Nick Berry spent ten years at Microsoft’s Casual Games division, and now is the president of DataGenetics, a data mining company based in Seattle. But in his spare time, he decodes the mysteries of board games like Battleship. Although the film adaptation had a lackluster opening this weekend, that hasn’t stopped Berry from reinvigorating the title.
Slate outlines Berry’s work and starts with a typical strategy called Hunt/Target. This is how most players play — make a shot and then fire nearby. But Berry divised something ingenious:
To improve upon the Hunt/Target method, Berry devised a tactic that combines Hunt mode with the concept of mathematical parity. Think of it this way: Imagine if the board were color-coded like a checkerboard, with white and blue squares. Even the smallest ship—the destroyer—covers two squares, and would therefore have to rest on both a white and a blue square. Fire only at blue squares and you will eventually hit every ship at least once. This method effectively allows you to reduce the number of targets on the board by half when you’re in Hunt mode. (When you register a hit you enter Target mode, and both blue and white squares are in play until you sink the ship.) This strategy yields a slightly better average than regular Hunt/Target mode: an average of 65 moves to sink your opponent’s fleet.
Berry then uses something called “probability density function,” which takes into consideration the different ways the ships can fit across the board. If you’re still dying to learn the nuances of Battleship from here on out, check out Berry’s site.