myself, doing well enough over several sessions for one sarcastic pit boss to offer to send me a weekly cheque for an agreed amount to save me the bother of visiting his casino day after day.

Even then, I was convinced that only progressive betting could beat the odds in the long term, and Oscar's Grind is basically a restrained, almost constipated, version of a simple double-up or Martingale.

It can do pretty well for quite a while, and is demonstrably a better way to go than setting bet values randomly, the way 999 out of every thousand gamblers tend to do.

I finally abandoned the method because it was boring.  After a while, winning is not enough.  There has to be at least a little excitement, which translates as jeopardy, however brief.

To put this in perspective, let's first clarify the conventional wisdom about casino games of chance.

The CW says that since even the smartest player is sure to lose more bets than he wins in the long run, it is a mathematical certainty that ergo, he must in time lose more money than he wins.

This is a line that is relentlessly promoted by casinos, their employees and their online shills, and it is simply not true.

You will have heard, no doubt, that it is possible to fiddle and fart around and fudge the facts for a while to come up with a betting strategy that wins against a relatively small sample of recorded outcomes (say 5,000) but the same strategy will crash and burn against the next data set of similar size.

That's a flat-out fib, folks, but we can all see that it is in the interests of the gambling industry to have us believe it.

For example, if we take 50,000-plus outcomes painstakingly logged from real-time casino games and simulations, even Oscar's Grind in its original version can show a modest profit:

The Ainslie version of OG calls for the method to be abandoned after a 20-unit loss, and if you try that, you won't see even the weedy long-term win shown above.  And you certainly won't cover the cost of a week in Barbados!

I tweaked Mr. Ainslie's (or old Oscar's) approach by tossing out the stop-loss, and that did OK.

One of the frustrations of the original rules set was that it made no provision for an opening winning streak in a new series or sequence of bets to be exploited, so I threw in a plus-one response with this effect:

Better, but still constrained by the rule that no bet can be more than the total loss to date (LTD) in the current series, plus one unit.

For almost three decades, I have been repeating the previous bet if the LTD drops below that amount, so if a 50-unit win reduces the LTD to, say, -20, I bet 50u again hoping for (but obviously never certain of) a repeat or "twin" win that will end the losing series or sequence with a 30-unit profit rather than +1.

Against the same real-time data set, NB = PB turns out like this:

Not too shabby.  But still shy of a proper return for all that hard work (53,700 rounds is about a year and a half of play five hours a day, assuming 150 rounds per hour).

So...what to do about pushes or ties?

Firstly, if baccarat is your game, never, ever bet on the tie.  And while you're at it, never bet on Banker either: the so-called 5% commission will cripple you.

Instead, simply double the bet after a tie, with this effect against the same sample, assuming OG rules:

Now we're getting somewhere!

I want to stress that this isn't a change in the rules to accommodate this particular sample of outcomes, any more than the rules of Target are ever tweaked to falsify results against whatever set of numbers comes its way.

Repeat readers will know that T-3 thrashes all the Zumma baccarat samples, as well as the thousands of simulated shoes posted online by the Wizard of Odds, aka the casinos' Number One shill.

So how does 3-Play do against the 53,700 mixed-game outcomes pitted against deal old Oscar?

Here you go:

Definitely better, we can all agree.

I am often asked how much Target needs to deliver a long-term win, and of course there's no definitive response to that question.

You might as well ask how much money you'd need to open a casino.

In both cases, the best answer is: A LOT.

As a general rule, a strategy player has to learn to think like the House, accepting that his entire bankroll is the bet, and once in a very great while (just as in high-stakes poker) all the money in the bank will have to be pushed out on the table.

The casino has "the arithmetic" on its side.

But so does the Target player, if random betting is set aside with disciplined consistence.

3-Play prevails again and again because the house needs three successive wins to threaten our bankroll, while we can get out of trouble most of the time with just a single isolated win, and when that doesn't cut it, "twins" will usually get the job done.

As for the mathematics of games of chance, a player will see tripled wins about as often as the house: -1,-1,-1 is less than a 15% probability and so is +1, +1, +1.

Simulations far smaller than 53,700 outcomes can sometimes defeat T-3 (and OG) but we have to ask ourselves if a real player would allow himself to get into serious trouble against a prolonged losing streak without taking any defensive action.

The best response to a killer losing streak is to walk (or RUN!!!) away and then resume play at another table, or even a different game in another casino, maintaining the NB and LTD values as if there had been no interruption.

I start to feel uneasy if in a given recovery series, the house gets five bets or more ahead of me.
I won't take flight after every run of five consecutive losses, because often a quirky slump like that can follow an equally unusual run in my favour, but by then, I am certainly paying attention.

Freak losing streaks happen, so the best rule is to suspend play the moment you begin to feel under pressure.

You'll be wrong about as often as you are right, but you will feel better for protecting yourself, and that's a good thing.

"Sims" rely in what I often call the inertia fallacy, a blatant falsehood which is critical to the casino industry's never-ending campaign to convince you that only blind luck can make you a (short-term) winner, and losing is lots of fun.

Sure, progressive betting pushes bet values to scary heights from time to time, but the moment you impose too-low limits or resort to unrealistic stop-loss triggers (like Tom Ainslie) the more certain you are to be a long-term loser.

The casino doesn't care how much you bet, as long as you bet randomly, and do so ever more erratically as "free" cocktails kick in to cloud your judgment and make you play like a fool.

The house can afford to do that because against random and/or low-limit betting, more losses than wins (an inevitable consequence of the house edge) will always result in more dosh losht than won.

Bet like a pro, and you will keep building a bankroll that will make you harder and harder to beat as time goes by, because smart play will put the numbers on your side, not the house's.

Better yet, start with a gigantic bankroll and courage and confidence to match.

That's what the house does.

And remember...If you can't afford to win, you shouldn't play.
(I have said all this before, more than once, but most readers don't take the time to scroll through 250 articles...and who can blame them?).

_ 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. One more piece of friendly advice: If you are inclined to use target betting with real money against online "casinos" such as Bodog, spend a few minutes and save a lot of money by reading this. _