Marathon Pacing

I first discovered the work of Veronique Billat in 2002, when I was working on my book The Cutting-Edge Runner. That’s a long time ago, but in retrospect I’m somewhat embarrassed that I hadn’t known about her even earlier, as she was then already well on her way toward titan status in the field of exercise science.

If you’re seeing Billat’s name for the first time, don’t be embarrassed. It’s not the job of everyday endurance athletes to maintain an up-to-date mental Who’s Who? in the area of sports science research. Now in her late 50s, Veronique is a Frenchwoman with a half-marathon PR of 1:18 who teaches at the Interdisciplinary University of Paris and has authored well over 200 scientific papers focusing primarily on the optimization of training methods in distance running. In 2018, she published a book titled Révolution Marathon that I never heard about because it was written in French. Correction: I hadn’t heard about the book until recently, when Johnathan Edwards, who is studying for his PhD under Billat, emailed me to ask if I would be willing to review an English version of it that he’s working on.

The answer was yes, of course, and within 24 hours I had a digital copy of the manuscript on my laptop. I’m not sure exactly what I expected, but The Science of the Marathon, as Edwards has rechristened the book, isn’t it—in a good way. Billat’s thesis is simple: The most effective way to run a marathon is not at the steady pace that is commonly advocated but at a variable pace, and by extension, marathon training should emphasize variable-pace running instead of being dominated by steady-pace running as it is for so many runners.

Billat makes a compelling case for this approach, beginning with the observation that the fastest marathons are run at variable speeds, and most often feature a fast start, a slower but oscillating middle, and a fast finish. She concedes that some of the underlying physiology that makes the variable-pace approach effective for the best marathoners in the world is absent in slower runners, but contends that a version of the same strategy is best for us mortals as well.

Why? A few reasons. First, Billat argues, starting fast allows a runner to get ahead of their goal time without putting themselves in a hole that they can’t climb out of, provided they slow down after 2 km (1.2 miles) or so. A fast start also primes the body and mind in ways that make the slower running that follows easier. Additionally, by generating high levels of lactate, starting fast creates a biochemical milieu in the muscles that stabilizes pH, preventing fatigue and making subsequent accelerations possible.

Furthermore, oscillations in pace allow runners to utilize more of their physiological toolkit than is possible when they lock into goal pace. They can, in a manner of speaking, rest one metabolic engine while firing another, generating a lot of lactate during surges and burning that lactate during slower segments. Managed properly, fluctuations in speed enable the marathoner to maintain a consistently comfortable effort level that all but ensures they are able to make a hard push in the final kilometers. I must confess, whereas some of the physiology Billat gets into flies over my head, this last rationale makes a ton of sense to me, as I know from studying Samuele Marcora’s psychobiological model of endurance performance that running performance is ultimately determined by perceived effort, and I know from personal experience that if you keep your pace steady in a marathon, your perceived effort level will fluctuate, and if you run by feel, your pace will fluctuate. As Billat puts it, by taking her approach, “You will no longer run to maintain a speed but rather run to maintain a feeling of racing intensity consistent with an ‘average’ effort on the marathon.”

The book mentions a 2006 study by Billat which showed that pacing by feel, hence variably, is more effective than steady pacing even at shorter distances. Subjects completed a 10K time trial by feel, and all of them exhibited fluctuations in pace. Later, the subjects were asked to complete a second 10K time trial running steadily at their average pace from the first. Amazingly, few of them were able to hold this pace longer than 7K.

It so happens that, when Edwards sent me Billat’s book, I was about six weeks away from taking a crack at running a 2:38 solo marathon time trial. My plan was to run 26.2 consecutive 6:04 miles. I am now considering the possibility of replacing my usual pacing strategy with a variable-speed approach. Having completed well over 40 marathons, I have some opinions of my own about how best to train for and execute a race of this distance, and I don’t intend to completely abandon my formula in favor of Billat’s. For example, she takes a dim view of high-volume training, whereas I know beyond the shadow of a doubt that high volume is critical to maximum marathon performance for me. But I have found her argument convincing enough to tweak my training according to Billat’s recommendations to prepare for this style of time trialing.

My first test was a 13.11-mile marathon-pace run, which again, I would normally have attempted to run at 6:04 per mile from start to finish. Instead I aimed to start with a pair of 5:52 miles, then run seven 6:10 miles, and finish with a cutdown from 6:02 to 5:56 over the closing four miles. In a word, it did not go well. After completing the first two miles in 5:49 and 5:51, feeling pretty comfortable, I slowed down to 6:10 pace expecting it to feel like an utter cakewalk, but it didn’t, and indeed I never really settled in at that slower pace. I was only able to cut down to 5:59 for the last mile, and doing so required a near-maximal effort (though I was slightly ahead of my target by then and wound up completing the run with an average pace of 6:03.5 per mile.

I’m smart enough not to conclude from this one experience that Billat’s variable-pace strategy doesn’t work, though at the same time I most certainly can’t conclude that it will work for me. Here’s the rest of my After Action Report on the test:

It was a hot day and I didn’t have my best legs going into the run. It’s possible I would have struggled even if I had used my usual steady pacing approach.

I’ve spent a lifetime making steady pacing second nature for myself. It’s not surprising, therefore, that the variable-speed strategy was a shock to my system. In the next test I will know better than to expect the slower running that follows the fast start to feel absurdly easy. I’ve done enough reading in exercise science to know that perceived effort is highly influenced by expectations. A relatively easy effort will feel harder if you expected it to feel easier.

Though Billat herself would probably disagree, at 49 I think I might just be too old to use the variable-speed approach effectively. In her book Billat points out that her strategy requires a large speed reserve. Her rule of thumb is that your maximum running speed must be at least double your marathon speed; otherwise the fast start and later surges will crush you. I’ve been working pretty hard on my speed lately because I’ve been participating in virtual mile races, so I’m already doing what I can to maximize my speed reserve, but it’s obvious that age has stripped me of my highest gears and they’re not coming back.

Be that as it may, I am at a point in my athletic career where nothing really matters all that much and I’m willing to take chances I would not have taken when I was younger. So I fully intend to continue guinea pigging Veroniqe Billat’s variable-speed marathon pacing strategy, and I’m confident it will be fun and interesting at the very least. More to come.