People are predictable even when they try not to be. Witness the "mind-reading machine" built by computer pioneer Claude Shannon in 1953. A human made a simple mental choice (to throw a switch "left" or "right") and the computer guessed which. When the computer guessed correctly, it won a point; when the human fooled the machine, it won a point. By spotting unconscious patterns in players' choices, the computer was able to outguess everyone. Today descendants of Shannon's mind-reading machine are all around us. They are the "big data" algorithms that use past consumer choices to predict what we can be persuaded to buy next.
In recent years, psychologists have learned much about the predictability of choices. The subject is anything but trivial: One way or another, we are all in the business of prediction. Anticipating thoughts and actions of others is crucial to winning an argument or a game, getting a date, a promotion, or a fortune. Success is often a matter of guessing a little more accurately than the next person.
Teachers try to make sure that a test's correct answers fall in random order. But most multiple-choice tests have patterns that are useful to those who have to guess on a tough question. On true-false tests, "true" answers are usually more common. The answer that was right on the previous question ("true" or [d], say) is less likely to be right on the current one. The biggest tip-off is "none of the above" answers. They are far more likely to be correct than they ought to be. Such answers can't just be dropped in; all the options have to be written around them. The effort discourages test-makers from supplying the strategic number of wrong "none of the above" answers. Such answers have about 50 percent chance of being correct when they occur on a 4- or 5-choice question.
Each month, three of Bernie Madoff's minions made up hundreds of fake numbers, representing trades that never happened, profits that were never realized. Scam artist try to invent "random-looking" numbers but almost always fall into unconscious patterns.
Here's a made-up number: $438,765.38. Notice anything unusual? Probably not, but it's got two telltale signs. One is descending sequences. The 4 is followed by a 3, the digit that's one less. Then 8 is followed by 7 followed by 6 and then 5! Such cascades are far more common in faked data than in honest numbers. Another tip-off is repeated sequences like 38--which appears twice. Madoff's alleged profits overused the digit-pair 86--and so did his self-reported golf scores.
A tennis player can serve to the right, left, or center. Studies show that a player who serves right this time is more likely to serve left the next time. The best players anticipate such patterns. The only countermeasure is to make sure your serve direction is truly random. One method is to use a wrist heart-rate monitor. Because your heart rate is always changing, the rightmost digit of a digital readout is for practical purposes random.
Maybe you want to serve to the right 40 percent of the time, to the left 40 percent, and to the body 20 percent. You might then decide that even digits 2, 4, 6, and 8 mean "serve right"; odd digits 3, 5, 7, and 9 mean "serve left"; and low digits 0 and 1 mean "serve to the body." Should the monitor read 167 when you glance at it, you'd serve to the left.
Poker players use "pupil reading." When someone sees an object of desire--a card needed to fill an inside straight, or even a sexy photo--the pupils enlarge. The effect is reliable enough to be a useful tell. Some serious players wear dark glasses as a countermeasure.
You can't turn around these days without being told to invent a new (secure!) password. The ideal is a completely random password. The problem is that mere humanoids have trouble remembering a password like EWsGR8Fx.
One bit of conventional advice is to convert a phrase to a password. "May the force be with you" would become "Mtfbwy". Of course, you wouldn't want to use that one.
A better idea is to turn this prescription on its head. Get a random password (from a website or browser) and convert it to an easy-to-remember phrase. Take EWsGR8Fx. That could suggest the phrase "Entertainment Weekly's Great Flix." Remember that phrase as a prompt for the truly secure password EWsGR8Fx.
The winners of most Oscar categories come as no surprise to anyone who consults online odds. Few players in office Oscar pools bother. Instead they play personal favorites. They might do better to bet their father's favorites. The median age of Academy voters is 62.
Simply by betting the favorites of online odds sites, you are on track to reap a profit. This strategy should be adjusted, however, when there are other high-information bettors in the pool. Then it pays to go mildly contrarian. In 2013 Internet odds makers slightly favored Tommy Lee Jones over Christoph Waltz for Best Supporting Actor (41 v. 39 percent). I went with Waltz. I knew I wasn't giving up much and suspected that the savvier players in the pool were backing front-runner Jones. Waltz turned out to be the right pick, and I looked like an Oscar genius.
Rock, Paper, Scissors
If you play rock paper scissors as an adult, it's as a way to decide who pays for drinks. When the occasion arises, say, "Let's do rock paper scissors for it!" Without waiting for an OK, begin pumping your fist. One... Two... Three...
This is the way the game begins; you show your choice on the count of three. So if you begin pumping, there's a good chance that your mark will join in. Then you throw paper.
Why? Novice players are prone to choose rock on a first throw, especially when they don't have time to think.
There is a safety net. Should you lose, immediately start pumping for the second throw. You're going for two of three, naturally...
William Poundstone is the author of Rock Breaks Scissors: A Practical Guide to Outguessing & Outwitting Almost Everybody, published by Little, Brown.