Marathon myths article in NYTimes

copy and paste the article

Most runners have heard the marathon lore: Your time will be best if the weather on race day is about 55 degrees and overcast, or even drizzly. And avoid dehydration at all costs, because it will cause your muscles to cramp and you could collapse at the finish line.

But none of that is true, researchers said at a recent marathon medicine and science conference in Chicago.

The weather theory “needs adjusting,” said Scott J. Montain, a research physiologist at the United States Army Research Institute of Environmental Medicine in Natick, Mass.

“Most of what we know comes from the lay literature,” he said.

Thousands of runners are no doubt monitoring the weather forecast for Sunday, when the New York City Marathon makes its annual tour of the five boroughs. (As of yesterday, it looked promising, with temperatures expected to be in the upper 40’s and partially cloudy skies.) But the weather nostrums for marathoning that are cited so authoritatively in journal articles and textbooks are not always borne out in legitimate science. Montain and his colleagues set out to conduct a proper study.

They gathered data from 28 years of the New York City Marathon, 35 years of the Boston Marathon and 23 years of the marathons in Hartford, Vancouver, Duluth, Minn., and Richmond, Va. The routes for those marathons have barely changed over the years, and each had a large field — more than 10,000 runners. The investigators looked at the average times for the top three men and women, and at the times for the runners who placed 25th, 50th, 100th and 300th.

Elite runners ran fastest in the coldest conditions — 41 to 50 degrees. But the slowing effect with heat was not as great as had been previously reported. For every five-degree increase in temperature, times slowed by 0.4 percent.

Warmer weather had a greater effect on slower runners. On a 77-degree day, an elite runner would be about 5 percent slower than on a 41-degree day. But a runner who finished in three hours on a 41-degree day would be slowed by about 12 percent on a 77-degree day, finishing in 3 hours 21 minutes.

One reason, Montain said, could be that slower runners spend more time on the course, and the temperature generally rises through the day. Or it could be because slower runners tend to run with a larger pack. A tightly clustered group of runners generates heat and blocks it from dissipating.

Montain and his colleagues also looked at whether marathon times were better under sunny or overcast skies. Only 13 percent of records were set on cool and cloudy days.

“It is more likely that a record will be set when it is sunny or when there are scattered clouds,” Montain said. He is not sure why that is; perhaps sunny conditions put runners in a better mood, he suggested.

Then there is the issue of cramping, that often excruciating, spasmodic, involuntary contraction of muscles that can occur during or, more often, just after a marathon. It almost always involves the muscles that were used to run — the hamstrings or calf muscles, for example. And it can last a minute or two — or much longer.

Conventional wisdom says cramps are caused by dehydration and that the solution is to consume salt and drink more fluids. Not true, says Martin P. Schwellnus, a professor of sports medicine at the University of Cape Town in South Africa.

At the conference in Chicago last month, he reported that he could find no relationship between dehydration and cramping. He has studied cyclists, marathoners and triathletes, measuring levels of electrolytes and body-weight changes, both of which are indicators of dehydration. Those who cramped were no different from those who did not.

Two other studies looked at how much weight ultramarathon runners and triathletes lost during races — a measure of fluid loss and a direct indicator of dehydration. Those who cramped lost no more weight than those who did not. If anything, Schwellnus said, those who did not have cramps were slightly more dehydrated.

The cause of cramps, Schwellnus believes, is an alteration in the electrical signals going to exhausted muscles so that the balance between those signals activating muscles and those inhibiting them is distorted. One way to protect yourself is with proper marathon training and proper pacing. “Racing at too high of an intensity is one of the single most important risk factors,” Schwellnus said.

When muscles cramp, there is a simple and effective treatment: stop running and stretch that muscle. And, Schwellnus said, realize that the cramping will soon stop.

“Almost no matter what you do, if you stop the activity, the muscle will come back to normal,” he said.

Beyond the finish line of every marathon are runners who feel dizzy, and some of them collapse. It is not as common as muscle cramps, but the condition can afflict up to about 5 percent of marathon runners, said Michael N. Sawka, head of the thermal and mountain medicine division at the United States Army Research Institute of Environmental Medicine. He wondered whether the cause could be dehydration, a commonly evoked mechanism.

Sawka looked at published studies. One compared 45 athletes who collapsed after an ultramarathon to 65 who completed the race and did not collapse. There were no obvious differences between the two groups: their body temperatures were the same (dehydration makes the temperature rise), as were their electrolyte levels. But those who collapsed were pushing themselves as hard as they could, were at or close to their personal records, or were medal winners in the race. Perhaps, Sawka said, “that final effort might contribute to collapse.”

The actual cause, though, does not appear to be dehydration, Sawka said. Instead, it is a pooling of blood in the lower legs and feet when vigorous exercise suddenly stops and the heart rate slows markedly.

Timothy Noakes, a professor of exercise and sports science at the University of Cape Town, said he had stopped giving intravenous fluids to collapsed runners.

“We completely changed the way we treat patients,” Noakes said. “All we do is have them lie down and put their feet higher than their head.”

Postmarathon collapse, Noakes added, “is a benign condition.”

“Just lift their legs and you will help the majority of patients,” he said. “That’s all you need to do to make most people recover very, very quickly. You can infuse as much fluid as you want, and you will not get the same response.”

wow malmo is going to roll over in his grave once he hears this! :)

this is a classic example of the headline not being supported by the "facts" in the article. and they have some facts way off.

for example:

"They gathered data from 28 years of the New York City Marathon, 35 years of the Boston Marathon and 23 years of the marathons in Hartford, Vancouver, Duluth, Minn., and Richmond, Va. The routes for those marathons have barely changed over the years, and each had a large field — more than 10,000 runners"

Hartford, Vancouver and Richmond have never had 10,000 runners. Richmond totally reversed the direction of it's course several years ago.

"On a 77-degree day, an elite runner would be about 5 percent slower than on a 41-degree day. But a runner who finished in three hours on a 41-degree day would be slowed by about 12 percent on a 77-degree day, finishing in 3 hours 21 minutes."

i'd say 21 minutes for a 3-hour marathoner is a pretty big "factor"

"None of that is true"???? I don't see anything in this article that makes me think that 40's and cloudy isn't the best condition for a marathon. Even a five percent slowdown is significant (15 seconds at 5:00/mile), yet the article presents this factoid as if it is statistically meaningless.

Gee, could it be that the reduced slowdown in elites has something to do with most of them being 120-lb Africans, who have ab incredible heat-dissipation capability? Also, the data for the 77-degree day are skewed if they don't somehow account for the inevitable increase in DNFs.

Also, slowdown isn't linear with respect to temperature. The five degrees from 40-45 are NOT the same as the five degrees from 65-70.

Poor use of statistics, poor interpretation.

asdasd wrote:

wow malmo is going to roll over in his grave once he hears this! :)

Not hardly. The "newly discovered facts" are what we've already known for 50 years - the ideal marathon temperature is between 40-50 degrees. The fact that this is a revelation to a hack NYTimes scribe should be of no surprise to you.

So you rise from your grave to post here malmo? Do you walk on water too? give sight to the blind? turn water into wine etc etc?

or was asdasd's news of your demise premature?

So, in fields of 10,000+ runners, they compared the times of a grand total of 7 men and 7 women over the years for those races? And the slowest place they checked was 300th? It's too bad no one's invented any sort of machine that could aid humans in correlating large amounts of data so that meaningful conclusions can be reached. Maybe such miracle devices will be invented some day, but until then we'll have to make do with not-even-half-assed research conclusions based on tiny, heavily skewed nibblets of data. A-lack-a-day.

I don't know ... I certainly remember dehydrating big time in Boston 2004 (87 degrees) and then feeling my calves, abs, and my quads cramp up to the boint they were rock hard.

Last I checked, your sodum/potassium levels are what keep the muscles able to contract/relax so I think that having some salt in your body and the potassium is important.

25 timer wrote:

this is a classic example of the headline not being supported by the "facts" in the article. and they have some facts way off.

What??? The facts are way off?? Can't be, impossible. I mean, this is the New York Times, the holy grail of accurate, unbiased, and profoundly meaningful reporting and editorializing.

i emailed the writer and the sports editor. wonder if i will get a reply?

Agreed. 5% is huge for any serious runner. The NY Times is reputable journalism - that's a myth they should debunk.

25 timer wrote:

"On a 77-degree day, an elite runner would be about 5 percent slower than on a 41-degree day. But a runner who finished in three hours on a 41-degree day would be slowed by about 12 percent on a 77-degree day, finishing in 3 hours 21 minutes."

i'd say 21 minutes for a 3-hour marathoner is a pretty big "factor"

Journalist at major newspapers don't write the headlines. It seems to me only the headline was misleading, the article was interesting.

Yeah, the only interesting part of the article was the 5% to 12% negative effect at 77 degrees. I ran Boston a couple of years ago when it was 86 degrees. The winner still posted a 2:11 or something like that. He was from Kenya, where I assume that 86 degrees is pretty common. I ran 2:44, which I was totally thrilled with given the conditions. I'm from the Pac. North West. It's never 86 where I live. 65 degrees with a cool breeze is my version of a warm summer day. When the article talks about elites fading less, I wonder if any attention was paid to the origin of the elites, and the weather they usually train in. Surely that makes a difference. Hell, anything above 60 degrees and I'm worried about the heat. If you're from Texas, I'm thinkin' 60 degrees doesn't make you crap your shorts.

Check your source wrote:

What??? The facts are way off?? Can't be, impossible. I mean, this is the New York Times, the holy grail of accurate, unbiased, and profoundly meaningful reporting and editorializing.

See, the thing is...that it is. People like you like to pretend you're above the fray by claiming that excellent newspapers like the Times are biased or generally inaccurate, but what you can't fathom is that by taking such a needlessly critical stance, you're really just protecting your own fragile, closed mind. I mean, you can feel totally secure in your opinions as long as you can pretend that the other guy isn't reporting the facts, right?

Fat Old Man wrote:

Yeah, the only interesting part of the article was the 5% to 12% negative effect at 77 degrees. I ran Boston a couple of years ago when it was 86 degrees. The winner still posted a 2:11 or something like that. He was from Kenya, where I assume that 86 degrees is pretty common.

86 degrees is rare in Kenya (Eldoret). The average number of days per year over 85 is 2.

M Avg Hi Avg Lo

Jan 72 57

Feb 74 57

Mar 74 58

Apr 71 58

May 70 56

Jun 69 55

Jul 68 54

Aug 68 54

Sep 70 54

Oct 72 56

Nov 71 57

Dec 71 56

Lazy L wrote:

Check your source wrote:

What??? The facts are way off?? Can't be, impossible. I mean, this is the New York Times, the holy grail of accurate, unbiased, and profoundly meaningful reporting and editorializing.

See, the thing is...that it is. People like you like to pretend you're above the fray by claiming that excellent newspapers like the Times are biased or generally inaccurate, but what you can't fathom is that by taking such a needlessly critical stance, you're really just protecting your own fragile, closed mind. I mean, you can feel totally secure in your opinions as long as you can pretend that the other guy isn't reporting the facts, right?

Do you believe in little green men from mars too?

Lazy L wrote:

See, the thing is...that it is. People like you like to pretend you're above the fray by claiming that excellent newspapers like the Times are biased or generally inaccurate, but what you can't fathom is that by taking such a needlessly critical stance, you're really just protecting your own fragile, closed mind. I mean, you can feel totally secure in your opinions as long as you can pretend that the other guy isn't reporting the facts, right?

"Check your source doesn't have a fragile, closed mind. He/she's on the mark. It's not so much the facts that are the issue - there's nothing new here. The issue is the headline and the opening premise from which the article is based. It reads, "Most runners have heard the marathon lore: Your time will be best if the weather on race day is about 55 degrees and overcast, or even drizzly." I can't speak for 'most runners' but I've never heard of such lore. (full disclosure: I haven't read Runner's World since the early 70s when it was actually about competitive running). 55 degrees certainly is great marathon weather, but runners have known for decades that the optimal conditions call for temps in the 40s, which the article presented as if this was a new revelation. It is not.

Furthermore, the author, who displays a working knowledge of numbers on par with a People magazine reporter, goes on to say that, "But the slowing effect with heat was not as great as had been previously reported. For every five-degree increase in temperature, times slowed by 0.4 percent." Not as great as previously reported? By whom? A 0.4 percent decrement in performance for every 5 degrees in temp increase might seem like a small number to a cub reporter assigned to the yearly marathon beat, but to experienced athletes and fans that sounds just about right. Furthermore, in the very next paragraph, the author suggests that on a 77 degree day a marathoner would expect to run 5 percent slower than on a 41 degree day. That's a 36 degree difference, and you should have figured out in your head that the 0.4%/5 degree "formula" doesn't jibe with 5% result cited by the author. Not that this 5% figure is wrong - I for one would expect that the decremental effects of temperature would accelerate as it got warmer, something like: 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0 = 4.9%

One would expect that a cub reporter would be confused by the ostensibly "small" numbers, 0.4%, 5.0%, etc. The cold facts are that just a 2% decrease in performance to a world class marathoner would mean he'd run a whopping 2.5 minutes slower. That's big. For 5% that number would be 6.5 minutes.

Finally, the author drew a faulty conclusion about the effects of sun on performance by putting forth a non sequitur stat that "only 13 percent of records were set on cool and cloudy days." Besides the obvious question of "what records?" this is an obvious case of the old adage "correlation does not equal causation." I would be willing to bet that close to 13 percent of marathons were run on 'cool and cloudy days' as opposed to 'cool and sunny'. What I want to know is not how many "records" were set on cool and cloudy days, I want to know how the athletes fared time-wise on those cool and cloudy days. One glaring omission from the article is the weather conditions of the most productive NYC Marathon ever: 1983. It was cool, overcast and drizzly. Secondly, note that most marathons are run in the early morning, where the angle of the sun is low and direct radiant energy would not be as critical as it would in a late-starting marathon such as New York and especially Boston.

A perfect example of how a the direct sun affects temperature was today in New York. At 10:00 this morning it was 41 degrees in the city. I called one of the top managers and told him to go outside and walk the streets. Walking in the shade it was very cold, however, walking on the sunny side of the street it was quite warm. Had one been running the marathon effects of the sun would creep up in latter stages of the race in spite of the cool temps. Had you been spectating in the shade you would have thought it to be downright cold. Had it been overcast today, it would have felt cold on both sides of the street - perfect conditions for marathon running. I can guarantee you that at 10:00 on Sunday morning you would much rather have a cloudy day over a bright sunny one.

The effects of sunlight and sun angle are no more evident than in the last Olympics (or any Olympics, for that matter). The starting temperature was very warm, in the high 80s/low 90s as I remember, so why didn't the top runners run 7-8 minutes (or more) slower than they normally would? It's the sunlight, stupid. While the ambient temperature was high, the low angle of the sun and impending darkness made conditions where no more heat is being radiated onto the athletes. Still not optimal conditions, but not near what it would have been had those Olympic marathons started 2 hours earlier.

Most of you have figured out this concept on your own. When would you rather run in the Summer, a 90 degree day at 4:00pm or a 90 degree day at 8:00pm?