Winning Is Like Coming Out of the Rain

t rains an average of 14 hours per day in far-away Soggyland. Occasionally, the rain is steady and persistent. Frequently, it's on and off. There are sprinkles; there are torrents. Unique local land, sea, and air conditions make the rain in Soggyland a nearly random process. Even short-term projections are unreliable.

Some Soggylanders have little choice about being outdoors. Maybe they work as letter carriers, or the podiatrist promised to squeeze them in right at 2:00 pm, or the dog won't take "no" for an answer. A few suffer cabin fever and lack the restraint to outride the rain. Over a long period, these civilians get wet about 14/24 or 58.3 percent of the time.

Other Soggylanders can better master their fates. Perhaps they retired, or joined a health plan with a walk-in clinic, or have a hamster, or are well-disciplined. The wisest such folks act on what they see outside and what 900-WEATHER forecasts for the next few hours. Often, they emerge only when the sky is clear. Sometimes, on especially dreary days, they try to decrease their unavoidable dampenings by dashing out when the deluge drops to a drizzle, figuring the weather will worsen if they wait.

Now and then, the best storm strategists get caught with their umbrellas down. They leave for the links in the sun; the clouds burst on the eighth hole. Or, just before dinner, they find they're out of Stove Top brand stuffing mix and run to the store, downpour notwithstanding, rather than endure a family fuss about potatoes.

Sure, folks with say over their schedules will get sodden once in a while. But presumably less than 58.3 percent of the time.

No secret system separates drys from wets. The drys aren't just lucky and don't regulate rain or play precipitation hunches. They base decisions on current conditions. Soggyland weather is still a random process. They use information to reduce the effect of the randomness on their own chances of getting drenched.

This is what information theory is all about. Applying data to overcome uncertainty, to reduce randomness. Information theory is the branch of science that gave the world computers, every sort of communication from the global telephone network to a single cable that brings hundreds of channels into your TV set, the tools with which biologists search for the genes that make us susceptible to certain maladies.

Since this column is about casino gambling, no great leap of imagination should be needed to get the point. Games of chance are random processes. Solid citizens can't control how cards fall, wheels spin, dice land, or balls bounce. Nor can they divine next results from trends they dream they discerned in the past. But they can choose when to cash-in based on the present state of their playing stakes.

Smart players can quit when they get ahead and begin to slip back, or cross from the loss to the win side, or make a significant come-back even though they're still in the hole. In doing so they employ information to reduce the effect of the randomness in the games. Sometimes, they'll take a bath anyway. Like Soggylanders who stroll to the strand under the stars and get soaked when a sudden storm gathers, they'll start confidently but soon drop behind and hit their fiscal or emotional loss limits. But, ultimately, players who understand how to exploit the element of advantage inherent in current information should do better than those who depend entirely on chance.

Or, should they? I'll stop here. Otherwise, we may find ourselves arguing whether multiple sessions in a casino are mathematically the same as one continuous game, and if the answer depends on why players quit when they do.

As Sumner A Ingmark, the Thoreau of information theory, intoned:

The fish all swim in pathways random,
No mystic patterns to command 'em,