Keeping Quality Run Data

It may be because I'm an engineer, or it may be because I just like tech stuff, the fact remains that I have logged my runs in one form or another for the 3+ years of my running career. While I don't mind running "naked" (without GPS), I don't want to run without obtaining the data that it collects. I upload my data to the Garmin Training Center on my PC and also to my favorite logging program at RunningAHEAD.com.

I found that there are three bits of data that are absolutely the most important: 1) Distance, 2) Pace, and 3) Temperature. Over the last few months, I am now convinced that a fourth bit of data is almost as vital: 4) Heart Rate.

So, let's suppose that I had never recorded the temperature. On a typical mid distance run, I might run at an average pace of 9:00/mile. Then, a few weeks later, I run the same course at 8:45/mile. The first instinct is to pat myself on the back and say "Wow, I'm really getting faster". While that may be the case, let's revisit this scenario again. Now assume I also recorded the temperature for my runs. In the first run at 9:00/mile pace, the temperature was 80°F. In the second run at 8:45/mile pace, the temperature was 70°F. Could that cooler temperature alone have made the difference in my pace?

We can spend just a few seconds of search on Google and find many hits on the effects of temperature and pace. While it would be nice if the data were the same for all runners, it doesn't take long to find that we are all a little different. Age, level of fitness, and so forth, are all variables in this equation. In my case, I logged enough data from my runs over long periods of time that I made this general observation that is applicable only to myself: For all other conditions being equal (fitness level, course, distance, etc.), I found that a 10°F change in temperature resulted in a 15 sec/mile difference in pace. I found this generally true from 50°F up to around 90°F with it skewing quite a bit beyond that.

While running these summer months in the heat, I pay very close attention to this temperature variable. Just yesterday, I ran a strong "race pace" run at 104° for 3 miles. My pace was 8:22/mile. While that's a bit beyond my range of quality data for my "15 seconds per 10°" rule, it still would suggest something like the following: Suppose the temperature were 84° instead of 104°. That's a 20 degree difference. It would suggest that I could have run the 3 miles at a 30 second quicker pace, or 7:52/mile. Okay, only I would know that's a little too fast for me at a still warm 84°, but the trend should be very apparent. Perhaps it would be closer to 7:59/mile. In any case, the emphasis is there is some significant effect of temperature changes, more than many of us may realize.

I have looked back at a lot of data and found that as long as I didn't try to stretch it too far, the 15 seconds per 10° rule fits very well for me. It also seems to work very well regardless of the distance. With that information, I take great comfort in knowing that a long run in the summer at 80° may be run with the same effort but 30 sec/mile faster for a cooler temperature run of 60°. That alone could make about a 14 minute difference in a marathon finish time.

The point of all this is not to reinvent the wheel for something that is fairly well talked over. Rather, the point is to encourage each of us to find what the effect is for ourselves. Collect that information in your daily run logs, then look for trends. Filter out obvious variables like your running fitness, terrain, etc. Do it smartly, and you will find something that is fairly accurate for you. Take comfort in knowing that the same effort you put forth on a warm summer morning run may actually be run at the same effort but 30 sec/mile faster on a cool fall morning.

Next time, I hope to make a few similar observations about heart rate data.