Data on Ultrarunning?

Elevation Trail World Headquarters

Elevation Trail World Headquarters

Join me, Tim Long, and Gary David today on Elevation Trail as we talk a bit about data and analysis available in trail and ultrarunning.  How do analytics impact a sport?  Hope you enjoy it!

About these ads

4 thoughts on “Data on Ultrarunning?

  1. I’ve got some ideas for ultrarunning statistics that might be leverages for analysis:

    1) Strava numbers. We would need a percentage of participation to make this interesting, but it could be part of an “elite” runners resume going into a big race. The numbers could be analyzed for the type of training leading up to the event (intensity, paces, elevation gains, mileage, etc). Once enough data is gathered we could even do analysis with respect to race performance (races would have their own statistics that could be matched up to a runner’s strengths/weaknesses, this could include elevation gains, course records, top ten spread time, technicality) I think there could be a wealth of analysis that could go on here.

    2) Ultra Signup statistics. As a racers portfolio is built up, there should be some data on what kind of races (again, elevation gains, CRs, technicality, etc). I know that Ultra Signup tries to use some of this to rank runners and rank entrant lists, etc, but I believe there is a lot of improvement that could be made in this area. And this information could be combined with Strava training numbers to really help you analyze.

    3) 5k, 10k, marathon PRs. These should be useful to give an overall idea of aerobic capabilities.

    4) Real time race analysis: SPOT device. SPOT device would allow you to upload gps positions at intervals that could then be collected for paces and positions during races. “Elite” runners could be outfitted with such devices that would be returned to a race org, or maybe even they might own one- specifics TBD. You could match up runner strength/weaknesses (whether their a climber, descender, a flats runner, etc), and their current performance in the race to make predictions about how they are doing or what they are going to do.

    Finally, this becomes important for race betting. Apparently the Super Bowl, even though it was a blowout, was still a TV hit because everyone was betting on the game. If you can make a race interesting enough from a gambling point of view, then money can be made available. Betting becomes more prevalent when statistics are available to give the house an edge…

  2. For rating a race with respect to “hardness”:

    The google cars use a form of smart radar to detect obstacles and through a pattern recognition system can make some predictions about various objects(are they moving, will they get in the way of the car etc.) We should be able to strap a similar system to a hiker, and have them hike a course to collect information about the trails with respect to rocks, roots, twists, bends, ups, downs, exposure, etc), then construct an algorithm that would give a number for various sections of a course (for instance per between aid station section), or for a climb, or descent, or whatever. This would act as a reliable way to rate the various races! The magic number that the algorithm spits out could be described for instance on a ten point scale with pictures/text description of what one might expect for a given rating…

  3. Good show guys.
    Increasingly Pikes Peak Marathon has provided more data (and attempts to provide it real time) in their race. Their historical stack of results are something I have spent ridiculous time geeking through: http://www.skyrunner.com/search/find.asp
    SOS certainly has done some analyis for marathons as well. And heck, pick up nearly any TnF news and it reads like a phone book: names and numbers with splits galore on every possible race, finisher. Of course track events (and to some extent the marathon) lend to that sort of comparison over events and over the years. Those sort of compilations ain’t real time but at good track meets, they give you the splits over the PA (although all the distance junkies are watching them on their own anyway). The debut marathons of Farrah (in London) and Bekele (in Paris) will be compared to no end even though the races are probably going to each take on a different flavor in how they are raced. Additionally it has been suspected that part of the resurgence we see in HS running is because kids can so easily see the data of some other guy’s results – across the country, in near real time.
    I think part of the lack of comparison in 100s is because there is such variation in the trail element of it. Hardly anyone says, “wow, that guy did such and such pace per mile at HR” because it does not mean jack. However, when Biter did 7:08s recently for that record run – that sort of thing got people’s attention on the stat side.
    As far as predicting, aren’t we sort of doing that with guys like Sage, King and until recently Krar? Based on their 800 meter to marathon times we were all guessing at what they were/are going to do at the 100 distance. I guess Krar’s R2R2R laid into that as well.
    Tim – I think the blogger you were talking about in terms of LT predictions / stats was Mitch Leblanc. For example: http://www.mitchleblanc.com/2012/07/leadville-2011-data-analysis/

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s