To answer questions about Oscar's Grind, it's better than flying by the seat of your pants, but target betting is better still.

_
I get questions about Oscar's Grind from time to time, usually from people who want to suggest that there is a more than coincidental similarity between Oscar's variation on the money management theme and my own.

The two methods are as alike as an apple and an orange (both grew on a tree, both are spherical and have pips, both taste good and...that's about it).

I have always enjoyed the story of the origins of the Grind, which legend has it was created by a crusty old geezer who played just craps and used a stubby pencil to record the result of each roll in a battered notebook. Too many adjectives!

I'm more inclined to believe that like card counting, Oscar's creation is encouraged by the casinos because most people mess it up, and when they do, they lose more money than they probably would otherwise.

The idea as I recall it from Tom Ainslie's book How to Win in a Casino is to freeze the bet after a loss, add one unit after any mid-series win, and revert to a minimum bet after recovering prior losses for the series, plus one unit.

If the next bet in a series would exceed 20 units, it's Game Over and back to a minimum wager, and that's the part that makes me suspect that "Oscar" was actually a casino employee with evil intent!

A 1-20 spread has zero chance of winning in the long term, with or without following the OG rules. That's an absolute factual fact. With the loss limit removed, "Oscar" does marginally better than a random bettor on a winning streak and can stay ahead for a good long while with a decent bankroll.

Against the BST blackjack data set, for example, OG earned about 5.0% of target betting's win to date while racking up more than 50% of my method's overall action. That translates to a greater risk with a far lesser reward.

Mr. Ainslie's description of Oscar's Grind makes no provision for a win progression, which seems to me strangely short-sighted unless there is an ulterior professional motive. Incorporating +1u after an opening win, continuing until a loss, then freezing the bet per the standard rules and playing on until the loss is recovered +1u, pumps OG's win by more than 30% while increasing the action by less than 10%. Why not do it?

OG is a progressive betting method, of course, so it wins more often than not. It's just not a very good progressive betting method!

A simple Martingale does much better, my version of a Martingale is more effective than the original, and target betting knocks them all into the proverbial cocked hat (one of these days, I must look that one up!).

Now seems as good a time as any to revisit the topic of which games to tackle with target betting.

Blackjack is by far the preferred option because of its low net house edge and the extra profit opportunities offered by double-downs and splits and 3-2 payoffs for naturals (never play at a layout offering 6-5, unless you are stuck at the last blackjack layout on earth).

Field betting at craps is a great place to start a series and take a break from blackjack, because of the x2 payoffs for 2 and 12 that are still x2 and x3 in some casinos. The negative odds are more than 5x blackjack's with x2x2, so you should never let the bets get too high before bailing out for a safer game.

Baccarat is the #2 option, but you should never, ever bet on Banker. That so-called "5 percent" commission is one of the great lies of gambling because it can easily swallow all your winnings and leave you in the hole in spite of your best efforts. I'll be posting more about that one of these days, with the help of the baccarat data sets from Lee Jones and Lorenzo Rodriguez (all 300,000+ of 'em!).

Roulette is a wonderfully streaky game best played (in my opinion) with the "wobble" method of picking black or red, odd or even, or whatever - never "inside" bets. Wobble is from WBL which in turn means Win Before Last, or as the French have it, avant derniere (or is it the other way around? I can never be sure!).

The big problem with roulette, of course, is its 5.26% house edge. So, again, kick off a new series (or several) at roulette if you must, but always back off before the NB gets out of hand, and take your LTD/NB numbers to a more player-friendly game.

Just don't rely on old Oscar, may he RIP. Mr. Ainslie claims to have paid for several gambling trips to the Bahamas with that method. Guess he's just a lucky cuss...?

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.
_

The conventional wisdom says a million wins don't matter and losses are all that counts. True, but false at the same time!

_
A mathematician arguing for the YCW (you can't win) philosophy will tell you with a straight face that progressive betting can never be profitable because more bets lost than won must always result in more money lost than won.

No amount of evidence to the contrary will be considered acceptable, in the same way that for centuries, the conventional wisdom dismissed data showing that Planet Earth was not flat.

It is of course completely rational to warn against a betting strategy that will occasionally suffer a catastrophic loss that will always wipe out all prior profits and bankrupt the player.

But progressive betting will only match that caveat if the method is applied the way the casinos and their house-trained experts want it to be, in real play or in computer simulations.

That means that the player must keep on betting through prolonged downturns, and meekly throw in the towel whenever his next bet is prohibited by the prevailing table limit.

The ideal gambler is, in the view of all casino operators and the mythematicians who promote the gambling industry's ordained right to the contents of all our wallets, an impetuous spendthrift who comes to the table with an inadequate bankroll. He then bets a very narrow spread, and sheds first his inhibitions under the influence of "free" booze, then his stash.

It is theoretically reasonable to suggest that since target betting my way lost just once in some 70,000 recoveries, that loss might have occurred in the very first session, and therefore the method is potentially too hazardous to merit consideration.

It is theoretically equally reasonable to argue that since 1 in 70,000 American motorists will be killed or seriously injured in a road accident on any given day, the car in your garage is potentially unsafe at any speed and you would therefore be ill-advised to drive it ever again.

I cannot refute a statistically supported recommendation that the only certain way to win in a casino is to keep all your money in your pocket and, better yet, never step across the threshold.

But short of an outright ban on gambling, the evidence shows that progressive betting is a better way to win than any of the alternatives.

Those of you who come to agree with me after a few hundred hours of cautious exploration and practice (I certainly do not expect you to take my word for anything!) will learn that target betting can be modified in all sorts of ways to suit your gambling personality and your resources.

In fact, the more modifications you develop, mixing and matching them as you see fit, the less likely you are to be identified by casino pit personnel as a player who wins "too often" and should be stopped.

Elsewhere in this blog is data that show that even the much-derided Martingale or double-up betting method (-1, -2, -4, +8 etc.) demolishes the house advantage in the almost 400,000 real-time, real-play outcomes that are the subject of the current trial.

Played right, double-up can be a powerful antidote to the house edge. If the casino you have selected allows you to play it at all, that is.

Target betting with my rules applied is far harder to spot, and the modifications that I recommend or that individual players will develop for themselves make it even more effective in the real world.

Here's the latest BST data set:

(Click on the image to enlarge it)

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.
_

Winning with target betting is easier, quicker and just plain better than all the alternatives. For the others, it's mostly impossible.

_
When the contest is between progressive betting and the only alternatives (flat/fixed or random betting), it's no contest at all: progressive betting wins, the other methods do not.

And since this has always been as much a mathematical challenge as a practical one, tailoring my version of progressive betting to the needs of the underfunded player seeking to beat the odds during occasional weekend trips to Las Vegas has never been a priority.

One simple truth I never lose sight of is that my application of the demonstrably sound principle I call target betting (the rules that I apply from first bet to last in each recovery series) cannot be claimed to be the optimum approach.

There may well be a better way to do this, and I am confident that if I fail to find it before my final visit to a casino, someone else will, and hopefully they will tell the rest of the world about it for free.

Why not sell my ideas rather than giving them away?

It is not because, as some of my critics have suggested, my ideas are worth even less than I charge for them.

It's because for me, this has always been about exposing as myth an axiom that I always felt served the gambling industry far better than the people whose wallets the casinos suck dry.

More than 30 years ago, I experimented with the simplest, most basic form of progressive betting in several Las Vegas casinos, and in every case encountered impressively efficient interference by pit staff.

Drunks, neophytes and other suckers were committing fiscal self-destruction all around me, and I was the only one singled out for friendly advice about how best to preserve my bankroll and save myself from certain ruin.

Why are all these nice people so afraid of a betting method that they say is a danger only to me and my money? I wondered, since fear (or at best, self-interest) seemed to me the only reason they would take me under their wing while leaving everyone else to their own devices.

In the years since, gambling industry representatives (many disguised as truth-seeking academic mathematicians) have either denied that casino staff will ever step in to stop progressive betting, or have patiently offered up implausible explanations for a policy that anyone will encounter within minutes of starting a Small Martingale.

Some examples:

1. Anyone dumb enough to try double-up or Martingale betting has obviously never wagered in a casino before and is in serious need of advice and counsel. Interesting, but implausible. Pit people in reality never offer advice unless asked, at which point they are usually courteous but non-committal, often starting out with, "The book says..." if the questions are about blackjack, or "Some people like this bet because..." Drunks hurling their money down the drain are only cautioned or removed from a layout if they upset other players by hurling something else, or knocking over drinks. Idiots who play or bet badly but don't ask questions are invariably left alone.
2. Pit staff know that a double-up player can do very well for a short period of time but will always crash and burn eventually. They are just being friendly when they advise against the use of this suicidal betting method. And pigs play poker sty-lishly! The entire gambling industry is geared to relieving players of their bankrolls as quickly and quietly as possible, while constantly reminding them that losing is "fun." The notion that a betting method that is certain to fail would be blocked is beyond silly, but no one is ever allowed to bet double-up without house interference. Casinos are constantly devising new long-odds side bets as sucker bait to bolster the relatively low house edge at blackjack and baccarat, and adding non-traditional table games that make roulette and pai-gow poker seem like easy money. If the Martingale was a surefire loser for gamblers, casinos everywhere would be promoting it as an "Exciting New Feature!!!" and printing new layouts to highlight it. Instead, pit staff respond to it like vampires to bright sunlight (not that I am suggesting that those nice people are bloodsuckers, mind you).
   
3. A progression will always fail because once the bet exceeds the table limit, prior losses cannot be recovered, and casino staff feel you should know the risks you face. This is one of my favorites, because it betrays how little the experts think of the rest of us. A $5 opening bet at a $5-$500 layout will bump up against the table limit after 7 consecutive losses requiring an 8th bet of $640. It doesn't matter that a losing streak that long is less than a 1.0% probability: it happens. A Martingale bettor who has a bankroll that is up to the challenge (quite likely, because he never loses if he is allowed to play) simply takes his business to a layout with a higher table limit (say, $100-$5,000) and continues betting at the required level. One win, and he starts the process all over again. If he is playing blackjack and he draws a natural, he's in clover! Given a house limit of $25,000 a double-up player will have to lose 13 bets in a row to hit the "green ceiling" and that's just a 1/8192 probability. And he's not dead yet...although that's a topic for another time.
   
4. If double-up really was a winning method, more players would use it. No one does because it's suicide, and casinos discourage anyone who plays irresponsibly. This is one of the biggest of the big lies that gamblers are asked to swallow whole. And most do. In fact, anyone who has ever spent more than a few hours at casino table games will have seen "Martingalers" at work, probably without recognizing them. They are the annoying people who place two or three bets, lose, and move on, leaving people who have been glued to their seats for hours wondering why their game has been so rudely disrupted (or so they believe). It happens a lot in a busy casino when empty seats are hard to come by and dealers and pit personnel have too much going on to monitor any one player's movements. The rule is three bets at the most at any one layout, then on to the next game. Casinos try to thwart Martingalers by prohibiting "entry" between shuffles at blackjack and baccarat, but that's just a minor inconvenience. Craps and roulette tables are never restricted, and field bets at craps have special appeal because of the 2x payouts for 2 and 12, which used to be 2x and 3x before the accountants that devised 6-5 payouts for naturals at blackjack did a little more dirty work.
5. Progressions are a bad idea because the ancient laws of mathematics state that any amount bet against a negative outcome must eventually have a negative outcome. Pythagoras and his peers and successors were and still are in essence stating a rule confirmed by logic, which is that because negative expectation will ultimately result in more player losses than wins, more money lost than won is an inevitable consequence. And as long as bets are of a fixed or randomly selected value, a long term win is highly unlikely. However, a progression is not random. Bet values are determined by the immediately prior outcome, and while that outcome is random and subject to the negative expectation for the game, the response to it never varies. A Martingale is a bad idea not because it does not win consistently but because it is too obvious. It requires camouflage to avoid quick detection, and that is what target betting is all about.
   

My conclusion long ago was that constant re-doubling is impractical in spite of the satisfaction gained from repeatedly recovering prior losses in a single bet.

Waiting out a losing streak by freezing or reducing the wager and then going for a one-bet turnaround seemed like a mathematically-supportable alternative, primarily because in most table games, the house's chances of winning a given bet are only marginally better than they player's.

Put another way, while each bet at a game with a 2.0% house edge faces negative odds of 49-51, there is no such thing as a fractional win or loss: you either beat the negative number, or you don't.

All of my blackjack and baccarat spreadsheet models track the performance against the same sample of outcomes of both a standard double-up method (the Small Martingale) and a variation that I call SM-Plus which modifies the progression to deliver a bigger win from drawn-out recoveries.

Against the blackjack outcomes (78,000 and counting) the standard double-up had one "bust" in 18,972 series, the pumped-up version had 3 busts in 18,055 series, and target betting had...no busts at all.

Here's the SM, SM+ and target betting summary for the blackjack trial to date.


(Click on the image to enlarge it)
The baccarat trial is especially relevant in this context because the data sets were supplied by two verifiable sources that are independent of each other and (more important) independent of me.

There were well over 300,000 rounds of baccarat in all, equivalent to almost three years of play for a very dedicated player putting in 30 hours a week with two days off each week but no vacation time.

Here's a summary of how a simple Martingale performed, along with data for the more aggressive double-up method, and for target betting.


(Click on the image to enlarge it)
I guess it's possible to make this stuff up, but why? The results achieved have been so consistent (predictable is another way to put it) that anyone applying the rules of these different approaches to target betting fairly and accurately is certain to see very similar numbers.

The other question always begged by results like these is how much better any disciplined betting strategy might do with human intervention allowed, meaning that prolonged negative trends might (or might not) have been avoided if defensive action had been taken.

High-speed simulations of any kind, my own included, require the wholly unwarranted assumption that a player in the early stages of one of those deadly downturns that we all know can occur (those of us who actually place bets in casinos, at least) would NOT back away from a punishing situation and resume play somewhere else. There are some certifiable masochists out there, I know, but I am not one of them.

The conventional academic argument is that since each outcome is independently subject to the same measure of negative expectation, attempts to escape a negative trend can have no long term benefit.

It may be true that a player who bets fixed or random amounts is as likely to get into more trouble after a move as he is to see an end to the negative trend that made him flee.

But for a target bettor, all that is needed is to get out of trouble is two consecutive wins, and the decision to bail out of a bad run is as much about morale as it is about math.

The math, though, supports preemptive "damage control" when a recovery begins to drag out uncomfortably, because the great majority of series (well over 60%) are over in six bets or less, and a tiny percentage drag on for 10 rounds or more.

As the Chinaman said, "A long losing streak must always begin with a short one, and it is better to lose a little than a lot."

Worth noting is that in less than 80,000 rounds of blackjack, target betting "won" (how I wish those quote marks were not necessary!) almost as much as it did in almost four times as many baccarat bets, confirming that blackjack is clearly the optimum table game for this strategy.

Here's the latest BST session...

(Click on the image to enlarge it)
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.
_