Thursday, October 13, 2011

Analyzing the Ironman Run for Kona - Pro Men

One of the things I am always intrigued about is what separates the top elite athletes from the rest, and there is no doubt that the ability to run off the bike is a key determinant of success in triathlon. It was usually very clear to me in my days of racing, that how the first mile went on the run during an Ironman was a likely determinant of my how my run split would  be. Was this a mental thing, which gave me the confidence and desire to push myself? Perhaps.

In professional Ironman racing, you don't see perfect pacing. Instead, we see a strategy more like, "go until you fall apart, just don't fall apart as badly as everyone else!" The run is raced very similar to the bike, almost as a belief of a "peloton affect", where if the pack is let go, they never come back. On the bike, even if the athletes are riding "legally", there is still some draft benefit from the group as a whole, plus motorcycles of draft marshals possibly shielding the crosswinds at times, and possibly media, photographers, etc. And obviously there is the mental help of having others around to keep the pace and intensity high and consistent.

But is this the best strategy for the run? Would someone being smart in their pacing for the run coming off the bike with the leaders or main pack of contenders, possibly be able to take the victory, simply by pacing better? Or would they at least maximize the potential for highest placing, should they not be an outright contender for the victory.

I tried to gather evidence of this last year, but only with a Garmin 310XT, riding behind the athletes on a bicycle for a brief time, but this data was such a small sample it was not accurate, nor did it take into account the gradient at the different points I measured each of the athletes at.

This year, I set out to stand at the mile marker, and with the help of some viewers at home, and a couple of spotters/recorders at the mile mark, we were able to determine the first mile split for the top men and women, in almost real time during the event. I was actually able to tweet these paces live, (@jimvance).

As a preface, here is the profile of the first mile, from (click on image to enlarge)

The 1st mile has a climb based upon this source, but overall has a net gain/loss of close to zero. This was relatively confirmed by some of the files my athletes had from 4 different run files from the race, via Garmins. (There was some variance, but so small, this doesn't really affect the data, as in a gigantic uphill or downhill mile).

The men's data according to place of finish, from those I was able to get splits for: (click on image to enlarge)

There seems to be a general trend of lower differential of first mile time to actual pace performed equating to higher placings, with some exceptions. Meaning that perhaps the better pacers do better in overall placings. Or is it simply the athletes who can best tolerate the fast pace at the beginning, with the least consequences? Perhaps those who train for the initial acceleration?

Courtney Ogden had a differential of only 2:07 from his projected pace at mile 1, relative to his actual performance. It appears he came off the bike in 31st place, meaning he moved up nearly 50% of the pro men's field. This was an impressive performance, as the next best of those I recorded was Bockel, who was the only other under 10 minutes differential. Bockel's pacing helped secure a 4th place finish.

I recognize I did not get each and every elite who came by, so there is some missing data, but I did get many of the leaders and those who I felt there would be interest in, as well as those likely to finish in the top 10, including 8 of the final top 10.

So here are the questions I pose to you, would these elites be better off pacing themselves better in the first mile, or is being a part of the race and getting into the lead group the most important thing? Or is it best to train for this surge, and be able to keep the differential in the 9-16 minute range? Is the mental reward of being in the race for the podium from the start better for the athlete's performance?

Coach Vance


Evan said...

Let me know if you'd like me to run some simple stats on that to find out what the best predictor of race finish time out of your variables is. There's not much power due to the number of athlete's but it might be a start...

Evan said...

Jim - Let me know if you'd like me to run some simple stats on your collected variables to determine what the best "predictor" of race time would be. It wont be to powerful because of the small sample, but it's a start...

Jim Vance said...


Sure, let's analyze this further. I have similar data to the women which I will post in the next entry.

Let me know how I can get you the info. My e-mail is coachjimvance at g mail dot com.



Unknown said...

It is typically known that over the course of an Ironman run athletes deplete themselves of glycogen and become increasingly dehydrated. As they deplete themselves HR goes up. It would be powerful to know the HR of a 5:30 1st mile split and how that correlates to 7 minute miles towards the end. Maybe HR remained constant during the entire run, but due to depleted states pace begins to suffer. Thoughts?

Sciguy said...

I think that Courtney Ogden was really the only athlete to run an intelligent pace for the first mile. Despite being pros these guys paced just like your typical field at a local 10k with nearly everyone going out too hard. Andy Potts was an especially huge surprise in this regard. It's a time trial to the end and the first mile isn't the time to "psych out" your competition. Cruising buy at mile 20 when you're not trashed is the time for the psych out.

David D. Daggett said...

Well, part of the mystic at Kona is that the mile markers are not accurately or consistently placed.

David D. Daggett said...

Well, part of the mystic at Kona is that the mile markers are not accurately or consistently placed.

Jim Vance said...

@David, Agreed, but the differences in time from the exit of T2 to that point, are a consistent measure. And for the record, I was skeptical of the measurement, but after checking my athletes run files on the google map, and also using an independent source,, I think the accuracy of the marker is within a reasonable margin of error to accept.

Also, the athletes are looking at their splits at the marker, so mentally, they are seeing an equal marker for each person.