Looking back at Heart Rate vs Pace: A Year Later
Heart rate in a single moment is not a great measure of running performance. But using averages and multiple days over the course of a 2-3 week time period, and it can be useful information. It becomes even more powerful when the heart rate is evaluated under similar weather conditions. That's why now a year later, I've got data to show improvement in a one year timespan from a Heart Rate vs Pace standpoint.
I find that my heart rate generally matches my perceived effort quite well. If I'm running Marathon Tempo, then my heart rate is generally 148-152 beats per minute. I'm doing Long Run pace (MP + 9%) and I'm around 138-142. Easy running is around 137 or less. HM Tempo or faster and I'm usually in the mid-150s to just inside the 160s. It's always been this way since I started tracking heart rate data in 2014. What's changed in that timeframe is how fast I am running with that relative heart rate.
The black line and yellow squares represents December 2016 data. The red line and red triangles represents December 2017 data. So how can it be interpreted?
-In 2016, a HR of 138 (which is the low side of long run pace) was a 8:09-8:16 min/mile. In 2017, that same HR of 138 is now a pace of a 7:25-7:33 min/mile. Same effort, but now one year later it is ~44 seconds per mile faster.
-In 2016, a HR of 150-153 (which is right near the middle of marathon tempo) was a 6:55-7:19. In 2017, that same HR of 150-153 is now a pace of 6:31-6:59. Same effort, but now one year later it is ~18-24 seconds faster.
-In 2016, a pace of a 6:55 min/mile was a HR of 152. In 2017, that same pace of a 6:55 min/mile was a HR of 140-147. That moved from high end marathon tempo HR to low end marathon tempo HR OR high end long run pace. If that 6:52 min/mile pace at a HR of 140 is real, then compared to the same HR of 138-140 in 2016 (8:09) it is a 77 second improvement at the same relative effort level.
So using the logarithmic curve and known HR to race distance relationships, we can try and estimate peak performances. Interestingly, this correlates quite well with the "race predictor" feature of my Garmin HR monitor based on my current VO2max.
While the projections may not be entirely accurate. They can show the level of perceived improvement from one logarithmic curve to the other.
So, in a year's time my race theoretical performances "improved" by the following:
5k- 19:51 to 17:48 (10.3%)
10k- 41:20 to 37:12 (10.0%)
HM- 1:29:06 to 1:20:20 (9.9%)
M- 3:05:19 to 2:47:39 (9.5%)
So a pretty consistent level of improvement across the curve generated by this data set. How would this look per Dopey 2017 5k/10k times if I took my time from last year and "improved" it by ~10% for 2018?
21:02 to projected 2018 time of 18:52 (5 sec faster than Red CF 2016 Dopey Model from my prediction explanation)
43:25 to projected 2018 time of 39:04 (1 sec slower than my CF estimate at the current moment from my prediction explanation)
Interestingly, none of the predicted data set model for my guesses had much to do with HR data. It was based on historical and perceived effort. So, it's quite neat to see these two separate models converge so closely. It makes it that much more interesting to see what the final times will be.
Can't do the HM because it didn't happen in 2017 and the marathon in 2018 should have a HM proceeding it which influences that data point. Without the HM, the M would be:
3:20:52 to projected 2018 time of 3:01:47
All in all an interesting look back at data from a year ago compared to now.
