Welcome to the Science of Sport, where we bring you the second, third, and fourth level of analysis you will not find anywhere else.

Be it doping in sport, hot topics like Caster Semenya or Oscar Pistorius, or the dehydration myth, we try to translate the science behind sports and sports performance.

Consider a donation if you like what you see here!

Did you know?
We published The Runner's Body in May 2009. With an average 4.4/5 stars on Amazon.com, it has been receiving positive reviews from runners and non-runners alike.

Available for the Kindle and also in the traditional paper back. It will make a great gift for the runners you know, and helps support our work here on The Science of Sport.

Saturday, September 19, 2009

Berlin Marathon: Revised prediction - no world record

Berlin Marathon revised prediction: The weather rules out a world record

I have to get a last minute revision to my prediction in. Yesterday, I called the upcoming Berlin Marathon as a world record to Duncan Kibet, some 30 seconds clear of Haile Gebrselassie.

However, this morning, two excellent sources - Letsrun (for news) and Globerunner (for great writing and stories) mentioned that the weather forecast for Berlin is not looking great. Peak temperatures of 25 degrees Celsius (touching 80F) mean the record, which really requires ideal conditions, might prove a touch out of reach. Add to this light breezes and relatively high humidity, and the record, now so tough that things need to be close to perfect, seems a longer shot than usual!

So I'm taking this 'gap' on the weekend to revise my prediction, and say that the record will NOT fall tomorrow.

How might it unfold? The SoS Crystal ball works overtime

Also yesterday, Haile Gebrselassie announced that he intends going through halfway in 61:30. He said "...I don’t want to be slower than 61.30 at halfway (compared to 62.04 last year), and would like to run 30k faster than last year”.

Last year, 30km was reached in 1:28:25. Given how quickly Geb finished last year, I dare say that if they're not at least equal to that in tomorrow's race, the record will elude both him and Kibet. Of course, in a head to head race over the final 5km, anything can happen, but I suspect the pace will not reach the levels it did last time, particularly if the temperatures get up there.

So what can we expect? To break the world record, the pace needs to average 2:56.2/km (or 4:43.6/mile pace). If Gebrselassie and co hit the target pace, they'll reach halfway having averaged a shade faster than 2:55/km, which puts them on course to crack the record by almost a minute, of course (a 2:03 projected time).

Look for the first 10km to be unaffected by the temperatures. I expect if the pacemakers are on target (which they have been for the last two years), then 10km will be reached in 29:09.

However, if the temperatures are high, even in the low 20s (last year it was 15 degrees the whole way), then I expect that the pace will start to drop from 10km onwards. Expect halfway to be reached in 61:57, which is slightly faster than last year (by 6 seconds), but which represents a significant slow down from 10km to 21km.

Progressive slowing due to the temperatures

That slowing of the pace will continue up to 30km, by which time the record will be out of reach. The crystal ball says 30 km in 1:28:41, which means the 20km interval from 10km to 30km will have been covered in just under an hour, a pace of 2:05:36. That will be testament to the higher temperature and humidity, but will mean that in order to break the record, the final 12.195 km will have to be run in 2:53/km (4:39/mile), which is a bridge too far, even in a competitive race.

So the record will slip away between 10km and 30km, which is typical when it's warmer - the body is too smart to simply carry on at the same pace until the athlete is forced to slow dramatically, and the physiology of pacing in the heat dictates that the slowing happens before the athlete hits that limiting temperature.

So the general trend will be for a fast start, then a progressive slowing up to about 35km, before it might pick up. However, warmer races are usually wars of attrition, and the winner will likely be the runner who maintains the pace, rather than lifts it. Sammy Wanjiru has rewritten the 'rule book' when it comes to racing in unfavourable weather conditions, but to crack a record as tough as this, I fear that every degree Celsius above about 16 degrees will cost the athletes a few seconds...

As for the race, I'm going to stick with the original projection - Kibet to win, Gebrselassie to finish about 30 seconds behind, but no world record for either. Look for the break to happen at about 38 km, as the pace just edges up to drop the Ethiopian.

So my revised finishing times, plus 10km, 21.1km, 30km splits, is shown below:

1. Duncan Kibet - 2:05:04 (29:09 - 61:57 - 1:28:41)
2. Haile Gebrselassie - 2:05:33

And referring to Giovani's question about confidence intervals...I'd say I am about 5% confident! But very confident it will be a great race!

Anyway, that's just a bit of fun, we'll see how far off the mark I am tomorrow.

Join us soon after the finish for our post race analysis and splits!


P.S. Good news for running - Martin Lel races again this weekend, in the Great North Run, where he has achieved success in the past. He takes on Jaouad Gharib in another great race this weekend. And then Lel heads to New York in November, so that's great news for fans of Lel, who, when he was on form about 18 months ago, was a fearsome runner, and my favourite marathoner.


Frans Rutten said...

The marathon is run from 9 to short after 11 in the morning. Meteorologically spoken it's already 3 weeks Autumn.

The temperatures will be not perfect, but also not desastrous I think. The onset of temperature in our area (Holland and Germany close by) is primarely in the late afternoon. Theoretically mostly between 16 and 17h.

So Berlin 2009 basically will ressemble Dubai Jan 2008 where Geb didn't have his pace right and faded (2:05.29)after scaring the 30K Road WR by a second. Perhaps the other weather components won't be ideal either.

But one thing is more important:
Are Gebrselassie en Kibet themselves in absolute top form?

Ideal weather is very useful, but this fact by itself isn't a guarantee for a WR by a long shot.

Gebr (or/and Kibet) might even attempt the 30K WR and deliberately dismissing his/their attempt breaking the Marathon WR.

Anonymous said...

up on this , but since you say it quite often...Letsrun is NOT a reputable or excellent news source. I know they love you over there, but it is nothing other than a blog site that links to articles written by others and a forum for either trolls or people who don't respect each other's opinion.
They may of course be right in this case and others, but they are not a 'credible news source'...you could get as much or more if you bothered to google.

That said, I would have thought a bureau of meteorology would be where one gets this info coupled to some good local knowledge as given by the poster above.


Anonymous said...

First line should read:

"Sorry to take you up on this,...

Ross Tucker and Jonathan Dugas said...

HI Frans, Anonymous

Frans, thanks for that. We will see how it goes. I still believe that if temperatures rise to anything around 20 degrees, it's too hot for a record, especially if the humidity is a little on the high side as well.

Not disastrous, but to run 2:03:58, you don't need good weather, you need perfect weather. Then of course, as you point out, you need perfect form, perfect pacing, it's too good now to be broken in a 90% effort, it has to be better.

So we'll see on the course, I hope the temperatures stay in the mid teens until the runners are done, but I suspect it will be just too warm.

Then to anonymous, you're right about the forums, but if you want to get a one-page view of what is going on in the world of track and field, Letsrun is the best single site to do it. Yes, you could go on Google, but I'll give you an example. I didn't even realise the great North Run was happening today, until I logged onto Letsrun and found the headlines, together with a link to an article. So credit to the source, of course, but the collection of that news is done by Letsrun.

I could spend 10 minutes on google, trying to think of all the running related news I might find, or I could go to one site, see it all, and then read further.

Also, their own analysis of races is excellent - their Berlin coverage was better than anyone's, race previews, race reports and so on. So I think it's a terrific resource - first page I visit in the morning when I wake up.

The forums, I agree with you, a lot of insults. But also some very sharp folks there, who do understand the sport. Good with the bad, I suppose!


Frans Rutten said...

Start 14gr C. Not much wind.
Start at WR pace for Haile and Kibet together in 10K 29:15 and 15K 43:56. Projected temp end race: 17gr C.

Frans Rutten said...

30K Road WR for Kosgei and (forever collecting WR's) Gebrselassie in 1:27.49.

laxrippe said...

Hi there,

I actually raced in this marathon (my first) and even thought the temperatures kept on rising a lot after 11am, which obviously didn't affect Gebresselasie, it was pretty warm since maybe 10:30. After my marathon I realized just how much the temperature affects your output in this kind of event (I never felt really affected in a 10k). Considering that Gebresselasie himself said after the race that the heat was definitvely a factor, I can clearly see that this was exactly the reason why he didn't go faster, at least a sub 2:05.

I lost my struggle for a sub 4 at the moment when I could not longer evade stopping for water (at 37km), which became a literal stopping since hundreds of runners didn't drink while running anymore but simply stood at the water tables, blocking everything. You had to stop in order to get some water, and well, at this point my body didn't want to go anymore and feeling the pain, I had to subsequently adapt my running style. Well, 2010 is on the line. Thanx for the fantastic insights your blog provides.