Case Study: Using Race Times To Diagnose Your Running Weakness

This week, I saw a post on Reddit from a runner whose marathon didn’t go to plan.

He had run a 1:31 half marathon in training, and he based his marathon goal – 3:10 – on that race result. Things went well for the first half, but the wheels came off in the second half … and he finished in 3:42. The last few miles were a walk-jog, and his average pace for the last five miles was 13:00/mi.

At the end of his post, he offered – in this order – the following possible explanations for the blow up:

1. Terrible sleep during training
2. Life stress
3. Not consistent mileage
4. The weather
5. Lack of stretching and strength training
6. Not enough salt intake

That’s a lot of possible explanations, and most of them are likely red herrings.

If we take a closer look at the background he shared, it’s pretty obvious what the problem was. And the biggest clue is his full set of recent race results.

Let me add a little more context, and then I’ll explain how you can use recent race times to diagnose your strengths and weaknesses as a runner.

The Runner’s Background and Training

In his post, the runner shared a few key details about his past.

He’s currently in his mid-20’s, but he ran in high school and some in college. He didn’t run at a highly competitive level, but he developed a decent amount of speed. His old PR’s were 4:51 (Mile), 17:49 (5k), 1:27 (HM), and 3:39 (FM).

The 3:39 full marathon was done with little mileage – about 25 miles per week. He started out in 1:31 and then crashed in the second half (2:08). His earlier training was described as “always smashing zone 3 runs and plateauing.”

After a significant (5+ year) break, he got back into running and took six months to build up to 35 miles per week. He then followed the Pfitz 12/55 plan starting in August. He also had his first child in August, and as a result he only ran 4 days a week. His weekly mileage ranged from 16-44 miles, and he averaged 34.1mpw (excluding the two taper weeks).

This year, he has run 5:19 in the mile, 18:31 in the 5k, and 1:31 in a half marathon. These times led him to target 3:10 in the marathon.

So what went wrong …?

Comparing the Historic Race Times

The major clue here lies in his history of equivalent race times.

First, take a look at his PRs from his younger days – 4:51, 17:49, and 1:27.

A 4:51 mile is pretty fast. It’s not D1 fast, but for most recreational runners that would be quite the accomplishment. So clearly he has some speed, and he developed that fairly well through his early years in training.

But if you plug that time into a race equivalency calculator (like the VDot calculator), you would see that a 4:51 mile is roughly equivalent to a 16:41 5k. The problem is that the 5k distance is much more dependent on your aerobic base than the mile, and you can’t muscle your way through it. He didn’t have a strong base, and this is pretty clear from his actual 5k PR – 17.49.

Then, take another step out to the half marathon. A 4:51 mile would be equivalent to a 1:16:29 half marathon. Yowza, that’s fast! But his 1:27 is nowhere near that. If you used the 17:49 5k instead, that’s still equivalent to a 1:21:43 half marathon.

A 1:27 is a good time for a half – but it is slow for someone who can run a sub-18 5k. It’s a clear sign of aerobic underdevelopment, and that jives with the runner’s history of running low mileage at moderate to high intensity.

He developed his speed – but he neglected his aerobic base.

Comparing the Recent Race Times

So has anything changed with his recent return to running?

While he started out by slowly building up to a decent chunk of weekly mileage – 35-40mpw – he didn’t sustain that through the marathon training block. It’s quite likely that he revived some of his old speed, but he didn’t invest enough time and mileage to more fully develop his aerobic base.

Analyzing the more recent race times confirms this.

In the visual below, I’ve plotted the equivalent paces for the mile, 5k, 10k, half, and full marathon, based on four performances – his mile, 5k, half marathon, and marathon. In this case, I used Peter Riegel’s formula to compute equivalent race times, and then I converted them to paces. Paces work better when you’re trying to visually plot things on a line graph.

The blue line is the mile result (5:19), and it predicts the best paces for the other events.

The purple line is the 5k result (5:57/mi). Although the lines look fairly close together, there’s a pretty significant drop off between the 5k pace predicted by the mile result (5:41/mi) and the actual 5k pace (5:57/mi).

The pink line above that is based on the half marathon result (6:56/mi). Again, there is a significant drop off between the pace predicted from the 5k result (6:29/mi) and the actual race pace.

If you follow that pink line to the end, the predicted pace for a marathon is 7:14/mi. That’s roughly 3:10, and that’s basically what the runner did for the first 15-16 miles of the race.

The problem is that he didn’t have the aerobic base to actually convert any of those paces. If the 5k is slower than the prediction from the mile, and the half marathon is slower than the prediction from the 5k, it stands to reason that the marathon would also be slower than prediction from the half marathon.

So What Went Wrong?

There’s an old saying in politics, “It’s the economy, stupid.”

And most of the time, you can boil down what went wrong in a marathon to this statement, “It’s the pacing, stupid.”

On any given day, there’s a theoretical best pace that you can sustain for 26.2 miles. The goal is always to get close to that line – without going past it. The further you go past it in the first half, the worse your second half will be. By a lot.

Based on a half marathon result of 1:31, the runner targeted 3:10. Based on a simple reading of equivalent race times, this makes sense.

But when you look at the full picture, it was most definitely not the way to go. The runner was clearly under developed aerobically, and he failed to convert any of his shorter distance efforts to longer distance efforts. With the clear drop off from mile to 5k to half marathon, he should have also anticipated a drop off from the half marathon result to the marathon.

Instead of targeting a pace of 7:15/mi, he probably would have had a much better day if he had targeted 7:45/mi to 8:00/mi. Maybe 7:30/mi? You could quibble over the exact target pace, but 7:15/mi is clearly too fast – as evidenced by the dramatic blow up at the end.

As for the underlying problem, it’s almost definitely one of consistent mileage. He was already underdeveloped aerobically, he had only been building a base for six months, then he jumped into the abbreviated 12 week version of Pfitz, and he failed to put in the prescribed miles in that plan. There’s really no reason to look any further than this for an explanation.

In some cases, the weather can cause a similar issue. If it’s hot, and you don’t adjust your goal pace, you will pay the price. But in this case, the training issue is a perfectly good explanation and the weather was definitely not bad enough to be a significant factor.

What Can We Learn From This Case Study?

As I’ve written before, the marathon is a race that rewards patience.

This runner clearly has some talent, and he was getting on the right track by slowly building up to a base of 35-40mpw. If he had kept that up for a few more months, done a proper build with an 18 week plan, and hit the actual mileage targets, he’d have fared a lot better. But even then, he probably should have been a little more conservative in his pacing.

The first lesson here is that you can’t fake a marathon. You can muscle your way through a 5k pretty easily. You can even fake your way through a half marathon. But you can’t fake it through a marathon. If you’re not prepared, and you target an overly aggressive pace, you will pay the price.

The second lesson here is that you should take a look at the bigger picture when trying to determine your goal pace. This is the most important decision you can make about race day – more important than how many gels to take, what shoes to wear, or how many miles to run the day before.

By comparing multiple race results, you can get a good sense of how well developed you are aerobically. If your 5k result lines up with your 10k result, and your 10k result lines up with your half marathon result, then you can safely use that half marathon to predict your goal marathon pace.

But if there’s a significant drop off as you increase distances, you should take that as a sign – and be a little conservative in your goal setting.

I’m developing a tool that will allow you to input multiple race results and see the predicted paces graphed – like we did for this runner. Check back in a week or two, and I’ll hopefully have that done.

If you have any other thoughts about what went wrong here – or your own story of a misjudged goal pace – feel free to leave a comment below.