For as long as I can remember, I have enjoyed “predictor workouts” in running. I recognize they have their shortcomings, and are not a guarantee of race days success. But, it is often fair to say “hey, if I can do this, I ought to be able to do that.”
One of my favorites (as I have written here before) was 6 x 800 on a minute rest, 3-4 seconds per 800 than 5k race pace. That seemed to work well for me. I also liked 3 x 800 at mile pace on a 3 minutes rest to predict mile performance but that never played out in a race for me as well as it did in the workout (maybe I left the race in the workout).
I got dozens more, including even guesses at what a run up the SAG route on Green Mountain in Boulder does for predicting a Pikes Ascent time. I love them. So when I recently heard elite miler and 2x Olympic medalist and miler Nick Willis provide a predictor of sorts for the mile, I got really interested.
In response to one of the questions on a podcast, Nick mentioned to be a 4 minute miler you needed to be able to run the 100 on a rolling start in 12 or better and a 4 mile tempo run in 20 minutes or better. I found this pretty thoughtful. A strong mile race requires both speed and endurance and these tests that Nick put forth were hitting right on that.
His idea reminded me of a discussion I have had with Lucho and Bob over the years about how to prepare for a distance race like a 5k. When jumping into runs with Lucho, he’d often want to do diagonals on a soccer field. That is sprint the diagonal (corner to corner of a soccer field) and take the hard turn on the end zone to jog that distance to the next corner. Do that 20 times and you’ll get pretty ripped up of you are not used it and Lucho thought it was good for helping him get in shape. Conversely Bob thought this sort of work did little to prepare for a 5k race and that this sort of work was not necessary until you demonstrated some ability to hold a good pace over a 20 minute tempo run. Where was it best to spend the training “buck?”


So who was right? Both, neither and “it depends.” The first “it depends” is it depends on what you are a responder to (and probably what you enjoy). Another “it depends” is it depends on what you are weak at. Those could be the same thing … or not. I have felt that while it depends on the athlete, a blend of training both of those ends is necessary to get you best prepared for a goal race – speed and endurance (in fairness to those guys, they do to, and the discussion was probably more a debate about what we wanted to do that day to play to our strengths so that we could hurt the other guy a bit more … Lucho would crush on short stuff, Bob would vampire grind us on long stuff … as long as it was not too hot).
Willis’ proposal seems to hit on both ends as well. I asked BrHS coach Greg Weich what he thought and I could see he was quickly going a review of the athletes he has known over the years in his head. The answer: it depends. There were guys who were speed guys that could go 50 in the quarter (and thus likely faster than 12 on a rolling 100) but would not be able to go sub 20 in a 4 mile tempo run. He knew of guys who’d go 19 on a tempo run but be lucky to run 13 in the 100. These were guys really close to the 4 minute mile so it seemed that Willis equation was in the ball park – or like a lot predictors, it worked for a lot of people some of the time but not all of the people all the time. The old George Box adage is “All models are wrong but some are useful.”
The Willis pitch got me thinking if his prediction points could be extrapolated for a 5 minute mile – or a goal that I have been thinking about. A 12 second 100 is 25 percent faster than the 15 second 100 that a runner needs to hold for a 4 minute mile (yes, I know that a mile is a bit longer than 1600 meters but for the simplicity of this work, go with that for now). A five minute mile pace for four miles is 80 percent the speed one needs to hold for a four minute mile.
Willis did not describe what he meant by a tempo run in the interview. Very classically, I think of a tempo run as an effort that you’d be able to hold for an hour – as in it is your race pace for 1 hour but usually are only doing for 20-40 minutes to get stimulus without a full on race effort. I am not sure if Willis mean that he expects 4 minute milers can also run a 12 mile effort in an hour race (or a 67 half marathon). In some regards, that seems appropriate … but that would be a stretch for some as well (it depends!).
More to the point here I think it is important to note that this is a tempo run for 4 miles – NOT an all out race effort.
In regards to what it means for a 5 minute mile effort – since a runner holds 18.75 per 100 there (or slightly faster because of the 1609 meters of a mile over 1600), 25 percent faster than that is 15 seconds. And following the 80 percent modeling for the tempo run, that would be a 6:15 pace or 25 minutes for the tempo run (check my math, I might have messed it up).
And now the fun of the predictor: can I do that and what do I need to improve (later followed up by the question – if I have done that can I do what I need to for the goal race)? A 15 second 100 is probably a bit out of reach for me at the moment. I might be able to do it but it would be very risky (injury). It seems do-able with some work, but that might just be an overactive imagination of an aging man.
On the other end, a 6:15 paced four mile tempo run … I think I could do this but today that would be well above tempo range. It is probably a bit above my abilities straight up for four miles (based on recent 5k times) but I think I could hit it … but hold that pace for an hour? Yeah, not likely. 6:30? Maybe. 6:15? Not right now.
Anyway … I need work at both ends – the strength and the endurance. And I don’t know if I am going to be able to get to that sub 5, but the Willis thought experiment has me chewing on how to get there a bit more.
(PS – I did not ask Bob or Lucho their permission on any of this so I am probably horribly misrepresenting them)