Predicting Performance

Tuesday, March 11, 2014 | By Hal Higdon
Email this article

Have a question about running? You're in the right place. Every Tuesday, world-renowned coach, author and athlete Hal Higdon posts and answers athlete questions here. You can submit your question by joining the discussions on Hal Higdon's Virtual Training Bulletin Boards.


I am using your Novice 2 program with bits of Boston Bound put in to help with the hills. Today I had scheduled a half marathon race, but sadly it got cancelled two minutes from start time due to power lines coming down. So I went home and ran my own half—a 13-mile workout--but it was just not then same. I had planned to use the half as a predictor for Boston, but I could not motivate myself to run 13 miles as fast in a workout compared to in a race. Unfortunately, there are no other convenient halfs that would fit into my training plan. Do you have any suggestions on how I can come up with a race pace for the Boston Marathon?


The various prediction engines work best if you feed them a time from a race distance as close to the goal distance as possible. Thus, given their availability, the half marathon has provided us with an effective predictor distance. I agree that it is difficult to motivate yourself doing a "time trial" half marathon vs. a real half marathon. There’s something about pinning a number on your singlet that motivates you to run faster, to hit 100 percent, to achieve a race time that translates to other distances—and that includes downward in distance too. But time-trialing is better than nothing. And maybe you will arrive at a conservative prediction, which is not always a bad idea given all the things that can happen in the closing miles if you go out to fast. The other option would be to do a race at a different distance. Even a 5-K or a 10-K time may allow yourself some insight into your current level of fitness.

Get the latest news

Join Us