If you have been following my recent articles, you’ll know that I ran two races on consecutive weekends. Not something I have ever done before, and from a peak performance perspective, not something that logic would recommend.
Like many things in life, this led to the often asked ‘what if’ question:
“If I had not run the first race, how would I have performed on the second?”
To answer this, I was able to use the power of TrainAsONE’s unique AI platform to re-write history and erase the first race (the Reedham 10 mile) and predict what my performance would have been for the second race (the Valentine 10k).
TrainAsONE’s prediction for the 2nd race (Valentine 10k), that considered all my training including the Reedham 10 mile (1st race) was a time of 42:30 (4:15 min/km). However, I exceeded this and completed the 2nd race in in a time of 42:15 (4:14 min/km pace). An difference of just 0.6%.
My approach to this time travel adventure was simple. With an up to date copy of my TrainAsONE account on my development environment, just delete the recording of my Reedham 10 mile race and then ask TrainAsONE for a retrospective prediction of a 10 km road race on the 12th February (the date of the Valentine 10k).
Like many things in life, things were not that simple…
As outlined my first scenario involved the simple deletion of my Reedham 10 mile race recording. TrainAsONE’s prediction….
44 minutes (on the nose)
Being slower than originally predicted, and that I actually ran, this did not seem right. I believe this was demonstrating a facet of the scenario that I had at the back of my mind. A facet which demonstrates the power of TrainAsONE.
A facet which demonstrates the power of TrainAsONE
My hypothesis being that TrainAsONE was looking at my running in the week between the events, and the sudden change (reduction) in my performance during that week (now unexplained as it had no knowledge of the deleted 1st race), led it to this drop in predicted time.
So I thought. What if I additionally remove all my training between the two races. It would be like having 7 days of complete rest before the 10 km. What would happen then? Using this approach, TrainAsONE gave me a prediction of:
42:40 (4:16 min/km pace).
That seems okay to me. It’s quicker than predicted for scenario #1, and it seems perfectly reasonable that a week of complete rest might result in a marginal loss of fitness.
Could we do better?
My final scenario to replicate was “what if instead of those now 7 days of complete rest, I mimicked a week of taper?” To do this precisely would be difficult, but I hoped that a good enough approximation would be to just repeat my previous 7 days of training (i.e. the final week of taper for the 1st race).
Using this approach, TrainAsONE predicted a time of:
41:30 (4:09 min/km pace)
Now I like that. Who wouldn’t – it would be a significant personal best for me!
So in conclusion it would appear that the effect of the 10 mile race 7 days prior to the 10 km race added 45 seconds onto my race time. And running on fresh legs, I would have been capable of a significant personal best.
Luckily (I think) for me, I have another 10 km race (the Mike Groves 10k) in just 11 days time. Naturally, I’ll get TrainAsONE to re-calculate it’s predictions (with all my data!) for this. One would surmise the prediction will be close (or faster) than 41:30, and fingers crossed a nice PB is on the cards…
I’d love to hear your thoughts. Please feel free to leave comments below.
Till next time, happy training.
This is very exciting. Are you planning to put a few training options into TrainAsONE over the next 11 days to minimise your race time the most?
Yes, I think this gives a glimpse into the future of training, and so very exciting. I’m using the next generation of our training algorithm, so yes I do have it configured to maximise my potential. Though regrettably, due to work and consequent lack of sleep my training this last week has not been quite the quality I would have liked (I have not missed any training, but have had to reduce some runs). I’m certainly looking forward to the next race.