Thorsten

One of the ongoing issues in Ironman racing is the disparity between Pro men and Pro women slots in Kona (50 for men, 35 for women). The main argument for the different number of slots is that there are a lot more men than women racing. Roughly, there is a relation of 2 men to 1 women that has been pretty much unchanged in recent years.

However, women’s races are often more exciting than the men’s race. In one of my older posts on this issue I’ve presented data indicating that the racing at the front of the women’s race is probably even closer than on the men’s side. Arguments for that include the number of lead changes on the run, or the time difference among athletes on the podium.

However, the decision by WTC not to increase the number of women’s Kona slots seems to be based on the perception that men’s field is „deeper“ than the women’s field. This post has a look at this claim and tries to provide data.

Suggestions for Measuring Depth Of Field

There have been a few „proofs“ that the men’s field is deeper than women’s. However, they don’t hold up to closer scrutiny.

Kona 2014

I’ve heard Kona 2014 mentioned as anecdotal evidence that the women’s field was lacking strength. The observation was that by the time the field reached Kawaihae, 30 men were still in the race. At the same point in the women’s race, there were only ten contenders left.

Of course, using an arbitrary point in a single race is neither convincing nor statistically significant. There are tons of counterexamples, take for example Kona 2014 (the same race) about 30k into the run. At that time, the men’s race was practically decided with Sebastian Kienle 10 minutes ahead of the rest of the field, while there were at least three women with a chance to win the race (Daniela Ryf, Rachel Joyce and Mirinda Carfrae).

KPR Points at #50

Another argument that is used very often is the number of KPR points at #50. Here’s the data from 2014 (not using Automatic Qualifiers):

• Schildknecht 3.915 vs. Bazlen 3.595

There is only a 320 points difference, probably smaller than some proponents of this argument make it sound. Also, the difference is a consequence of #50 being interesting for the men and not interesting for the women. We can easily refute the argument that this shows a lack of depth for the women by looking at #35:

• Jurkiewicz 4.280 vs. Wee 5.040

While Bree has a lot more points than Jeremy, I don’t think that this shows that the men’s field is less deep than the women’s field – it just shows that #35 is interesting for the women and not for the men.

Time Differences

As an example of many similar lines of argument, Andrew Starykowicz has posted a detailed look at the time differences between the TopX in men’s and women’s fields. His data shows that the men’s races are much „tighter“ than the women’s races:

However, this ignores the fact that the women’s fields are smaller and that a smaller field leads to larger gaps between the finishers. To simplify a bit, if the women’s field is half as large as the men’s field, this leads to time differences twice as large (e.g. if the male have a difference at #10 of 9,5%, the expected female difference would be in the order of 19% instead of the observed 11%). So rather than show that the women’s fields are less competitive, the larger difference is more a sign of the smaller fields.

Races with Small Women’s Fields

Each season, there is at least won Pro race that has a very small women’s Pro field. In 2013 IM France had only five starters, in 2014 IM Wales had only two. Of course it’s always bad if prize money goes unclaimed (France paid 8 deep, Wales 6 deep), but I think that this is again a consequence of the lower number of women racing: There are too many races for the smaller women’s field. I have suggested in my post on the Registration Procedures that WTC should keep Pros informed about the number of athletes that have already suggested for a race so that these very low numbers can be avoided.

My Suggestion

It’s actually surprisingly hard to come up with a measurement for the depth of the field, considering the different field sizes and different finishing times. A way that I suggest are Relative Finish Distribution Charts („REFIDCHs“). A Relative Finish Distribution Chart shows the percentage of finishers for different percentages of the winner’s time. Here’s the chart for the men’s Kona finishers from 2011 to 2014 (the years that the KPR has been used to determine the fields):

On the horizontal axis we see the relative finishing time (for example 105% means that the time equivalent to 105% of the winners time in a given year). On the vertical axis the chart shows the part of the field that has finished faster than the time on the x-asis. (For example, roughly 35% of the athletes finished within 105% of the winner’s time.)

REFIDCHs can be used to visualize the depth of the field. A „deep“ field will finish pretty close to the winner, resulting in a very steep graph. A „less deep“ field will have a flatter curve, showing that more athletes have finished further away from the winner. So if the women in Kona were less competitive, they would have a flatter curve, one that would be to the right of the men’s graph.

So let’s add the women’s distribution:

The two graphs have a very similar shape – if there is a difference, the women’s graph is to the left of the men’s graph, indicating a slightly deeper women’s field.

Let’s control for the influence of the different field sizes. One could argue that the smaller women’s field cuts away the slower end of the field. (Anecdotally, this isn’t the case: Some women that missed Kona this year included athletes such as Amy Marsh, Angela Naeth, Rebekah Keat, Laura Bennett or Eimear Mullan that could have placed well in Kona.) This is pretty hard to factor into the data, so I’ve done the next best thing by removing the men’s qualifiers in #35 to #50 from the results. (For 2014, this would have affected such well-placed finishers as 2nd Ben Hoffman, 6th Nils Frommhold or 10th Romain Guillaume.) Adding in a third graph for the „reduced Men’s field“ produces the following chart:

While reducing the men’s field has led to a steeper graph (indicating that more athletes would have been cut from the „slower“ end of the field), the graphs are still very close together, crossing each other at various points. Using this chart, there is no indication that the women’s field is any less competitive or deep than the men’s field or even the reduced men’s field.

Conclusion

I couldn’t find any data that supports the claim that the depth of the women’s field is any worse than the men’s field. (If you have other suggestions, please let me know!) While the lower number of athletes leads to bigger gaps in the Ironman races across the globe, at least the women that made it to Kona are as competitive as their male counterparts. In my eyes, the women’s race in Kona would be even more exciting than it already is if there were 50 Pro slots for the women.

Last year I wrote a post on procedures around Pro registration for Ironman races. In it, I have made a few suggestions for improvements and wanted to check on the progress.

Suggestions from Last Year

Here is a quick overview of the changes I suggested:

1. Information on how many athletes have already registered for a race
2. Standard registration deadline of four weeks
3. Easily available start lists after end of registration
4. Stricter withdrawal process

Let’s have a closer look at what happened with each of these suggestions.

Registration Levels

I am not aware of any information that is made available to Pro athletes (or the triathlon press) that indicates how many athletes have registered for a race. If this information had been available, it might have helped to prevent the really small Pro fields (most notably IM Wales which only had two Pro women). From what I heard, some more athletes were interesting to start in Wales after the start list was published, but were declined as the registration deadline had already passed.

As I have suggested last year, I don’t think that putting out the information has to be complicated: A monthly email newsletter by Pro Services to the athletes could just note the current number of athletes registered for the races in the next months, and athletes could adjust their plans accordingly.

Registration Deadline

For most races, a registration deadline of three weeks was used (and in some cases, strictly enforced). I think that almost all Pro athletes are aware that they have to plan their races in time. Except for a few licensed races, I don’t really see this as an issue an more – a three week deadline seems to be standard and accepted.

Start Lists

Ironman has made big steps forward in making start lists available: They have a standard location on their website for Pro start lists (Results – Pro Athletes, then Event Registration – Pro Start Lists). A big thumbs up!

However, there are still some improvements that should be made for the 2015 season:

• While it’s been working well for North American and Asia/Pacific races, it was hard to get timely and up-to-date information for European races.
• The start list should be available within days after the registration deadline, but there were a lot of cases when it takes much longer.

Withdrawals

This is the area where more progress should to be made. Quite often, athletes that have announced on social media that they won’t be starting are still on the start list – apparently because Pro Registration has not been notified. It seems to me that the sanctions that Ironman has in place (basically starting at losing 500 KPR points) are not sufficient to ensure that all athletes that won’t start properly notify Pro Registration services.

Also, the information about withdrawals should be reflected in the published start lists. At this point, it is not apparent when start lists are updated and where the updates have occurred. I think two simple changes could help here:

• On Ironman.com, note the date that the list was last changed in addition to the race name (so instead of „Ironman Lake Placid“ it should be „Ironman Lake Placid (updated June 10th)“)
• In the list, cross out athletes that won’t start, and note late additions in a special way (e.g. color or by adding „LA“ at their bib).

Overall

I think that Ironman Pro Registration has put a lot of work into improving the information that is available. In addition, I’m thankful that Paula and Heather are alway very helpful and quick in answering my requests. (From what I heard, they are also great in answering questions from Pros and the triathlon press.) However, I think there is still room for improvement in the information that is publicly available, and I hope that this will also make their job a bit easier by reducing the number of requests they have to handle.