Wednesday, August 29, 2012

LT100 Race Analysis

So, being the OCD runner that I am, I grabbed all of the race results (with splits) off of the website and threw them in a spreadsheet. After massaging the data a bit (e.g., calculating cumulative time columns for each racer at each aid station), and sorting it in a variety of ways, I was able to create the following tables.

Why bother? Well, mostly I'm just curious. But I also find it helpful to write down a few details about each split, along some thoughts about how I might be able to improve, before my memories of the race start to fade.

2012 Results vs. 2011 Results

2012 Results vs. 2012 Goals

2012 Results vs. 2012 Average

The tables above compare my splits from this year to my splits from last year, my goals for this year, and the average splits for this year. So, for example, I ran from the start to May Queen 13 minutes faster than last year, 3 minutes faster than my goal for this year, and 5 minutes faster than the average.

A few clarifications: Obviously, my results from last year were on the shorter course. Also, I didn't adjust my goals for this year when I found out that the course was going to be ~3 miles longer. Once I heard about the extra mileage I knew I probably wouldn't be hitting my goal splits from Twin to Winfield and back (and I let my crew know as much). I try to be as brutally honest as possible when I set my goal times for each aid station. My goal times aren't when I hope to be an aid station. They're not best case scenarios. They're my best guess as to when I'll actually be at an aid station. My crew rely on these times and I owe it to them to try to be as accurate as possible. If I've done my job right, it should be as likely for me to arrive at an aid station 10 minutes ahead of schedule as it would be for me to arrive at an aid station 10 minutes behind schedule. Of course, during the race there are a million variables that I can't predict or control, but that's what makes running ultramarathons so interesting, right?

I should also note that the average times were calculated from all of the runners who completed the split-- regardless of whether or not they ultimately finished the race. So, they're almost certainly a bit slower than the average split times for just the finishers because they will include the split times where folks blew up and DNF'ed at the next aid station. Finally, I want to point out that these are the average times-- not the median times. I'm guessing the median times are slower... maybe?

2012 Results vs. 2012 Median
Actually, no. It looks like the median times are actually faster than the average overall. I'm no statistician, so I'm not sure which one is more "correct" in this case. I think I prefer median because it's more "real". Someone actually ran the split in that time and half of the runners ran it faster and half of the runners ran it slower.

-160 places in 40 miles! That's -4 places/mile...
The table above tracks my place for each section of the race (and how much it changed from the last section). Also, my time for each split is ranked. So, for example, when I arrived at Half Pipe (inbound) I was in 288th place. From Twin to Half Pipe I passed 53 runners, and my time for that split was the 157th fastest. (As a reference point, I counted 796 people starting the race.) Again, as far as place and rank go, I'm considering all the runners who completed the split-- not just the finishers. There are a few minor holes in the data because a couple of split times weren't recorded (meaning that I couldn't calculate cumulative times for absolutely everyone). I really finished in 183rd place, so the numbers might be off by one or two places-- with slight errors being more likely for the later splits.

Start to May Queen

Not much to improve upon here. If I want to run this split faster I, um, need to run this split faster. The question is: can I do that without it coming back to hurt me later in the race? The fastest I would probably ever feel comfortable running to May Queen would probably be around 2:05-- and even that would make me a bit nervous. I mean my PR for the Turquoise Lake Half is 1:56. I'm just not that fast. In a lot of ways, I feel like your precise time to May Queen is a bit beyond your control due to the conga line effect around the lake. You can jockey for position before the mini-powerline climb, but after that you're kind of stuck. I certainly don't think it's wise to try to pass folks in the dark on the single track. My philosophy is that if you're ever forced to go slower than you'd like, don't get frustrated; just take the opportunity to eat more.

May Queen to Fish

Again, not much to improve upon here. Just like the first split, I know I could run this split faster, but it's probably not a good idea. Maybe I could run this section in 2:00 without killing myself? Doubtful. If I'm going to push a bit harder in the first 23 miles, maybe it makes more sense to push during this section rather than to May Queen? I'm not sure... I don't want to waste too much energy pushing uphill this early in the race.

Fish to Half Pipe

I might be able to shave a minute off of my stop at Pipeline, but I couldn't run through the Fish aid station any faster than I did. I don't think I want to run this split any faster-- certainly not to Pipeline. Maybe I could shave off a minute or two from Pipeline to Half Pipe-- but that's insignificant in the big scheme of things.

Half Pipe to Twin

Hey, here's an idea: don't get ITBS and stop four times to stretch for ~10 minutes on the way to Twin! Oh, and actually run the downhills faster than limping at a 12:30 min/mile pace! Now that I have more confidence in relying solely on energy drink for calories, I could definitely move through the Half Pipe aid station 5-10 minutes faster. So, there's lots of room for improvement for this split. This was my worst split, but it's also the split that I have the most confidence in improving upon. (Last year I ran it 24 minutes faster.)

Twin to Winfield

So, this split includes the time I spent at Twin and the time I spent at Hopeless (about 10 minutes each). Plus, it includes the ~3 minutes or so I spent icing my knee in Lake Creek. I could certainly move through the aid stations faster. I also feel like I got a little behind on calories/fluid during this section. I was really surprised how strong my split from Hope to Winfield was. I was expecting it to be much worse given my ITBS. So, I think 3:55 should be possible for this section. I've never broken 4:00 for this section (even on the shorter course), so that would be a big psychological boost.

Winfield to Twin

This split includes the time I spent at 3 aid stations: Winfield, Hopeless, and Twin (because I stopped to meet my crew before crossing the timing mat). I easily spent 10 minutes at each of those aid stations. It also includes the time I spent puking right after leaving Winfield. What a waste of time: eating for 10 minutes at Winfield only to spend 5 minutes puking it all up! If I had stayed on top of my nutrition/hydration during the previous section that probably wouldn't have happened. I spent a few minutes at Hopeless icing my knee (to no avail). Limping from Hope down to the river crossing sucked-- I lost a ton of time. I'd really like to run from Winfield to Twin faster than I run from Twin to Winfield. I'd also love to arrive at Twin before sunset. I think both of those goals are within my reach as long as I can keep my stomach and my IT band happy.

Twin to Half Pipe

Here's where I started clawing my way back into the race. At Twin, I was 41 minutes behind my time from last year; at Half Pipe I was 4 minutes ahead! Could I run this split faster? Probably, but not by much. There were several sections where I had to walk a bit due to knee pain and/or to settle my stomach (so that I could swallow Advil). I spent no time at aid stations during this split, so there's no room for improvement there.

Half Pipe to Fish

My strongest split, surprisingly. And my time includes 5+ minutes changing into long pants and different shoes (my Hokas) at Pipeline. I don't realistically think I could run this section much faster.

Fish to May Queen

This split includes the time (approximately 5 minutes) that I spent at Fish. It also includes the time I spent walking the road to Powerlines (and puking). I was severely limited by my IT band on the way down Sugarloaf. So, even though I set a PR for the section, I think I could possibly run it 10-15 minutes faster. That seems pretty fast to me, but it appears to be doable.

May Queen to Finish

This split includes the time I spent at May Queen-- where I had to spend some time (~5 minutes) in a porta potty. Strangely, I felt like I was moving faster during this section than I had in the past, but the results show otherwise. (My best time for this leg was in '10 when I was terrified of not making the 30-hour cut-off. I blew through May Queen without stopping, which made all the difference.) My ITBS almost completely limited my running except for a couple of miles on the Boulevard. I probably should've taken more Advil for this section, but I was worried about my kidneys. Assuming my knees are okay, I believe that I could run this section much faster. I don't think 3:20 is out of the question. There is a lot of room for improvement here. I would guess that the final 13 miles of the race hold the most room for improvement for almost everyone. In my opinion, too many racers kill themselves trying to shave 10 minutes off their time over Hope when it'd be much easier to make up that time (and more) in the final miles.

Best Decision

To blow through Half Pipe (inbound) without stopping. I recognized that stopping to eat "real" food was unnecessary, just slowing me down, and causing my knee to lock up. (And Pipeline was only 2.5 miles away.)

Worst Decision (Tie)

To not mix more energy drink at Hopeless (outbound). I needed those extra 300 calories and I puked shortly after Winfield as a result.

To chug a full bottle of 5-Hour Energy while approaching Powerlines. God, that stuff tastes awful! Next time: dilute it!

Hardest Decision

To stop and stretch for ~10 minutes on the way to Twin (outbound) to try to combat my emerging case of ITBS. I don't regret it, though I'm not sure it did anything meaningful, but it definitely meant sacrificing my short term goals for the long term.

Biggest Regret (If Time Travel Were Possible...)

If I had just continued to stretch/strengthen my hips in the 3-4 weeks leading up to the race I firmly believe ITBS wouldn't have been an issue. The score so far: ITBS: 2, Me: 1.

  1. Don't get ITBS and limp all the downhills!
  2. Spend less time at aid stations!
    1. Trust in energy drink to provide almost all of my calories. Cut back on stops for soup/potatoes.
    2. Don't fiddle with gear so much. Don't waste time changing shoes.
  3. Umm... puke less?
While I don't think there are enough savings to be had to go sub-25 and big buckle, I do think that a 26-27 hour finish is within my reach-- and without altering my training, except to include more time stretching/strengthening my hips (to prevent ITBS).

My knee pain has mostly faded, but I can still feel my left Achilles tendon. That's going to take a while to heal. I'm icing the tendon once or twice a day and I'm beginning to stretch and strengthen my hips and calves. I haven't run since the race. (I don't think I could run without tendon pain altering my form, which is not advisable.) Last week all I did was bike around the Mineral Belt Trail, hauling my daughter in the Chariot (and setting a PR for the route!). I think the bike really helped my recovery. I may try another bike this week and maybe a short run on the weekend if I'm feeling limber enough. I have surgery scheduled for mid-September to repair an epigastric hernia, and that'll knock me out of commission for 2 weeks or so. (Yes, I ran the 100 with my guts bulging out of stomach! How awesome is that?) I'm not precisely sure when I came down with the hernia, but I noticed it a few days after the Silver Rush 50, I think.

I guess you could say I ran my guts out this summer. :)

As for the fall, I'll probably be taking it pretty easy. Part of me really wants to continue to race because I'm at peak fitness right now, but another part of me is content to simply be done for the season. I'll probably run casually in September, enjoying the fall colors when I can. Maybe a marathon in November? Who knows? I've got to complete my recovery before I make any serious plans...


  1. Andy, where did you get your placement data in the last chart?

  2. Well, once you've created columns for each racer's cumulative time to each aid station (by simply adding their splits together) you can then sort the data by the cumulative time to each aid station and take note of your row to figure out your place. If you simply sort data by each split time, your row will indicate your splint rank. Make sense?

  3. Does that mean you could easily compile similar data for other runners? :-)

  4. LOL! Yeah, like Jim P. Said... sorry I was lost after the word column. Happy Friday!

  5. Sure! Jim: all it took was for me to follow the link to your blog and notice that you ran the San Juan Solstice 50 in 10 something for me to immediately know that you certainly beat me in the 100! Solid! That race always kicks my ass...

    MQ: 136th
    Fish: 100th
    Half Pipe: 88th
    Twin: 90th
    Hope: 148th
    Winfield: 150th
    Hope: 179th
    Twin: 149th
    Half Pipe: 144th
    Fish: 140th
    MQ: 119th
    Finish: 102nd

    Again, there are some discrepencies in the data (you really finished 104th), but that should give you some idea of where you were during the race. Awesome job!

    No pressure: but based on your SJS50 results you should be able to finish at least 50 places faster, imo! :)

  6. Andy... Could you calculate mine too please???? That wold be awesome!

  7. No problem, Andy!

    MQ: 256th
    Fish: 213th
    Half Pipe: 204th
    Twin: 222nd
    Hope: 195th
    Winfield: 184th
    Hope: 178th (You beat Jim by 7s!)
    Twin: 149th
    Half Pipe: 144th (Jim's <2m behind!)
    Fish: 122nd (You've got a gap now!)
    MQ: 95th
    Finish: 93rd

    Wow. A super consistent run! You only lost places on one leg. That's sold pacing, imo.

  8. Dude! That is so awesome to actually get to see. Thank you! I felt like I was consistent and or at least moving up all day but this shows it. And losing the spots on the way in to TL on the outbound I knew... something about that drop into TL get's me everytime. THANK YOU so much for doing that for me.

  9. No problem, man. Your 2:45 from Fish to MQ is sick. Great job!

  10. Yeah, you ran the 42nd fastest split for Fish to MQ!

  11. Can you calculate mine? Wait ...

    Impressive analysis. I love "phone book" (names and numbers) posts like this.

  12. Wow. Thank you! I think you could have a little cottage industry going here...teasing apart LT100 data and doling it out to less-data savvy runners!

    The data you did for me is telling...lost 60 spots between Twin and Winfield. (Laying down at Hopeless and sitting in a chair for 45 minutes at Winfield will do that to a guy.), then started clawing it back, especially between Fish and the Finish.

    Really cool data to see.

    Thanks again.