"There's only one hard and fast rule in running: sometimes you have to run one hard and fast."

Monday, May 2, 2011

How to Overanalyze, part 1 of 74

The things that got me interested in doing long trail races is the difficulty in figuring out how fast one should be able to do them and then figuring out how one should prepare. If you run a road marathon in one place in 3:30, you can expect to finish about that same time in another, but how do you decide how fast you could run 100 miles on technical trail if you've never done it?

One thing that ultras have in common with marathons is that all the ideas floating about pertain only to those trying to win or to those just trying to finish. The second-most common reason people find this blog is that they're trying to find a schedule of how to train to run 100 miles (the most common is "why men can run faster than women") and there are a few plans out there for first-timers. Otherwise, one ends up trying to follow how the top finishers train; while Geoff Roes and Tony Krupicka have very good blogs, it's impossible to get anything useful out of analysing their training. Ask around and you get a lot of "just find what works for you."

In trying to figure out how to run the Superior Sawtooth 100, I looked at how the top few runners trained and I had lengthy conversations with some of those who finished whose training seemed inadequate or unusual. Again, that gives you the winner or finisher, not anything in between. The problem with studying the middle is that there's too many people and too much variation; the numbers become intractable.

Since the course was changed - made more difficult - Superior's only been held a few times and the weather's varied from cold to rainy to hot. The finishers divide into one or two per year running under 24 hours, then a gap of a few hours, then a gradual increase in numbers up to the cut-off of 38 hours, often with packs finishing close to each other.

I decided to take a closer look at those who finished at the top of the pack, around 27 hours, but not in the race to win it outright. There are 6: Duke Rembleski, Chris Hanson, Scott Meyers, Adam Schwartz-Lowe, Nolan Ming and Helen Lavin. That turns out to be an interesting collection, all of whom I know at least a little, a few converts from other sports, a few 100 mile specialists. Duke's the closest to me in type, as an expert at shorter distances (he ran 61 minutes for 10 miles recently). Though they'd argue it, I'd say the majority of these were capable of running the course a couple of hours faster, except for Meyers (who's finished in 26:59, 28:27, 28:48, 29:48 and 34:52) and Hanson (27:05, 27:55, 28:34, 28:42, 29:21); these two "specialists" both had one very good day there.

People often say I train way too hard for the results I get. If you can run 24 hours, it doesn't take much training to run 27. Meyers and Hanson trained a lot more miles, but not as fast, as the others. Mileage helps guarantee finishing, but doesn't make one faster. The closer you're trying to get to your best possible time, the more mileage you'll have to do; there's a law of diminishing returns, but more... is more.

Looking at these runners' other races, the one most commonly run is the Voyageur 50 Mile. So what time there corresponds to 27 hours? Helen's run 8:06 and 8:19; Duke 8:28; Chris 9:04, 9:26, 9:26 and 9:26 (talk about consistency!); Scott 9:33, 9:42, 9:47, 10:16 and 10:51; Adam 10:24 and 10:29. What can one possibly tell from that wide a range? Well, one can assume that Helen and Duke raced Voyageur all-out and didn't put as much into Superior. Adam apparently jogged through Voyageur as a very easy run. That leaves Chris and Scott; it looks like Scott's more of a 100 mile guy than Chris and that makes 9:30 at Voyageur sort of a minimum to break 27 at Superior and 9:00 probably the goal.

So... three times the Voyageur time equals Superior time??? Chris Gardner's 6:55 at Voyageur (he also ran 7:38 and 7:42) would then be under 21 at Superior (he ran 21:58) and Brian Peterson's 7:52 (he also did 8:26) would be just under 24 (he ran 22:35). Give or take an hour or so, it's not a bad rule of thumb.

And that makes my 10:50 there (ignore the 11:22 in 2009) good for finishing Superior in about 33 hours. That sounds about right.


Glaven Q. Heisenberg said...

Uh, I didn't follow this. Which is fine. I didn't want to anyway.

wildknits said...

Interesting theory... can't wait to see how this all works out.

Adam said...

I can clear up my stats a bit. I only ran Voyageur once, and I got royally spanked at it. But pretty much from that point on I refined my training and that culminated in a good Sawtooth run.

Check out realendurance.com, and go into the 'Relative Race Finish Times' from some good numbers on comparing different races. I think they regularly farm numbers from ultrasignup.

Anonymous said...

Actually now that I look at the numbers about the Voyageur times 3 for Sawtooth seems to be pretty accurate. Makes me feel less crappy about such a "slow" 100 mile time.