Over the past 2 years I have used an indicator that I came up with to monitor my fitness and predict my race times and have been surprised at how accurate it has been. Now I would like to put it out there for others to test to see if it works for the general population or just for me.
I call it feet-per-beat, or FPB, and, as the name implies, it measures the average number of feet I travel per beat of my heart during aerobic runs. The theory is that as fitness improves, you will get faster at the same HR, or run the same speed at a lower HR, both of which are reflected in a higher FPB. But do too much and the buildup of fatigue will cause your FPB to decline.
I will get to how I use it, but first I calculate it by taking the info from my Garmin after a run (time, distance and avg HR), and put it into an Excel spreadsheet to calculate and track the FPB. The equation is:
FPB = (Feet traveled) / (HRavg * Minutes)
Feet traveled is (miles * 5280), or if you are metrically inclined, (kilometers * 3280.8399096).
HRavg is what the watch reports as your total HR avg. I have noticed some glitches in my HR with my Garmin 225, but far fewer in my 935. If the glitch is fixed soon enough (take watch off and wipe sensor and wrist), the result is still usable.
Minutes is in decimal form so 23:12 is 23.2 and 45:36 is 45.6, etc.
Note that I only trust results from aerobic runs over 30 minutes. Shorter runs and runs with significant anaerobic effort tend to less reliable in the context of the FPB trend. Speaking of trend, the graph will be a bit erratic due to the nature of round-off errors and sensor inaccuracies so I apply a 5 day moving average so highlight the trend.
To predict a race, take the FPB of the last easy run before the race and plug it into the equation above, along with the race distance in feet and expected HR, and solve for Minutes. The closer your estimated HR is to reality, the better the prediction will be. I got my HR right in my last prediction, which was 1:34:05 for a half marathon. I actually ran 1:33:55, but I do not always get that close.
If you want to get more accurate, you can make adjustments based on course or weather differences between the last run and race day, but that is an exercise left to the reader.
There it is, now I would like to hear from people who apply this metric to their log books to see if its predictive powers holds true for them. It would be interesting to hear from someone who has followed several well-known training plans/philosophies to compare the FPB trend – was the success or failure of the plan reflected in the FPB trend? And, can this be used universally as a predictive tool?