Baseball is a game of statistics. Detailed statistics are kept on every player and are widely available (from the web to backs of collectable baseball cards). There are so many different statistics that it is not clear what is the "best" way to measure the ability of a player or a team.

For batters, for example, batting average, slugging percentage and on-base percentage are all recorded. (How to compute these statistics is available in many places, for example, an Alaskan softball league gives formulas along with information about snowshoe softball.

You will need to find statistics for starting players of major league teams. Remember that this is hard, major league general managers and owners try very hard to answer exactly these questions. But fans have been smarter than owners since the game began, so give it a try.

1. Which of the batting statistics for a team, average batting average, average slugging percentage or average on-base percentage, is the best predictor of a team's success (say in number of games won over a season). How can you tell which statistic is a better predictor?
2. Is it better to have a few very good (high average) hitters and the rest mediocre or have all moderately good hitters? (That is, if you use only the averages of the 4 best batters on a team, does the predictive power of the batting statistics get better or worse?)
3. What about pitching statistics? Is ERA or strikes/balls ratio a better measure of pitching performance?
4. Which is more important, hitting or pitching?

The key to answering the questions above is deciding which statistic more accurately reflects the outcome of games. Radio announcers and the people that call them on the phone argue endlessly about what is more important. Few of them have more than a "feeling" about how to judge the which statitics are more reliable predictors. Based on your work above, come up with a quantitative criterion which allows you to say one statistic is better than another.

Communicated by G. Hall.