Pick Your Pace

(This piece is for my book-in-progress titled See How We Run: Best Writings from 25 Years of Running Commentary. I am posting an excerpt here each week, this one from July 2005.)

My second marathon team of the year is well into its training now for the Portland Marathon. The questions that the first group asked are replaying regularly again.

Among the most popular requests is, “How fast should I train on the long runs?” My answer isn’t the same one I would have given before working with team one last winter and spring: “run slower than your projected marathon pace.”

A runner named Rebecca Montgomery from team two had heard similar advice, only more so. “I have read every book I can find about training,” she said, “and there’s something I don’t understand. Why, as many books say, should the long run be slower than your pace for the marathon?

“I figured out that I could probably run a marathon in about 4:30 [10:20 pace], and so my long runs should be 11-minute-plus miles. That feels excruciatingly slow.”

Then she popped that big question: “What do you think should the pace be for my long runs in preparation for the marathon?”

Rebecca didn’t get the sound-bite answer she wanted. My explanation was multi-parted. The first part was that slower-than-race-pace training works best for faster runners who have higher racing gears available to them.

I could average 90 seconds per mile faster than training pace for my early marathons in the three-hour range because I’d go faster yet in shorter races. But when the marathon times later reached four hours or more, I’d lost the higher gears and now trained for and completed marathons at about the same pace.

My second answer to Rebecca addressed the common mistake of calculating bass-ackward. Runners set a marathon time goal, then try to train at that pace. It seldom fits them because the goal – qualifying for Boston, breaking a round-number time – has little to do with current fitness realities.

My advice: let the training pace set the marathon pace instead trying to work out the timing in reverse – forcing a guesswork pace onto the training. And then let whatever happens happen on the long runs. The pace that feels right usually is.

I never suggested an exact training pace to anyone on marathon team one, but let it settle out naturally. That team had its graduation day in June. All 16 of our group finished at Newport, Oregon – in times starting at 3:26 and ranging far upward.

Two runners hit major walls that day for health reasons. The others averaged within 15 seconds per mile of their pace for the longest training runs (which for them was 21 miles). Three runners finished their marathons within two seconds of that pace.

For all but the fastest runners, this might be the best measure of marathon potential: continuing to the finish at the pace you’ve practiced. As with any reliable gauge, this one isn’t based on what you dream of maybe doing someday but instead on what you really have done lately.

UPDATE FROM 2015

That first team of 16 was a tiny sample from which to draw a conclusion about judging marathon potential marathon times. Now we’ve had hundreds of finishes, and I’m even more sure that the guideline above applies to most of them: the best single predictor in the race is the per-mile pace of the longest training run, multiplied by 26.2.

[Hundreds of previous articles, dating back to 1998, can be found at joehenderson.com/archive/. Many books of mine, old and recent, are now available in two different formats: in print and as ebooks from Amazon.com. Latest released was Going Far. Other titles: Home Runs, Joe’s Journal, Joe’s Team, Learning to Walk, Long Run Solution, Long Slow Distance, Marathon Training, Run Right Now, Run Right Now Training Log, See How We Run, and Starting Lines, plus Rich Englehart’s book about me, Slow Joe.]