Hard science is the study of measurable, quantifiable, predictable results. A falling object will accelerate at the same calculable rate every time, and given quantities of oxygen and hydrogen will combust to form an exact amount of water every time. Soft science deals with ingredients that do not always react the same way no matter how hard you try to keep the situation uniform. The difference between these ingredients and those studied by hard science is that the loose cannons are life forms. Put five cats, two dogs, and a ten-year-old boy in a room, and you have no idea what’s going to happen. And if you put the same bunch in the same room again (assuming they all survived the first time), something different will happen.
Game theory started as a mathematical study of how certain types of games play out. But the math only worked in theory because it assumed that each player would play predictably—for his or her own maximum benefit in the most efficient way. However, human beings often, even usually, don’t play that way. Whim, stupidity, vengefulness, other emotions and motives, carelessness, and the whole range of human complication play a role in games large and small.
When hard math didn’t fit real life, game theory became more of a soft science that probed, with experiments, how people played games. There’s plenty of math involved, and astute game players of all kinds—military strategists, politicians, businesspeople, lawyers, spouses, lovers, parents, criminals—can improve their odds by studying it. This isn’t a math that says two plus two always equals four and parallel lines never meet. It’s a math that penetrates the house of mirrors that springs up when people in negotiation and competition second-, third-, and fourth-guess one another’s motives and plans. It describes what might happen, what should happen but probably won’t, how to avoid worst outcomes if not always how to attain the best, and how to plot the likeliest strategies—but rarely about what predictably will happen in any particular game.
We’re all used to the unpredictability of real life. I found it fascinating that there is, however, an underlying math that can be methodically applied to life games to improve the odds of success, reduce those of failure, and for the spectator, to understand better why people do what they do.
Davis generally writes clearly and organizes the material logically. I suppose by necessity for nontechnical readers, he didn’t explain what lay behind some of the calculations, and by the end it got rather complicated, but still, it was a short, engrossing read. One of his most interesting approaches was to begin each chapter with a set of real-world problems designed to get readers thinking and speculating. Often in the middle of the chapter I’d suddenly see what I had missed in my first stab at a problem. At the end of the chapter, he explains the solutions, and it was refreshing, in a math book, that sometimes there was no clear-cut answer but just guidelines to possibilities.
Game theory is no dry, arcane math but a realistic study of what people are likely to do in certain situations and how to predict and deal with their range of possible reactions.