What I learned from watching baseball's Division Series...

Game 1 is the key game, especially in a short series of five games. Did you know that 21 of the 32 Division Series contested since 1995 have been won by the team that won the first game?* That underscores the importance of setting the tone for the series by jumping out in front of your opponent and getting momentum on your side.

The importance of winning Game 2 cannot be underestimated, especially in a short series. If you lost the first game, it is crucial not to go down 2-0 and put yourself in a position to have to win three straight, in particular if you opened the series at home. If you won the first game, winning Game 2 really establishes momentum and puts you in the driver's seat to win the series early.

Game 3 is always a critical game, especially in a short series. If you're down 2-0, of course, Game 3 is a must-win game to stave off elimination. If you're up 2-0, you need to pound the nail in the coffin so you can finish the series early and put your pitching rotation in order. If the series is tied 1-1, then Game 3 is the turning point -- whoever wins seizes the momentum and forces the other team to the brink of elimination.

Game 4? One word: vital, especially in a short series where one team must be on the verge of elimination. If you have the upper hand, it's time to finish off your opponent and give your pitching staff that all-important extra day of rest. If you're down 2-1, winning Game 4 not only keeps you alive, but it puts the momentum back in your clubhouse for the final game.

In a short series, Game 5 is, of course, the decisive contest. Winner moves on, loser goes home. Game 5 is for all the marbles, so forget about momentum, forget about regular-season records, and forget about your pitcher taking the mound on three days' rest. This is October baseball, when leaders emerge and champions prove their mettle.

---

*BONUS PROBABILITY FEATURE!

During last Saturday's A's-Red Sox game, the announcers put a graphic on the screen showing that, up to that point, teams that won Game 1 of a five-game Division Series had won that series 21 of 32 times. This statistic floored the announcers, who discussed it with reverence and awe, and seemed to agree that to win Game 1 confers powerful psychological advantages on the victorious team, as well as giving them precious momentum. (Momentum is a magic substance secreted by faeries when a team wins a game. Smart teams keep their momentum in an enchanted box, crafted by the elves of Mfriggrynddor, to be used at a critical juncture later in the series. Strangely, when a team loses a game, all of its momentum disappears.)

Anyway, should we be surprised that the team that wins the first game has won 21 of 32 Division Series? (Let's try to forget for a moment that this year's Game 1 winners lost 3 of 4 Division Series.) Let's suppose that Team A is playing Team B, and Team A wins Game 1. Let's further suppose that Team A and Team B are evenly matched; that is, that the probability is 50% that Team A will beat Team B, (or vice versa), in any given game. There are only a few ways the series can unfold.

Team A wins series:
AAA (probability = 50% x 50% = 25%)
ABAA (probability = 50% x 50% x 50% = 12.5%)
AABA (probability = 50% x 50% x 50% = 12.5%)
ABBAA (probability = 50% x 50% x 50% x 50% = 6.25%)
ABABA (probability = 50% x 50% x 50% x 50% = 6.25%)
AABBA (probability = 50% x 50% x 50% x 50% = 6.25%)
Total probability that Team A wins series: 68.75%
=============================================
Team B wins series:
ABBB (probability = 50% x 50% x 50% = 12.5%)
AABBB (probability = 50% x 50% x 50% x 50% = 6.25%)
ABABB (probability = 50% x 50% x 50% x 50% = 6.25%)
ABBAB (probability = 50% x 50% x 50% x 50% = 6.25%)
Total probability that Team B wins series: 31.25%

Out of 32 Division Series, then, we would expect that the team winning the first game would win 22 ( 32 x 68.75% ) of those series. Game 1 winners have thus done worse historically than we should expect, especially when we consider that the team that wins the first game is probably the better team. This means that it should stand a better-than-50% chance of winning any given game, which would skew its expected winning percentage upward.

+posted by Lawrence @ 10/07/2003 06:02:00 PM