What does TrainAsONE predict my time will be for tomorrow’s 20 mile race?
It’s race day again tomorrow, and this time it’s a long one! The Wymondham AC 20 Miler, a 20 mile road race hosted by Wymondham Athletic Club. The race starts and finishes in the town of Wymondham (Norfolk, England), with the bulk of the course being along country roads and taking in a number of the surrounding villages. Being located in Norfolk, it is an undulating but fairly flat route profile, though I have been warned that there is a hill towards the very end.
This is the 4th race of the 2023 Sportlink Grand Prix, so it will no doubt be a well attended event, with some fast runners in the field.
Since my casual Blickling Half Marathon last Sunday, my training has been very nondescript, with no speed work and only easy economy runs. This would have been a result of both running the half and the lingering effects of my previous flu episode necessitating a good amount of rest in order to be as prepared as possible for tomorrow’s outing.
Racing any further than a half fills me with dread.
Whilst I have ran a couple of road marathons in the (distant) past, I have never run a 20 mile race before. So this will be a new experience for me. And being honest, it’s been so long since running a marathon that the thought of racing any further than a half fills me with dread.
I’m sure (know) I will be fine, but doubt that final 10km will be pretty. I’m just hoping the excitement of the occasion will carry me through to complete a good time. So what time am I predicted to do? Let’s find out.
My Race Time Predictions
As with my previous few races this year we’ll compare how TrainAsONE’s unique AI race prediction compares to the estimate provided by the use of Riegel’s formula.
| Algorithm | Days Since Assessment / Race | Pace (min/km) | Time (mm:ss) |
| Riegel (3.2 km Assessment) | 18 | 4:59 | 2:40:34 |
| Riegel (6 min Assessment) | 32 | 4:41 | 2:30:55 |
| Riegel (10 mile Race) | 48 | 4:31 | 2:25:40 |
| Riegel (10 km Race) | 20 | 4:25 | 2:22:18 |
| TrainAsONE | n/a | 4:23 | 2:21:41 |
As depicted in the table above, the different algorithms estimate quite different outcomes for tomorrow’s race, from a slowest time of 2 hr 40 mins to a fastest of just under 2 hrs 22 mins.
History tells us that the 3.2 km assessment (which is predicting the slowest and most different time to the others) is probably safe to ignore. I was holding out more hope with the 3.2 km calculation for this race than previously, given that I have completed another 3.2 km assessment since the last race, and (for me) put in a good performance.
My most recent (best effort) race was the 10 km, where I managed a personal best. So one would suspect that this provides a good benchmark for tomorrow. This fits, as out of all the Riegel prediction formulae it is estimating the fastest time.
TrainAsONE, with a pace of 4:23 min/km and time of 2:21:41, is predicting a slightly faster, but not significantly different, performance to the Riegel 10km.
What is more interesting this time around, is the comparison of the box plots for the various predictions. In all cases the estimate for tomorrow is above the upper quartile. If the errors hold true for me, then one would calculate that my fastest time tomorrow (based on TrainAsONE) would be just inside a 2 hr 29 min time.
It would seem that these box plots (and with those of the previous shorter races in mind) nicely demonstrate that as race distance increases the average runner is less and less prepared for the requirements of fast endurance running.
From a personal viewpoint, it is nice to see that the TrainAsONE box plot is overall tighter than the Reigel variants.
Targeting around 2:22 seems mad
At the start of the year, when I first decided that I would enter the Wymondham 20, my thought was to be aiming for 2 hrs 30 mins. Targeting around 2:22 seems mad, especially as I just managed to race a 70 minute 10 miler only 6 weeks ago!
Let’s see that tomorrow brings… (I’ll keep you posted.)
