Analysis

WTC has announced a new schedule for the Kona start:

• 6:25 a.m. Pro Men
• 6:30 a.m. Pro Women
• 6:50 a.m. Age Group Men
• 7:00 a.m. Age Group Women

This will mean that the gap between the Pro Women and the Age Group Men will shrink from 25 minutes (where it was in 2013) to 20 minutes. This blog post has a look at what this might mean for the WPRO race. For a number of reasons, we will focus this analysis on the end of the bike:

• Running together isn’t much of an advantage
• Most of the Pro Women are actually running faster than the fast male age grouper
• most of the potential benefit occurs on the bike (e.g. illegal drafting or potential benefits from following within a legal distance)
• most of the potential problems occur on the bike (e.g. women being blocked by fading male or males not accepting to be overtaken by a women)

Gap of 25 minutes

In case you are not familiar with my „race development“ graphs, here is a quick introduction using last year’s Women’s Kona Pro race: The horizontal x-axis shows different points during the race, such as the start, after the swim, and various points on the bike and the run up to the finish. The vertical axis shows how far back each athlete is from the front of the women’s race. So for example, you can see all the athletes start together, then some of them being as far back as 20 minutes after the swim, and – towards the end – the final change in the lead around 15mi on the run. Each of the red lines represents one women’s race. Now let’s add the Age Group Men (blue lines) to this graph: You can see the men starting 25 minutes behind the women, and the fastest of them slowly making their way through the slower Pro women. Here are some more facts that you can’t easily discern from the graph:

• The first Age Grouper off the bike was Marc Unger. His AG time of 5:38 corresponds to 6:03 for women. This means that Marc would have been the „12th women“ (i.e. all but eleven Pro women were behind him).
• Linsey Corbin finished the race in 10th place. She had a great run and was in T2 at just under 6:10 (corresponding to an AG time of 5:45). When she hit T2, there were already 33 age groupers in T2, 21 of them less than five minutes ahead of her (a average gap of just over 12 seconds or 120 meters).
• Mirjam Weerd finished the race in 25th place. Her time into T2 was just above 6:20. When she hit T2, 123 age groupers had overtaken her, 48 within 5 minutes (a gap of just over 6 seconds or 60 meters).

Gap of 20 minutes

Now lets assume that the Age Group Men started five minutes earlier. You can easily see how many more men overtake the women and how much sooner they start to do so: Here are the corresponding facts to the 25 minute section:

• The first Age Grouper in T2, Marc Unger, would have been the „8th women“ (instead of 12th).
• Linsey Corbin (10th women) would have had 75 age groupers in front of her (instead of 33). In the five minutes in front of her would have been 42 athletes (instead of 21) with an average  gap of 6 seconds (instead of 12) or 60 meters.
• Mirjam Weerd (25th women) would have had 227 age groupers in front of her (instead of 123). In the five minutes in front of her would have been 104 athletes (instead of 48) with an average  gap of less than 3 seconds (instead of 6) or 30 meters.

Assessment

While the difference between 20 minutes and 25 minutes sounds very small, the difference to the second half of the women’s Pro field is very noticeable. Around the 25th women there is an average gap of 30 meters – technically still enough to avoid illegal drafting (12 meters from back of the bike in front of you to the front of your bike), but it seems obvious that there will be some larger groups forming that might either help the slower bike riders or cause problems for those with a good finish after pacing their bike well. Therefore, WTC should strongly consider leaving the gap at 25 minutes.

Options

Here are the options that I see that result in a 25 minute gap between the Male and Female Pros:

1. Move all the age groupers five minutes back. If the cutoff can be extended a few minutes past midnight, no one will be impacted by this. If that is not possible, the only group that would really be impacted by this are the slow women age grouper who are finishing within the last few minutes before 17 hours. (Looking at last year’s data, there were two women between 16:55 and 17:00: Harriett Andersen at 16:56 and Karen Aydelott at 17:00:48 – I don’t know if she was an official finisher.)
2. Move just the male age groupers five minutes back. This will shorten the gap between the male and female age groupers to five minutes. There is probably good reason for a ten minute gap – a five minute gap won’t help much to decrease the congestion in T2 and the early part of the bike. With the recent WTC announcement, I don’t see this happen.
3. Have the Pro men and Pro women start five minutes earlier. This probably not possible because it’ll still be dark at 6:20.
4. Have the Pro women start before the Pro men. The idea of this is that the faster men overtake the women on the early part of the bike and that both races would then be „clean“. I don’t know if this has ever been tried before, and an important race such as Kona is probably not the right place to start an experiment.

Weighing all the options, I think that the first option (move both age group starts one minute back) is the best. To alleviate the concerns for the 16-hour-women, maybe it is an option for the slower age group women to start with the age group men. It’ll be interesting to see if WTC acknowledges the concerns of the Pro Women for a clean race and will be open to more changes of the Kona start times.

Race Conditions

We’ve seen about the same conditions in Frankfurt as last year (adjustments of 17:55 vs. 18:27 in 2013, leading to a new course rating of 13:30) – confirming Frankfurt as one of the faster courses on the IM calendar. However, this year was especially quick on the bike (adjustment of 13:19 with a new bike rating of 8:27), and the hot conditions slowed things down on the run (adjustment o 1:33 and a new run rating of 2:10).

Male Race Results

By winning his first Ironman race, Sebastian Kienle had a great 30th birthday.

He took control of the race with a fantastic new bike course record of 4:12 (eight minutes faster than the old record by Andreas Raelert), then also finished with his first marathon under 2:50. Frederik van Lierde ran a bit faster than Sebi, but didn’t have a chance after the bike – even if his 4:21 was the second fastest bike split. In his IM debut, Jan Frodeno had some bad luck: First he tore his wetsuit before the swim (luckily he found a replacement in time) and then had three flats and had to wait for tech support to change his rear wheel. He was cramping even before starting the run in the changing tent, then had to stop and stretch and walked most of the aid stations in the second half of the run. I have no idea how he managed to have the fastest run split of the day with a 2:43!

Ronnie Schildknecht also had a good day finishing fourth, probably not quite good enough for a July Kona slot. There were a few notable DNFs: Bas Diederen injured his foot during the swim exit, rode well but couldn’t run at all. Also, Andreas Raelert stopped after 25k on the run while running in forth position. Kona is probably not in the cards this year for these two.

Female Race Results

While none of the women broke three hours on the run, the race changed a lot on the run. After the bike Jodie Swallow, Camilla Pedersen and Mary Beth Ellis were at the front, but fell back on the run or even DNF’d (Camilla with hip issues). There was a lot of talk after the race that a lot of the Pro women were caught in groups of strong AG men. (Gina Crawford described her frustration in her Frankfurt race report.) While there is personal responsibility by each athlete, this situation could have been avoided by not starting the Pro women together with the fastest age groupers. After Brasil and Frankfurt it seems obvious that there should be separate starts for Pro men, Pro women and age groupers, with as large a gap as possible between these three groups.

Jodie held the lead until about the half way on the run, then faded to 7th place. Corinne Abraham had the strongest bike/run combo and took her second Regional Championship after Melbourne 2013 and a long, frustrating injury. Elizabeth Lyles finished second, running down Gina Crawford who finished third.

Kristin Möller had the best run split, but as for Ronnie on the men’s side, the points for fourth place are probably not enough for her to qualify. Amy Marsh in 5th and Natascha Badmann in 6th (posting a 4:44 and new bike course record!) raced well, but started with probably too few points to get a July slot for Kona. Jodie in 7th and Mary Beth in 8th already had a lot of Kona points and should now be safe for a Kona slot. Eva Wutti was also fighting for a good place, she was near the front of the race and apparently collapsed exhausted just a kilometer before the finish.

Photo Credits: Michael Rauschendorfer (via tri-mag Twitter stream) and Siri Lindley (via Twitter)

This post adresses a consequence of my suggested qualifying system with a mix of direct and points slots: The use of the KPR as a Ranking System.

When the KPR was introduced, it’s stated reason for existence was to decide which athletes get a Kona slot. However, WTC and some parts of the triathlon press were also pushing its use as a Ranking System, for example by assigning the bib numbers in Kona based on the position in the KPR.

However, I don’t think the KPR is a good Ranking System:

• The „most important position“ in the KPR is whether you are over or under the cutoff line. This is what almost all athletes are focused on.
• Hardly anyone is really interested whether you are placed 8th or 21st.
• The only other position of some interest may be the #1 spot – and that spot is mostly decided after the 70.3 Championships and Kona. Last year, no one was able to challenge Leanda Cave (having won both races), this year Sebastian Kienle is almost assured the #1 spot (after winning 70.3 champs and placing 3rd in Kona).

There are a couple of elements that would have to be addressed in order to create a ranking system:

• Athletes don’t race very often during one year, therefore a ranking system would have to include results from more than one year.
• Even though WTC has most of the IM-distance races, there are other races that should be included in a Ranking System. This certainly includes Challenge races, maybe also off-distance races such as the ITU long distance championships or Abu Dhabi.
• In order to rank „the best athletes“, some factors would have to be weighted differently than for Kona qualifying (e.g. number of races, inclusion of 70.3s, how to deal with bad results).

As an example of an alternate ranking system, my own TriRating uses a totally different approach than the KPR:

• based on time instead of placing
• based on all results of an athlete (with older races having less of an influence into the overall rating)
• all races have the same weight

It is much more „stable“ than the KPR and – while certainly not perfect – probably a better ranking system.

It might be a good task for the ITU to develop a good ranking system that has a chance to be universally accepted. But the way the long-distance racing scene is fragmented between WTC, Challenge, other smaller races and the ITU, I can’t see any one organization to step up to this challenge.

My assessment: The KPR is not a good system for an overall ranking, but I can’t see a universally accepted alternative to evolve. I will certainly try to improve my own Rating System and hope that its use will increase.

In the last posts, I have been laying the groundwork for the changes to the KPR that I propose. This post brings these thoughts together and introduces the main elements of my suggestion for a modified Kona qualifying system.

Have some Direct Qualification Slots 

As I have outlined in „Why have a points system“, I think there are a lot of advantages to slots that you can get from having one great result in a limited number of Ironman races. My base line suggestion is as follows:

• Top 10 from Kona get a direct slot (while still requiring validation, winner still has an AQ status and doesn’t count = 9 slots)
• Top 5 from the three Regional Championships get a direct slot (3 races * 5 slots = 15 slots)
• Top 3 from the five „Continental races“ get a direct slot (5 races * 3 slots = 15 slots)

You can certainly play with the numbers and reduce the number of direct slots (e.g. 6 + 3*3 + 5*2 = 25 slots), but the important point is that there are direct slots (from racing very well in one race) and points slots (from racing well in multiple races).

Award the Rest of the Slots through a Points System

This leaves at least 11 slots open for those athletes that are willing to work for their slots (i.e. race more often) – in fact there would be some more as there won’t be a rolldown for the automatic slots. The points slots could be assigned by a system that is very similar to the KPR system today, with points from all of the Ironman races offering a Pro category counting towards Kona qualifying.

The number of races that count towards Kona qualifying could be similar to what we have today (up to 5 races including up to 3 70.3s), but as I discussed in “Which and how many races should qualify for Kona” I suggest to lower the number to reduce the impetus for racing many IMs over the season. My suggestion would be to have up to three IMs plus up to two 70.3s.

I would also suggest to simplify the points schedule: After the direct slots, all races should be P-2000 races (so that 11th Kona = 6th in the Regionals = 4th in the Continentals = 1st in other IMs = 2000 points).

Some more details

Of course there are a lot more details that have to be clarified for a complete system. Here are some that I have been thinking about:

• 50 slots for both the men and the women
• no roll down for the direct slots (slots declined or not needed go to the points slots – so there will probably be about 15 to 20 points slots)
• points slots are awarded at the end of July
• only August slots would be the direct slots from Mt. Tremblant
• the Regional and Continental races should be spread on the calendar so most months have one direct qualifying race (e.g. October-Kona, November-Arizona/Florida, December-Western Australia, March-Melbourne, April-South Africa, May-Brasil, June-France, July-Frankfurt, August-Mt. Tremblant)

I think that such a system could be a great compromise between those that want to focus on Kona and those that are willing to work hard for their Kona slot. It’ll be interesting to see what ideas WTC is considering and what system they will decide on.

Please add your voice to this discussion, it is important to Professional IM racing to have an open broad discussion before any final decisions are made!

After a nice long weekend (with a short trip to visit friends), here is the next post in my series of thoughts on different aspects of the KPR. The focus of this post is the type and number of races that count towards Kona qualifying.

This question has two different aspects:

• How many races should count?
• Should 70.3s count towards Kona qualifying?

Number of races

An issue that has been discussed for as long as the KPR has been announced is the number of races that count for Kona qualifying. There are different views on this issue: One type of athlete would like to be able to qualify in once race (and therefore argue for just one or two races counting), other athletes want to be rewarded for racing often (and want as many races as possible to be included). It is probably impossible to come up with a number of races that would satisfy both types of athletes. The 2013 changes have been aimed at pleasing the first type (better points for top 3 places), but have probably resulted in even more racing by those willing to race often and thereby increasing the required cutoffs.

My view is that WTC shouldn’t encourage a behavior where athletes have a chance to qualify for Kona by racing themselves into the ground. We are now at a point where Kona qualifying by racing often makes it virtually impossible to have a real off-season. Racing often may work for a season or two, but will usually catch up with athletes after some time and force a longer, usually unplanned downtime. There will always be athletes that are willing and able to race four or five Ironmen a year, but this is probably too much for the typical athlete and also doesn’t allow proper peaking for all these races.

Therefore, I think that the KPR should limit the number of Ironman races that count towards Kona qualifying. The current system allows for five scoring Ironman races. If you add Kona (which is the goal after all) and maybe another sub-par Ironman performance, this could easily mean six or seven IMs per year! I think that no more than two or three IMs should count for Kona qualifying. Last summer (when analyzing the impact of the new KPR points system), I was also simulating the impact of only two races for Kona – and except for some „edge cases“ close to the cutoff line there was hardly any difference in the athletes that would have qualified.

It has been argued frequently that one of the goals of the KPR was to force athletes to race more WTC races and therefore increasing the depth of the Pro field in races. The depth of Pro fields in the 2014 races indicates that there is sufficient interest in racing Pro and „forcing“ athletes to start often is no longer required from a business viewpoint.

70.3 for Kona?

Another question that has been discussed for a long time is whether results from 70.3 races should count for Kona qualifying. The idea behind including 70.3s is that if an athlete needs just a few more points, this shouldn’t force him to race another Ironman. Instead, by finishing reasonably well in a 70.3, the last few remaining points could be collected in a race that’s easier on the body. I’m not sure that the current KPR fulfills that goal, to me it looks more as if the 70.3 points also increase the cutoff points and therefore encourages even more racing.

At the time that the KPR was introduced, the 70.3 series wasn’t as well established as it is now, and being able to get Kona points in 70.3s was probably an extra benefit for the 70.3 series. I don’t think that this is necessary any longer: The 70.3 series is attractive by itself and manages to attract a lot of interest without any cross-promotion from IMs.

One issue that the 2013 KPR changes addressed was the relative merit of 70.3s and IMs. In the old system, a P-1000 IM was hardly any more „valuable“ that a P-750 or P-500 70.3. After the changes, all IMs have more points that every 70.3 (the only exception is the 70.3 championship as a P-3000 race). Therefore, winning an IM is more valuable than racing well in a 70.3 – as it should be for Kona qualifying.

I don’t really see a problem with 70.3s counting towards Kona qualifying. As far as I can see, the influence on the cutoff number by adding 70.3s into the KPR is relatively small if you limit the number of 70.3s. In addition, if you reduce the number of IMs counting for Kona qualifying, you also have to reduce the number of 70.3s. My suggestion would be to reduce the number of 70.3s from three (as it is today) to one or two (depending on weather two or three IMs count for Kona qualifying). In order to not reduce the number of races counting for Kona too much, I’d suggest points from two IMs plus one additional 70.3 or three IMs plus two additional 70.3s.

My assessment: A reduction of the number of races should be seriously considered to limit over-racing, but 70.3 should remain a part of Kona qualifying.