Jim2's Running Page

Predicting A Marathon Time From A 10k Race

Home
About Me
Favorite Links
Contact Me

1997 

Many runners use running calculators to project a marathon time from their performance in shorter distance races. All of these calculators simply multiply the short race time by a ratio to arrive at a predicted marathon time.  The ratio used by the current RWOL calculator (see http://www.runnersworld.com/home/0,1300,1-51-58,00.html) for predicting a marathon time from a 10k performance is 4.6. The ratio used by the Merv calculator (see http://merv.deas.harvard.edu/runcalc) is 4.67.

However, many runners find that the accuracy of the predictions varies, sometimes widely. Most runners opine that it's frequently due to being inadequately trained for the marathon. But just what does that mean?  What training factors are more likely to cause the deviation?

BrianW, an RW participant, aeronautical engineer and 2:40 marathoner, who was responsible for the creation of an RW Competitive Forum that later led to the creation of the Merv Competitive Forum, was of the opinion that training mileage in a marathon program is the most significant factor in the accuracy of the predictions. He conducted an extensive survey to collect data, which he used to correlate actual marathon/10k performance to mileage. He reported the results in a post on 12/16/97 in response to a question from another forumite concerning whether it's better to use training paces/performance or a 5/10k race performance to predict a marathon time. Brian's post follows:

"I think a race provides a more reliable measure than trying to compare training (although this can be done too). The longer the race the better. And using the RW calculator blindly is probably not a great thing, unless the race is longer than 10 miles. 5k's are unreliable, but I'd convert it to a 10k time then use a factor based on what the person mileage is. I never posted the results of my survey on marathon to 10k ratio versus mileage. Of course, there are variations among people, but considering the mileage helps reduce the variation. Here's what I came up with based on the survey:

30-35mpw: 5.5

40mpw: 5.0-5.3

55 mpw: 4.9

60 mpw: 4.75-4.85

70 mpw: 4.70-4.80

80-100 mpw: 4.55-4.65

Depending on how long a person has been running and how many marathons they've run, you can choose the slightly higher or lower values for a given mileage. The RW calculator gives 4.67, the world class runners varied for 4.55-4.8. But the 4.55 people never really tried to excel in the 10k, and the 4.8 people only gave the marathon one to two shots. A RW survey of 2,000 readers gave a number closer to 5.0. I would guess the average mileage of these readers was about 45 mpw. I wonder if they even thought to include that in their survey? I don't have that article."

Brian's data supports the premise that total mileage is an major element of quality, perhaps the most important element, in marathon training.  The higher mileage the runners in his survey ran, the lower the marathon-to-10k correlation ratio, i.e., the faster their marathons were relative to a 10k time.  And, it appears that race equivalency calculators are based on the high mileage run by very advanced, highly competitive runners.

Jim2