'
(PDF)
Game Theory and Predictioneering's Roots
As a discipline within math, game theory reflects calculated circumstances,
also called games, where a person's success is based upon the choices of
others. It is mainly used in economics, political science, and psychology.
While at first designed to investigate contests where an individual does
better at the cost of another player, also called "zero-sum" games, game
theory applies a wide range of class relations, and has developed into an
umbrella term for the logical side of science, to include both human and
non-human decision theory.
Classic uses include a sense of balance (called
an "equilibrium") in numerous games, where each person has found or
developed a tactic that cannot successfully better his results, given the
other person's approach.
Components of game theory can predict how a given player should make decisions; in other words, how to construct a strategy (a set of decisions based on what decisions other players make so that precisely following this set of decisions will result in the best possible outcome for the player). Game theory can also predict biological and political equilibriums.
The Origin of Predictioneering
Predictioneering, as originally developed by Bruce Bueno de Mesquita, works first by leveraging the core game theory concept of self-interest to predict human decision-making. The theory and application assumes human decision- makers will always act in their self-interest. Given this assumption, we can look forward and reason backward through a series of decision steps or iterations to understand what any given player within a game simulation will choose to do.
Let's say there are two companies, A and B, who are trying to buy company C. For simplicity's sake, let's assume there are only 3 players: the boards of companies A, B, and C. Let's further assume price is the only metric to negotiate. And let's additionally assume there is only a single round of sealed bidding, meaning companies A and B must present to C their bids in sealed envelopes. In this overly simplified example, we can easily establish C's best strategy: always pick the highest bidder between A and B. A and B's strategy is only minimally more complex: strive to predict the other's price and make a purchase offer fractionally higher based on the degree of uncertainty about the price prediction. In such a scenario, it's trivial for the human brain to comprehend every possibility, since there are only two: either A's offer is higher than B's, or the inverse. (Technically, there's an unlikely but at least mathematically plausible option that both offers are the same.)
