Thursday, November 13, 2008

A long answer to a short question

Mark asks regarding Clearwater: When are you going back to implement the changes you discuss? ;)

That's a bit of a tricky question to answer. Of course, part of the trick is needing to qualify. I had a pretty good race and was able to wrangle a spot,so at least I've shown that I have what it takes to get into position for a rolldown slot. As I continue to grow in this sport, I would hope to be able to finish even higher, increasing my chances, but it's still a bit of a crapshoot.

It won't even be a consideration for a few years. With family commitments, my wife and I have agreed on a plan of unquestioned training time every five years. (Thinking about it now, maybe I should have bargained for every 4 like the Olympics?) Since this year had a lot of volume just getting my feet wet, I'll be scaling back for the next few years. My plan is to focus on my short-course speed, doing sprint races for the next two years, build to Olympic or an under-prepared local HIM in year three, be slightly more serious with a HIM in year four building volume, so that year 5 would be either Ironman, or something similar. That means, at best, I wouldn't be looking at Clearwater for four years at least.

There's also the issue of scheduling. The two month period between Muskoka and Clearwater didn't work out well for me this time. Perhaps I can figure out a better plan, but it's such a weird stretch to deal with. If I could qualify at an earlier race, that would be optimal, like Calgary in late July, or Disney in May. Enough time to properly rest and allow for a good build. So what races I can get to, and get into, become part of the equation.

But then there's also the very short answer to the question: As long as the race remains in Clearwater, I'm 90% certain I wouldn't go back. I'm glad I went, and I would recommend to any first-time qualifier to go and have fun, but once you've been there and seen how the race plays out, I can understand why so many people won't return until a venue change is made.

It seems to be the cool thing to complain about the drafting after IMFL and Clearwater, and I don't want to fall into that trap (but I'm about to). Rather than just stamping my feet and proclaiming, "It's loaded with cheaters. I was never passed a single solo rider, but tons of groups wheel to wheel" (which is true) I decided I'd do a little analysis, being the analytical sort. I took a look at the top 50 times to T2 in the men's 35-39 and 40-44 age groups. Why these two? Well, M35-39 is my group, and M40-44 is the first men's wave that followed us, and most of the people who passed me were from this wave. Both of these age groups were divided into two waves, separated by five minutes, based on last name (A-L, M-Z for M35-39; A-K, L-Z for M40-44). My table below has the time to T2, total bike leg, and time for the last section of the bike, color coded by wave (Wave 1=blue, wave 2=green). All times were taken from the Ironman results pages, which list the top 100 finishers in each AG. Any missing values are due to the athlete not being listed in the top 100, or in the case of third bike leg time, no intermediate timing.



I should start by pointing out that none of this is proof of any drafting. An athlete is legal at 7m behind, which is about 0.8 seconds at 38 km/h. It should also be noted that the last leg of the bike ride has the causeway (which is one of the few places to break up a group), as well as the finishing stretch (people dismounting and running at different speeds).

Here are a few things I see: Obviously, there's some clustering. Either speed is dependent on your last name, or groups are forming on the road. Inside some of the clusters, you can find up to 3 minutes difference over the entire bike course, but only a few seconds over the last leg, and that the difference to T2 is very close to the difference in the last leg times. Probably the most interesting observation comes from a few riders down the list. If you look at M35-39, around positions 41-45, these guys do the last leg much faster than the riders immediately above them (by 90-120 seconds). These are all wave 1 athletes. If you look up, arriving exactly five minutes earlier (by race time) is a cluster of wave 2 athletes (positions 9-12), who ride the last leg at an almost identical speed. They all get to T2 within seconds by clock time. (There's a similar thing happening in M40-44, see 48-50 vs. 20-26.) There are probably a few other minor observations that can be made, but that should be enough.

None of this includes the riders even further back, or the women's field caught between these two groups, or the upcoming M45-49 and M18-29 waves that started later. If I had the time and the data, I'd love to do this for everyone on the course, and see what patterns arise.

I want to make a point that there doesn't seem to be any real drafting at the very front of the race. These folks are absolute hammers, who may benefit from the slingshot draft of passing everyone, but they seem to be that much stronger than everyone else. It's the MOP and behind where the issues are the greatest. During my ride, I didn't see a single marshall or any riders in the penalty tents. Everything seemed to go unchecked. I just wish I could redo this for a larger sample.

As I said before, this may seem like sour grapes, and maybe it is. I was 154 out of the water, and moved up to 153 by the end of the bike. I admit that I'm not the fastest biker on the block, but I've never been in the second half of bike splits, nevermind almost the last, so to only get one spot is surprising. I know I rode legally, and I know what I saw other riders do (less riding legally). And this exercise is, at least, some vague evidence that the complaints about Clearwater might not be that far off.

So that's why I wouldn't be going back. Unless you're at the absolute front, there's little chance of getting a decent placing without resorting to drafting (in my view). If they change the venue to some place for fairer competition, I'd love to attend another Worlds, but I don't see myself returning to Florida.

2 comments:

  1. Calling a horse "a horse" isn't sour grapes.

    Your analysis makes it pretty clear that the numbers are speaking for themselves. Without any event-specific information, I'd very much expect competitors to cross the finish line according to some sort of Poisson-like distribution.

    Do you have the numbers in a form that makes it easy to test that hypothesis? I bet the p-value will make me feel sick.

    From a selfish personal point of view, I'm glad to read that you're focussing on shorter races -- I'll have someone to race with ;)

    ReplyDelete
  2. I haven't gone through the formality of statistical testing, but even the "eyeball check" leads me away from any smooth distribution. As I said, I'd love to have the times for all waves, so I can compare clock-time arrivals (to find the inter-wave drafting). I'd also like to further examine the 80-120 place groups, since I have a feeling that that is where some of the worst drafting may be happening.

    I'm not sure at what point we'll be racing together, since you'll always be 5-10 minutes up the road, and when your running comes around, I won't see you until the finish line. But we'll at least have to coordinate which races you beat me at.

    ReplyDelete