Monday, June 26, 2017

Who is the most technical woman in the season 2016?

Usually, people believe that winner of Middle distance at the main IOF competitions is the most technical athlete. But the problem is that you have the opportunity to win the race with imperfect navigation simple cause you are in really good physical shape. I have developed the method of performance analysis with the aim to estimate the O-technical level of international level athletes (but the original idea was generated by mathematician and O-athlete Egor Kostylev).

Step 1. Take all main IOF Middle events and manually mark all technically challenging legs with straight running (actually, legs with route choices are important as well, but Jan Kocbach has developed really good tools to performance analysis of these skills).

Middle WC #1 (Poland):


Middle EOC-2016 (CZE):

WOC-2017 (SWE):

Step 2. Define the level of an athlete. The purpose of this step is to prepare split data to further analysis. Some runners are quite stable and technically good, but simply not so strong physically like Matthias Kyburz or Tove Alexandersson. Let say for some runner, his best time on the legs is -7% of the best time, for other it is 0% (top runners). Every leg inside this threshold is perfect. But if you are able to show best split time then there are no needs to use for you normalized split-times.

Step 3. Create thresholds of mistakes and compare leg time of particular runner with superman time. Obviously that someone can be really technically but not so fast like a winner.

Step 4. Make computer analysis of several top races (EOC, WOC, WC) to find time-loss statistics of particular runners.

Step 5. Make ranking based on the normalized split-times for straight technical legs.

And here is the rank!
The full list and all calculations are available here. I was so surprised by the outcome of my analysis that checked all GPS-tracks of Anja Arbter. But that's true - she performed really stable races all three times!

There are some limitations in this method:
1. Selection of straight technical legs - we have to use expert opinion. It's hard to make it automatically.
2.We can not make a valid conclusion based on the one race. But then:
3. The ranking system is influenced by some subjective assumptions (let say, those who was running only two IOF-races were down in the overall rank - like Marika or Ida Bobach).
4. Running in a pack formation. Take a look at the picture from Poland (World Cup, Middle):
Helena, Svetlana and Emma significant amount of time were running together. Definitely, it gives some gains. But I don't know how to eliminate this factor from the analysis.

This performance analysis method first time was reported at the nice Elite coaching seminar in Turkey (together with some other tools). And these ideas just develop discussion about the role of sports statistics in Orienteering:
Olav Lundanes tells about future of sports statistics in Orienteering
Anne Margrethe Hausken Nordberg tells about future of sports statistics in Orienteering
Ida Bobach tells about future of sports statistics in Orienteering 
Jerker Lysell tells about future of sports statistics in Orienteering
Helena Jansson tells about future of sports statistics in Orienteering 
Emil Wingstedt tells about future of sports statistics in Orienteering
Håkan Carlsson tells about future of sports statistics in Orienteering
Leonid Novikov tells about future of sports statistics in Orienteering
Petter Thoresen tells about future of sports statistics in Orienteering
Antti Harju tells about future of sports statistics in Orienteering
Bjørn Axel Gran tells about future of sports statistics in Orienteering

P.S. The author of the photo is Dobre Światło.
P.P.S. Mats Troeng and Jan Kocbach are two great guys! They have developed a lot of useful tools for the performance analysis (Quick Route, 3DRerun, WinSplit Pro, SPAS-analysis, 2dRerun Course Planning, etc.). It is easy to say that Mats and Jan have changed our sport! Let's join these amazing specialists and start to develop more tools for the sports statistics and performance analysis in Orienteering. 

11 comments:

  1. Thank you! More analysis are coming soon.

    ReplyDelete
  2. An interesting idea, but I think you are just determining who makes fewer mistakes, which is not the same as the technical ability. You can walk the course and do it perfectly and get a high score, but it does not mean you will still be able to navigate flawlessly at a proper speed. I think this is why Arbter got the highest score - she is probably too careful and is too afraid of making mistakes, so runs it more slowly than she is able to navigate. This would still be useful information for her, but it's just not the same as measuring the technical skills. I think a proper measurement should include some comparison with running without navigation. Perhaps compare technically easy and hard legs? If an athlete runs easy legs fast, but slows down significantly for the hard ones, then her technical ability is not very good, even if she does not make any mistakes.

    ReplyDelete
    Replies
    1. Thanks for your comment. I will try to test your idea during men races analysis.

      Delete
    2. In any case I don't think that Anja is scared by mistakes and has to slow down for tricky legs. But to be sure: it is better to hear from her or some of her coaches

      Delete
    3. Interesting analysis and cool to be far up in the list! My analysis have also shown that I usually do less technical mistakes than others in particular when it gets technically difficult. No doubt that I am on a lower physical level than the top runners but I would not say that I slow down when it gets difficult. In my opinion, lower physical capacity does not necessarily mean that you have more time to prevent mistakes as it still is the maximum speed (pulse, lactate) of that person. I hope this could answer your question. See you!

      Delete
  3. For the records:
    WOC16: 1. Tove 33.57 ... 52. Anja 47.44
    EOC16: 1. Tove 32.37 ... 40. Anja 42.29
    WC17: 1. Helena 38.31 ... 52. Anja 47.58

    It is obvious that higher physical capacity needs to be answered with a sharper technical performance to succeed in orienteering.
    Neglecting this makes your definition of "most technical" as your method (normalizing running speed) as your results rather questionable.

    ReplyDelete
    Replies
    1. Definitely, in the modern Orienteering you have to be in the incredibly good physical shape to the success. I guess that no-one is in doubt about this! The main question is how to measure your technical level. I am coach, and coaches need to have some methods to answer for the question: "Is my runner getting better after technical trainings or stuck in some-kind of plateau?" There are some approaches to answer for this question (and I will publish it), but there is no 100% perfect way to do this by a single method. If you know the answer for the question: "Is my runner getting better with technical trainings this season?", please, let us know! Share your wisdom.

      Delete
    2. Make a Re-run test:
      1) Let the athlete run a middle distance course.
      2) Let her/him rerun the course along the chosen route (there you get the time lost by orienteering technique)
      3) Let her/him rerun the course along the fastest route (there you get the time by not ideal choices). Sometimes the "better" choice cannot be extracted from map though.

      Delete
    3. One could define the technical level as the degree your overall performance is close to your shear running potential...
      http://o-zeugs.blogspot.ch/2018/04/what-is-good-re-run-curve.html

      Delete