Sunday, 24 November 2019

UTMB 2019 - Pacing stats and DNF rates


I have created a pacing chart based on the average time finishers made it through each checkpoint. It doesn't format well so I have put it in a separate post. 


I've looked at the DNFs by checkpoint but was curious to see if the drops were an even slice through the field or just the back of the field being systematically picked off by the cut offs.


Aid station Still in DNFs Distance DNF per k Finish P
Les Houches 2540 0 8 61%
St Gervais 2534 6 21 0.46 61%
Contamines 2497 43 31 3.70 62%
La Blame 2428 112 39 8.63 64%
Les Chapieux 2383 157 50 4.09 65%
Lac Combal 2287 253 65 6.40 68%
Courmayer 2203 337 78 6.46 71%
Courmayer exit 2098 442 78 74%
Arnouvaz 1939 601 95 9.35 80%
La Fouly 1803 737 109 9.71 86%
Champex 1684 856 123 8.50 92%
Trient 1607 933 140 4.53 97%
Finish 1555 985 170 1.73 100%


These graphs are all in the same format. X-axis is the time of arrival at checkpoint. Y-axis is the probability of going on to finish. As the race goes on, it converges to 100%.


Checkpoint 1 shows an interesting pattern. First runners through we less likely to finish. The Elites in the race have a much higher drop rate. Mid packers have the best rate with those at the back least likely to finish. 

I made it through in 57 minutes which is roughly 50% chance of finishing. There is no cut off at the aid station. My chart has 54 minutes as the typical split for someone finishing after 46 hours.


Same pattern again. There is a sharp drop off at about 3:20. The cut-off is 4:00. I made it through in 3:20. About 30% of those who arrived at about the same time finished.

This CP has a big drop off in finishing odds - almost linearly between 3 hours and 4 hours. My chart has 3:12 as the typical split for someone finishing after 46 hours.


5:20 - up to a 40% chance now. My chart has 5:10 as the typical split for someone finishing after 46 hours.

7:10- up to a 60% chance now. My chart also has 7:10 as the typical split for someone finishing after 46 hours.


9:58 - down to a 60% chance now. I was pretty slow over the Col Du Bonhomme. My chart also has 9:55 as the typical split for someone finishing after 46 hours.


14:58 still at 60% chance now. My chart also has 14:48 as the typical split for someone finishing after 46 hours.

18:06 somewhere between 40% and 60%. My chart also has 17:57 as the typical split for someone finishing after 46 hours.


18:30 up to 65%. A lot of people dropped here so getting out of the aid station improves your chance a lot. The trend for those near the cutoff to have massively lower chances of finishing had reduced.

23:30. Up to a race high of 75% compared with the field average of 80%.

I dropped about an hour later on the upper reaches of Colin the Ferret. My chart has 23:22 so I was still sufficiently clear of cut offs. Maybe I should have carried on through the thunderstorm and over into Switzerland...



Odds have greatly improved at La Fouly. Getting here in 27:36 was typical - being an hour up on cut-offs was plenty

By Champex, all expect those within an hour of cut off had roughly the same chance of finishing (92%).

97% finish rate from Trient. Cut off dodgers have a lower chance but still pretty high despite 25km to go

No comments:

Post a comment