The Yin-Yang of Fast Marathons – Mathematical Runner

When marathon historians look back at the 2018 New York City Marathon, they’ll quickly note that both Lilisa Desisa (2:05:59) and Mary Keitany (2:22:48) ran the second-fastest times ever at New York. Course records: 2:05:05 (Geoffrey Mutai, 2011) and 2:22:31 (Margaret Okayo, 2003).

But mainly they’ll talk about the way Keitany and the other top women covered the course.

In a sport where physical and metabolic homeostasis is everything, endurance experts have long counseled even-pace running or perhaps a slightly slower start and faster finish (negative splits). Sunday in New York, the women took it to another level. They ran black-hole splits. This is not necessarily a good thing.

Keitany covered the second half of the course nearly nine minutes faster than the first half (75:50 vs 66:58). No one else came close to her, but everyone else also ran big negative splits of four or five minutes. The first and second half of the NYC course are roughly equivalent.

This leaves observers to choose between two reactions: Wow, those women sure ran strong the second half. Or roughly the opposite: Why did they lollygag the first 13.1 miles?

Some research has indicated that women are less likely to hit the wall and slow down in marathons than men. Are women tougher? Do they burn fat more efficiently? Or do they simply start slower? These questions don’t lend themselves to easy resolution. The answer could be none of the above, or all three, or more than three.

Regardless, the men’s and women’s world-record marathons were both achieved with almost identical negative splits. Paula Radcliffe and Eliud Kipchoge sped up by a bit more than 30 seconds for their second 13.1 miles. In her American record marathon, Deena Kastor ran perfectly even splits; in his American record, Khalid Khannouchi increased his pace by two seconds over the second half. The following Table compares these efforts, and a few others, with the women’s race on Sunday.

Marathon 1st half 2nd half Difference Final
Radcliffe London 03 68:02 67:23 0:39 2:15:25*
Kipchoge Berlin 18 61:06 60:33 0:33 2:01:39*
Kipchoge Monza 17 59:57 60:28 0:31 2:00:25**
Desisa NYC 18 63:57 62:02 1:55 2:05:59
Keitany NYC 18 75:50 66:58 8:52 2:22:48
Cheruiyot NYC 18 75:50 70:12 5:38 2:26:02
Flanagan NYC 18 75:50 70:32 5:18 2:26:22
Flanagan NYC 17 76:18 70:35 5:43 2:26:53
Kastor London 06 69:48 69:48 0:00 2:19:36**
Khannouchi London 03 62:50 62:48 0:02 2:05:38**
Huddle NYC 18 75:50 70:54 4:56 2:26:44
Tusa NYC 18 75:50 71:23 4:27 2:27:13
Linden NYC 18 75:51 72:00 3:51 2:27:51
Kieffer NYC 18 76:34 71:38 4:56 2:28:12

* World record

** Nike Breaking2 time trial

The Best Marathon Pacing Plan

It’s important to ask what if anything is known about optimal marathon pacing. Alex Hutchinson looked into this and other endurance issues for 10 years while researching his book, Endure. “The women undoubtedly left some time on the table in New York,” he says. “But I don’t believe that either scientists or athletes really know the absolute best way to pace a marathon. What we know is that it’s going to be awfully close to even-pace.”

Coach Brad Hudson had an athlete on the streets of New York on Sunday: Allie Kieffer. “We had hoped for about a 1:14 halfway,” he says. “I thought she might be able to close about a minute faster and run low 2:27s. I like slower starts, just not as slow as Sunday.” Kieffer finished in 2:28:12.

Another well-known coach, Steve Magness, co-author of the book Peak Performance, was among many who noted the unusual running Sunday. “That big a negative split is surprising when everyone does it,” he observed, “especially at New York where the runners are usually hanging on when they hit Fifth Ave. and Central Park.” Magness said he would advise a 2:24 marathoner to aim for 1:13:30 or so for the first 13.1 miles.

The Mayo Clinic’s Michael Joyner has been analyzing fast marathon performances for more than three decades, and was the first to propose the possibility of a sub-2-hour marathon. He believes every marathoner has a set of pacing tools that can improve performance. These include: 1) Run slightly slower the first half; 2) take advantage of downhill stretches; 3) get in a pack and draft as much as you can; and 4) think about how good it’s going to feel to pass others in the final miles.

In his thesis at BYU, U.S. Olympic marathoner and statistician Jared Ward reached similar conclusions. He and a colleague investigated split times at the 2013 St. George Marathon. Conclusion: “Start the race more conservatively, and run the downhill portions faster.” Ward finished first among American men in New York with a 2:12:24 (65:25 first half, 66:59 second half).

A Modest Proposal

The elite runners in New York don’t compete in a vacuum; they’re part of a New York Road Runners event. That organization long ago stopped hiring rabbits to assist in world-record attempts, but it does offer generous time bonuses, presumably to motivate fast times by elites.

What’s missing is any strategy to link the two–fast runners and fast times. So here’s an idea. The NYC Marathon has pacers for 3-hour, 4-hour, and 5-hour runners. Why not do the same up front? Why not hire elite pacers to run the first half roughly 60 seconds slower than course-record pace? In the women’s race, the pacers might aim for a halfway time of about 1:12 to 1:12:30.

That’s an honest effort that would still allow the fittest women to run a faster second half. Meanwhile, the less fit could soldier on to finish in the time-bonus zone. There would be no lollygagging.

Top marathoners only get a handful of prime race opportunities during an entire career, and half of those are scotched by bad weather. They want the chance to run for the win–always!–but also the opportunity to record personal bests (and maximize income) when they can’t win.

The weather in New York on Sunday was basically 100 percent perfect. The pacing was not. As a result, many ran less than their best.

Is It The Shoes?

In the last several years, a new Nike shoe, the Vaporfly 4%,  has added another wrinkle to the marathon equation. Can a pair of shoes make you faster or more enduring?

Before Sunday’s marathon, Des Linden and Molly Huddle, who run for Brooks and Saucony, sounded half-convinced. “It makes you pause and scratch your head,” Linden told “We’re seeing marathons closed faster than ever before. Is that the shoes? I don’t know, but it’s on the table.”

Huddle also had concerns. “I’ve beaten athletes at distances shorter than a marathon who were wearing VaporFlys,” she said. “I think in the marathon is maybe where its strengths come out.”

Research conducted at the University of Colorado’s Integrative Physiology lab has shown that the VaporFlys can reduce the “energetic cost” of running by 4 percent. However, that doesn’t mean that other shoes can’t do the same, or that the VaporFly shoes work for everyone. Also, energetic cost is not an Olympic event, and will never win anyone a gold medal or big prize money.

Besides, running faster has an underappreciated yin/yang quality to it. On a stopwatch, it looks glorious. But it also causes more total body stress, which a runner might not appreciate after 22 miles of marathon running. “Theoretically, running faster might result in more muscle damage during a race, which would impair running performance,” explains Wouter Hoogkamer, PhD, of the Colorado lab. He and his team are continuing to explore the role that shoe cushioning plays in marathon running.

At any rate, Keitany wore Adidas Boost shoes, not Nikes. Linden and Huddle wore customized shoes from their sponsors. The unusual women’s performances at New York didn’t come from energy-return shoes but from human decisions in the emotional and physically-demanding cauldron of marathon competition.

In the elite racing world, you have a plan, sure. And then someone says “Ready-Set-Go!” At that point, the best-laid plans often evaporate. Which is why elite marathon pacers deserve consideration. They could improve a situation fraught with nerves, indecision, fear of leading, and months of intense marathon training.

Share this:

This content was originally published here.