Based on some theory, some experience, and some modeling, I've come up with a simple formula to scale running results and allow a fair comparison among people of diverse ages, sizes, and shapes:

age + body mass index - 50 + 10 (if you're female) |

The result is a percentage adjustment. Take that much off your time in a race, and you've got your scaled result. For example, with a BMI of 24 and an age of 51, my current handicap is 24+51-50 = 25%. So when I run a 10 minute mile I get to subtract 25% from my finishing time, or equivalently add 25% to my speed. It's like doing a 7:30 mile when I was half this age --- which makes sense and feels about right.

The logic, in brief:

**age**hurts performance by ~1% per year, as is widely recognized**body mass index**is a standard measure of obesity or lack thereof. To compute BMI take your weight in kilograms and divide by the square of your height in meters (or use one of the widely-available tables or online calculators). Since each pound of excess weight traditionally is thought to slow a runner down by ~2 seconds per mile, converting the units shows that a 1 point change in BMI means ~1% penalty at a typical 7:30 minutes/mile pace**-50**is simply a linear shift in the zero point of the scale, a reasonable choice so that most people in their mid-twenties end up with near-zero handicap factors**+10**for females corrects for the normal difference in body fat percentage and strength between the sexes

One certainly could design a much more complex equation, and likewise could quibble over various of the terms in the formula --- but **age**+**BMI**-**50**+**10** *(if female)* is probably about as accurate as the uncertainties in the input factors allow, and it has the singular advantage of being trivial enough to compute, just barely, during the final stages of exhaustion during a long run.

*(see also NeedForSpeed (10 Aug 2002), LogbookTyrannicide (17 Oct 2002), ... )*

TopicRunning - TopicScience - 2003-10-08

*(correlates: AppearVersusIs, SpeedUpSlowDown, At My Pace, ...)*