Marathon Madness

When Robert Lawson of Division of Labour (one of my favorite econ-blogs, by the way) recently participated in a marathon in Ohio, he was possessed to perform a regression analysis using data from the previous year's race, and posted the resulting formula derived from that race's data in "The Laffer Curve of Life". Using this data, Robert was able to demonstrate a positive correlation between a runner's age, gender and residency (in-state vs. out-of-state for where the marathon is held) in their average time to complete the race.

Now, ordinarily I would let this go, but Robert may have kicked off a trend in the blogosphere, since a reader of his original post was inspired enough to perform the same exercise for the 2004 Chicago Marathon. Not being one to buck a trend, if multivariate marathon regression analysis is one, Political Calculations^(TM) is ready to leap in with a tool to do the math in the resulting formulas. Enter the appropriate data in the table below, and click the "Just Do It" button for the estimate of average time results:

Runner Data
Input Data	Values
Age (Years)
Gender	Female Male
Residency (Relative to Location of Race)	In-state Out-of-state

Estimated Average Time Results
Marathon	Time (Minutes)
2004 Towpath Marathon (Ohio)
2004 Chicago Marathon (Illinois)

While modeling marathon race results this way may not be the most useful of pastimes, the exercise undertaken by Robert Lawson does underscore that it's possible to analyze reams of data and to extract meaningful relationships from it. These relationships, described as mathematical formulas, may then be used to estimate results in the future or to fill in missing segments of data, should there be any. Properly done, regression analysis is a powerful tool for representing how certain individual factors may influence overall results.