Haile Gebrselassie Dubai splits

Haile Gebrselassie's splits from Dubai

Following on from yesterday, here are the splits from Haile Gebrselassie's 2:05:29 in Dubai. Thanks to Ray for pointing out the LetsRun thread with some splits, but especially to Sean Hartnett, who kindly provided these detailed times from the race (Sean was responsible for the "RaceTracker", which was featured before the race in an IAAF article. Thanks Sean, hope your food poisoning has been conquered!)

The graph shows the splits from Dubai, compared directly to those from Berlin last year, when the 2:03:59 record was set. I've indicated the pace required for the world record with the dashed purple line, but bear in mind that the numbers you see are a direct comparison between the races.

The boxes at the bottom are the comparison. Dubai is shown in blue, Berlin in Orange, and then the green and red boxes show the time difference in the 5km intervals between the two (green when Geb is ahead of the record from Berlin, red when he is behind). At the top of the graph I've shown the overall gap between Berlin and Dubai. (you might need to click on the graph to enlarge - had to squeeze information in)


Brief analysis

It's very apparent what transpired in Dubai - the record was on up to about 30 km, and then the final 10km cost him enormous time. The weather reportedly worsened at about 35km, which would seemingly track the gradual slowing down. Bear in mind that in Berlin, Gebrselassie finished astonishingly fast - his final 10km were covered in 29:09! So given that he was 8 seconds faster than Berlin at the 30km mark, he still required an exceptional finish in Dubai.

In fact, he needed to speed up - his time at 32km was 1:34:17, which meant that he needed to run the final 10.2 km in 29:43, which is a pace of 2:54.8/km. That is the SAME SPEED as he finished with in Berlin (a 29:08 10km), and is FASTER than he'd been going up to that point (2:56/km), even without the weather complication.

Therefore, the weather ultimately put paid to any hopes that may have existed. However, as I wrote yesterday, I don't think it would be correct to say "the weather denied him the record". That assumes that the record was a guarantee, and given that he need to at least match the spectacular finish in Berlin, the outcome in Dubai was anything but certain. Very importantly, he needed it without any company (the pacemakers were gone by then. Berlin was a little different), and that would have been tough, even on a good day. So what the weather did is deny him the CHANCE - he had put himself in an excellent position, equal to Berlin, but he still needed a mighty good last 10km. I guess we'll never know.

One other observation - the early pace in Dubai was quite conservative - 14:50 for the first 5km, and then it really started to pick up. That slower than in Berlin (which was too quick), and different from previous years. I can't say that's down to anything. The other amazing thing is the consistency of pacing between about 10km and 26km. Apart from one 'aberration' at 22km, he reeled off 16 consecutive kilometers within two seconds of 2:55. Quite amazing.

So that's the splits, as promised. The rest of the week holds the promise of some more (hopefully) interesting discussion of physiology of performance. Also, the Australian Open, the year's first Grand Slam, starts tomorrow, and might warrant a post, if I can get my hands on some interesting data I once saw regarding it!

Join us then!
Ross