What Am I Doin' Here?

A few years ago I was eking out a living on the Don'ts at a $1 craps table that was taking no prisoners. Periodically a middle-aged woman would come to the table, make a few Pass Line bets, lose about $10, and leave. Halfway through her fourth or fifth visit she noted that I was holding my own. She held out her hand with five $1 chips in it to me and asked, "How would you bet this if it were your money?"

My response was, "What is my goal?"

I immediately regretted my frivolity, for the look on the woman's face was quite pitiful. She had obviously been taking a beating, some of which I had witnessed, and had turned in desperation to me, one who was ostensibly succeeding, for advice. To be sure, I could have been more sensitive to her plight, yet I defend my intent. Without a hint of what the woman wanted for her nickel, I could not suggest a betting strategy, for the best chance to turn $5 into $10 is not the best chance to turn $5 into $100, or to make $5 last for 30 minutes. Her problem, besides the dark cloud that had been following her around, was that she had no goal other than "to win."

My point is that there is no one-size-fits-all betting strategy, one that gives every player the best chance of achieving his/her goal for a session. The reason is simple: different players have different goals. Oh yes, everybody wants to win, but how much (relative to starting bankroll), and is that the primary goal? How important is playing time? What about comps, or the prestige or thrill of having a lot of action? These (and I am sure other) considerations all contribute to the gambling experience in different proportions depending on the personality and attitude of the individual.

For a player, then, the trick is to employ a betting strategy that accents the factor(s) of greatest importance. Step 1, of course, is to identify those factors, to answer the question posed by the title of this article. Step 2 is to identify the factors accented by a system one is contemplating using. Unfortunately, few authors mention what those are for their systems. Obviously an exhaustive analysis of every published system is not possible, so instead I present the following categories of strategies along with the goals that realistic players who use them should be hoping to achieve.

Winning progressions call for one to increase bets after each win, dropping back to some base level after each loss. The simplest winning progression is a full parlay for some number wins, but typically less aggressive increases, 50% or less, are used. This is a "go for the gusto" method aimed at maximizing profits for a long winning streak while riding out losing streaks with minimum bets. Users of winning progressions accept frequent but relatively small losses in return for the rare bonanza.

Losing progressions are the inverse of winning progressions, calling for bets to be increased after each loss. The simplest and perhaps oldest such progression is the simple Martingale, in which bets are doubled so that each win shows a net profit of the first bet in the series. Variations on this theme soften the blow of a losing streak by recovering gradually from prior losses instead of making them all up on the first win of a series. Users of losing progressions seek frequent but very small wins, in return for which they are willing to risk the rare catastrophic loss when they inadvertantly sit down in a handbasket.

Regressions start out with the biggest bet(s) of a series, then reduce the bet(s) after each win. This strategy is typically confined to games like craps and roulette, where a combination of bets can be made which offer a better than 50-50 chance of winning one of them. A simple regression strategy in craps is to place the 5 and 9 for $10 each, and the 6 and 8 for $12 each, a bet combination known as "44 inside." The player is a 3:1 favorite to win $14, after which the player takes down the bets on 5 and 9, and reduces the bets on 6 and 8 to $6 each, thereby locking up a $2 profit while still having a 5:3 chance of winning an additional $7. Users of regressions have a lot in common with users of losing progressions: they seek small frequent wins at the risk of larger but less frequent losses. The difference is that regression losses are limited by the first bet and therefore tend to be smaller, but alas, more frequent, than the loss when a losing progression fails.

Hedging strategies use offsetting bets (i.e., one bet wins when another loses) to reduce the magnitude of bankroll fluctuation. A simple example from craps is betting $1 on Any Craps in conjunction with a new $5 Pass or Come bet; the Any Craps bet "protects" the other bet by winning $7 when the $5 Pass/Come bet is lost to rolls of 2, 3, and 12, collectively called "craps", while a natural Pass/Come winner will net a $4 win. Another flavor of hedging, unique to craps, takes advantage of the fact that once a Don't bet is established on a number, the player is the favorite to win even money and can therefore guarantee a profit by making a Place bet on that same number for a lesser amount of, but which will pay more than, the established Don't bet. Either way the player is paying a premium in expected value for shrinking the variance, so hedges are best suited to players who value playing time and/or comps.

Waiting strategies presume that recent events can be used to predict the near future. In the case of card counting, this predictive value is mathematically provable; the value of other waiting methods (e.g., charting, where the player tracks some number of decisions to determine a trend and then bets that the trend either will or will not continue) remains debatable. Nevertheless, even if the assumptions they make are invalid these methods reduce their users' exposure to the house edge (by having the player make no or minimum bets) while they wait for the triggering event to occur. Hence, the patience of waiting systems users is rewarded with longer playing time.

Hybrid strategies integrate two (or maybe even three) of the above basic strategies into a system to offset the weakness of one strategy with the strength of another. A perfect example is the New York progression, so named because the beginning sequence of bets (in units), 2-1-2, matches the area code of New York City's Manhattan. By starting with a regression this system requires only one win to show a profit for a series instead of two, three, or more for a straight winning progression, but retains the potential of cashing in on the winning streak of the ages with the sacrifice of only a few units. Hybrid strategies appeal to players who seek a compromise between the goals of the strategies being combined, for example, not as many losses at the expense of the really big win in the case of the New York progression.

Three to four years ago a Kansas City dealer, using the handle "mykey", posted the best analogy I have seen on betting strategies to the rec.gambling.craps Usenet newsgroup. (I am modifying it slightly, but the basic idea is his.) The player is presented with a panel of 100 unmarked buttons. When pressed, 48 of the buttons produce a pleasant sensation, while 52 of them cause the player some mild discomfort. By varying his/her bets the player is rewiring the panel. Unfortunately, the mathematics of any casino game dictate that the total amount of pleasure and pain each remain constant. If 99 buttons are wired to produce pleasure, that pleasure will be barely detectable, and pushing the pain button will be fatal.

So ya bets yer money and ya picks yer own schematic diagram. No wiring can guarantee a positive experience, but with the right ratio of pleasure to pain total disasters can be made blue moon events.