Friday, August 30, 2013

On one's pace, etc.

Steve Quick has written much over on his blog about the so-called "best" way to pace your race at Sawtooth: here, here, here are three recent and good examples of his thoughts on the topic. They generally focus on course-record splits and pacing. Should you run hard to Crosby and try to make it there before sundown? Should you try to run even splits? Push the overnight and try to negative split?

But he's also put this post up: Two Thoughts on the Superior Sawtooth 100. Thought no. 1 is about of all things, spectating Sawtooth, specifically where to go to make sure you see all of the faster runners who will finish and all of the mere-mortals who will finish. Now ultrarunning outside of a 12/24-hour course (think a track ultra or something like FANS) is not a spectator-friendly event. At a race like Sawtooth or most every other ultra, spectators only see runners at three locations: 1) the start; 2) aid stations and then only for a few minutes barring the runner's serious needs; 3) and the finish.

When you extrapolate Steve's time slots to the pace a runner must travel at to be there while the dedicated spectator is spectating, his groupings make sense. For example, compare that pace with the mileage necessary and someone should fall within one spectating time slot or the other, barring a drastic decline on the overnight that pushes a faster runner into the second spectating time slot as well. This is what happened to me at Sawtooth 2011.

In 2011, I fell into both groupings. I came into County Road 6 at almost 7:30 PM exactly, and came into Sugarloaf at about 9:30 AM after my overnight sufferfest to and through Crosby-Manitou. In sum, it took me 13 or so hours to do 30 miles. Pretty unacceptable to me for future races, but it was first overnight run, I had no idea what I was getting into, etc.

As noted in my previous post, I'd like to crank out 30- to 32-hour finish. Something 29:XX:XX would be fantastic, and I think I am capable of it. Looking back on 2011, my run to County Road 6 felt great. Perhaps a little fast at the start with race jitters, but things eventually calmed down because they had to. And I just kept pushing. Then the overnight hit and I took a huge cratering nosedive.

I've put together a rough pace chart, based on the premise of running 55 percent of the distance (to Sonju) in the first half of the allotted time and running the remaining 45 percent of the total mileage with the latter half. That means you're running the fast pace at least through some darkness, whether that's 15 miles or more. It's a 22 percent slow down, and the percentages are based off of some ultrarunning discussion that I've read places other than NPR. (Storkamp provides 50-mile splits for 100 mile finishers, so perhaps I'll have to compare my theoretical slow down to actual slowdown and somehow control for runner's ability.)

What does it look like? Surprisingly, it puts me at County Road 6 at 7:30 - just like 2011 - for a 30 hour finish and puts me at Sugarloaf at just before sunrise for a 32-finish and damn-near middle of the night for a 30-hour finish. Based on Steve Quick's groupings, that makes me in the faster grouping. It's also puts that last marathon at eight to nine hours, and the last 50K at 12 to 12.5 hours. Something crazy-slow just reading on paper, no? But absolutely realistic under course and race conditions.

I have these thoughts of my ideal race looking like the first 43 miles of 2011 and the 50-miler from last year, with a little fudging to get myself from County 6 to Finland. I didn't write a race report from last year's race (and I haven't raced an ultra since then), but I came in at 13:45 on basically no training and ran comfortably up until I hit the roots next to the Cross River in the section between Cramer Road and Temperance. From there, I slowed down but muscled through it and felt like I had my fluids and nutrition dialed in. I ran evenly, and led a large group for a good chunk of time with people commenting how even keeled I was. I ran with no watch - not an option at Sawtooth, me thinks, but I haven't totally kicked the idea - and went on feel.

I feel comfortable with the first daytime section. My key will be keeping that first section mellow and keeping moving despite having long section lengths (three of 10 miles and one of nine) - if we were to run this race backwards, I'm sure you'd see some more serious carnage that you presently do, Carlton Peak, Moose and Mystery Mountain be damned.

The overnight will see headphones get plugged in and a push forward through present, in-the-moment mindfulness. I know I likely slow down, but that needs to be minimized. Geoff Roes has a post on his blog or on iRunFar regarding maintaining one's self can save a minute or so per mile when you're feeling well, but several minutes per mile when you're feeling terrible. The latter will be key in the overnight.

The goal of the overnight is to simply get to Russ in those pre-dawn or early morning  hours. With new lighte my circadian rhythm will kick back in and I will be more enthusiastic to push onward now that my vision is not limited to a headlight beam.

That's how I plan to run 2013 Sawtooth - on feel, namely if it feels too fast, it probably is. We're going to consistency and avoiding a sufferfest and the death march that it induces.


samh said...

I love how much of a nerd you are, Matt.

SteveQ said...

So that's why my blog's getting so many weird hits! A good post of mine on Superior:

If you want, I can dig out some old records on splits for people who finished in 30-32 hours.

SteveQ said...

Oh... and before anyone thinks they know what a bad day at Sawtooth is like, check out

Matt Lutz said...

For a bad day at Sawtooth, see also:

Matt Lutz said...

@Steve: what I may do is to put a chart together of finishers whose 50-mile split is known - first 50 on X-axis, second on the Y-axis. The slope of the resulting formula that best fits the dots should give us the average amount of slow down.

SteveQ said...

Don't bother with 50 mile splits. As I recall, they only predict within 4 hours of finish (or, for Andy Holak, 15 hours). Especially if you combine different years.