Last Sunday I ran the Mike Groves 10k race, hosted by Coltishall Jaguars Running Club at RAF Coltishall. The 3rd race in the 2023 Sportlink Grand Prix series of races.
On leaving home, my car indicated that the outside temperature was 8 degrees celcius. However, during the journey it steadily dropped and upon reaching the event it had fallen to 3 degrees. A little chilly for a T-shirt and shorts run, but at least it wasn’t raining!
But then, the heavens opened, and it absolutely poured down. I left things as long as I could before dashing through the rain to get to the registration area to collect my race number. Whilst torrential, the downpour was relatively brief. And as I walked over to make a last minute toilet stop, the rain ceased, and a distinct lack of wind, made the 3 degrees feel less chilly.
The conditions were looking favourable for a good performance.
My Race
For those that have not been following my articles, my target for this race was to achieve a 10 km personal best time of 41:30. A time indicated possible by TrainAsONE’s AI prediction algorithms. So a 4:09 min/km pace was firmly in my mind.
I edged myself close to the front of the start. Far closer than I would normally position myself, but I was keen to have as few fellow runners to navigate around in the starting straight as possible. Valuable seconds could be, well, valuable today!
We were soon off, and our 1st lap of the airfield was underway. I was firmly focussed on trying to achieve an evenly paced race.
I hit the first km marker at 4:14. Being 5 seconds behind plan, this was a good start, and I knew that with a little extra effort I could get on schedule by km 2. My little extra effort was obviously a little more than planned, as I ran past the 2nd km marker at a 8:08. I had jumped from 5 seconds behind to 10 seconds up. A correction to my pacing and 3 km was made with just under 4 seconds to spare. The 4th km was ran exactly at my target pace of 4:09 min/km, however during the 5th I found my mind wandering a little and my paced had dropped under target. Before I knew it the half-way point was upon me. I glanced at my watch. 20:45. I was halfway through, and exactly on schedule.
Halfway through, and exactly on pace.
I needed to focus. I knew that letting my mind drift would loose all important seconds. So by km 6 I reversed the temporary blip, being back exactly on pace, and with 4 km left to go, felt in a good place. At around 200 metres before the 7 km marker a small group of runners who had been steadily catching me drew alongside. The natural tendency to fall into their synchronisation took hold, and I calculated that I was 3 seconds ahead at the marker.
During the 8th km, the group very slowly began to pull away from me but I did my utmost to stay close on their heels. The 8 km marker seemed an age away. By the time I passed it, my mental maths was beginning to fail me. I estimated that I was roughly 10 seconds ahead.
Time to forget the maths and just run. Run hard!
2 kms to go! Only 500 metres and that’s a 6 minute assessment! Time to forget the maths and just run. Run hard! And run I did. So much so that I hit 9 km at 37 minutes. 37 minutes! And I was not slowing!
A new pair of runners began to creep up from behind, but I was just keeping ahead of them. At 500 metres to the finish they began to kick, but my legs had nothing left, and they accelerated passed me. Whilst I began to run through treacle, wading through the finish line in 41:15.
Analysis
My official time was 41:12. Sitting here writing this article 2 days after the race, I still cannot quite believe that I managed to run 10 km at a 4:07 min/km pace. That’s over 3 and a half minutes faster than my previous recent best which was only in November of last year. And nearly a minute faster than my all time best around 20 years ago!
Algorithm | Predicted (pace) | Predicted (time) | Error (seconds) | Error (percent) |
Riegel (3.2 km Assessment) | 4:25 | 44:10 | 178 | -7.0 % |
Riegel (6 min Assessment) | 4:20 | 43:21 | 129 | -3.9 % |
Riegel (Parkrun) | 4:17 | 42:57 | 105 | -4.1 % |
Riegel (10 mile Race) | 4:10 | 41:40 | 28 | -1.3 % |
TrainAsONE | 4:09 | 41:32 | 20 | -0.9% |
As can be seen in the box-plots above, I exceeded all the ‘most likely’ predicted expectations, and came within the 3rd quartile for the 6 min Assessment, 10 mile Race, and TrainAsONE calculations. It is worth pointing out, that whilst TrainAsONE was the most accurate in this case, the Riegel (10 mile Race) was still only 1.3% out.
On looking further into the prediction data, an aspect that the box-plots do not illustrate is that TrainAsONE’s predictions are skewed towards under-estimating performance. Consequently, it would appear that I had a 56% chance of hitting my 41:12 time (or faster). Whereas the ‘Reigel (10 mile Race)’ prediction errors are more uniformly distributed, and indicated only a 30% chance of achieving my time of 41:12. Interesting statistics, that provide more context to a prediction than the simple value and box-plot. Surfacing such information within TrainAsONE is on the roadmap, and I believe will prove to be very useful.
Parting Thoughts
The course and conditions were favourable on the day. On another course (i.e. a normal road route), and another day (i.e. warmer and/or with wind) I’m sure one could have added at least 15 seconds on to my performance. Making the predictions even more accurate.
What doesn’t kill me, makes me stronger.
Friedrich Nietzsche, 1888
It may nearly have killed me, but it didn’t. So do I now dare to dream of going sub 41… (according to the stats, I have a 50 : 50 chance!)
Thank you to all involved in putting on a great event. Despite the course being expectedly uneventful, with only some sheep and banks of solar panels for any real distraction from the pain of racing, I had a thoroughly enjoyable time.
Till next time.