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.

Sunday, April 11, 2010

Paris Marathon splits and analysis

Paris Marathon splits and analysis

Tadese Tola has won the 2010 Paris Marathon in 2:06:40, in a good start to the Spring racing season.  It represents a massive 9:05 improvement for a man who before this race had a best time of 2:15:45.  Tola's move came at the end of what was a fascinating race, which saw a relatively slow start (around 2:07 pace for the first 15km), a super-fast middle segment, then a tactical battle with an elite group that had a dozen athletes all the way to 35km, before Tola managed to pull away in the attritional final 5 km.

The race was full of changes in pace.  First a surge at 15km brought them to half-way in 63 minutes, and the course record of 2:05:47 was on.  However, with about twenty athletes in the lead group, the third quarter was always going to tell, particularly when the pace-maker dropped out.  That happened at 30km and suddenly, the pace slowed.  The section between 30km and 35km was covered in 15:31, which saw any chances of records disappear altogether.  At that point, five or six men ran abreast, no one wanting to take the pace decisively.

The moves eventually began at about 35km.  It was the marathon debutant Wilson Kipsang who went first, and the group was cut by half.  Then, with about 5km to run, Alfred Kering of Kenya threw in the first real decisive move.  Only Tola could respond, and the race was down to two.  The section from 35 to 40km would be covered in 14:48, a substantial increase in pace from the previous 5km.

Tola proved stronger and he gradually began to eke out a lead over Kering, eventually winning by the relatively comfortable margin of over 30 seconds.  Kering was almost caught by Kipsang, but hung on for a 2 minute improvement in his own PB.

The 2009 Paris race was remarkable for its depth - six men broke 2:07, and eleven went under 2:09.  This year, eight sub-2:09 performances were recorded, but only one man went under 2:07, and so the overall quality was not as high.  It was, however, a great race and a good start to what should be a fantastic few weeks of marathon running.

The splits and projected times are below, as well as my thoughts throughout the race.

Summary split times

Below are the split times from the race.  They sum up the race at a glance.  Below the tables follows some thoughts put down in real time, as the race developed, for those hungry for more detail!

As it unfolded...random musings, in real time

5 km  -  15:07 (Projected time 2:07:34)

A relatively slow start.  The conditions seem very good for running, and as you'd expect, the lead group is enormous this early on.  The pace did pick up at 5km, suggesting that the requested pace is substantially faster than what they're currently on.  I expect that to pick up a good deal between 5 and 10.

So far there have been a lot of aerial shots of various landmarks of Paris, and little of the elite runners.  Not that Paris doesn't look great from the air - it is my favourite European city, but I would enjoy a little more actual marathon action!

10 km - 30:11 (last 5km 15:04, with projected time of 2:07:22)

Well, it hasn't really picked up a great deal.  The lead group is still huge - I'd guess that there are forty runners, tightly bunched.  Last year in Paris, three men broke 2:06:30, and the winner was a 2:05:47 (Vincent Kipruto, running in Rotterdam this year).  It seems unlikely that 2:06 will be broken this year, and even 2:07, unless things pick up over the middle of the race.  All the pre-race favourites are there, including Joshua Chelanga, James Rotich, and Benjamin Kiptoo of Kenya.  All have PBs under 2:08, which is of course highly credible, but in the current era of 2:05 competitive races, fall slightly short of the 'platinum standard' that we'll see later this season.  Rotterdam promises to deliver the first heavy artillery later today (join us later for splits on that race)

15 km - 45:08 (last 5km 14:57, with a projected time of 2:06:58)

The lead group is now considerably smaller and thinner, and the pace has picked up, now projecting a sub-2:07.  Just over twenty men are in the group, but it is strung out considerably more than the bunch from earlier, and within the next ten kilometers, will likely drop to a dozen, before the real racing begins at 25km.

There have now been two falls in the elite group, both around the water points.  The athletes have twice dived over to the water points and cut across runners behind, leading one to come down.  Both have returned to the race and remain in the lead group, but the impact of the fall is usually quite substantial - remember, they are traveling 20 km/hour - a bicycle accident at that speed is not insignificant.  The falls have happened because of the size of the group.

Halfway - 1:03:00 (last 6.1km 17:52 (14:39 per 5km pace), with a projected time of 2:06)

The pace has really picked up now - they didn't show a 20km split, but the halfway point was reached in 63 minutes.  That projects a 2:06:00 finish, which is now substantially faster than it had been.

That comes thanks to a very fast section - the interval between 15km and 21km was covered at a pace of 2:56 per kilometer (14:40 per 5km).  Not surprisingly, runners are beginning to peel off, though the group is still large, which bodes well for a fast time.

Can they maintain this pace?  With such a large group, it seems likely that at least a few could.  The telling part of the race comes now - the third-quarter, which is usually where the pace drops off.  It's also where the pace-maker is likely to step off, probably at 30km.  If this pace of 14:50/5km can be maintained up to 30km, then the course record is certainly within sight.  After a slow start, that seemed unlikely.  It promises to be intriguing.

25 km - 1:14:58 (last interval at 15:20 5km pace, with a projected time of 2:06:32)

The pace has slowed considerably - the last 3.1 km were covered at 3:04/km, which is the slowest segment of the race.  Perhaps not surprisingly, that drop stimulated a sudden surge at 25 km from one of the pacemakers and Benjamin Kiptoo (winner in Rome last year).  The field was split for a short time, but has largely regrouped.  They're running past the Notre Dame Cathedral, which is garnering significantly more air time than the race (it is an amazing building), but the race seems to have heated up at the front.

30 km - 1:29:50 (last 5km 14:52, with a projected time of 2:06:21)

The lead group is now down to a dozen or so.  The pacemaker stepped aside at the 30km banner, and allowed Benjamin Kiptoo, who was paying closest attention, to move into a temporary lead.  The group is thinning out, many athletes are hanging on at the back, including a couple of the race's sub 2:08 men.  The fast pace between 15km and halfway is beginning to tell, because the last ten kilometers have not as aggressive (2:07:13 pace).

The acid test comes now that the pace-maker has gone.  With still a dozen guys in contention, the odds are that no one wants to pull them along to the finish, and so I suspect we might see quite a slow interval up to 5km.  Perhaps then the pace will lift, but too many men may slow things down a little before a final assault for the finish.  You can see that this already happened because the shape of that lead group changed within about 100m of the pace-maker stepping aside.  Having been an elongated "missile", it now spans five men across.

The helicopter is working over-time, this time showing the Eiffel Tower and Trocadero, while the runners hit the 30km point.  At this stage, I'd guess the proportion of race coverage to city coverage has been about 0.8:1.

35 km - 1:45:21 (last 5km 15:31, with a projected time of 2:07:00)

There are still seven men in the lead group, which means this race over the final 5km is going to be very aggressive.  As expected, the pace has dropped considerably - the slowest 5km split of the race.  That was, as suggested above, down to men looking at one another and biding their time after the pace-makers had gone.  The last split might have been even slower but for the the last few minutes, where it has picked up to cause the splitting of the lead group from twelve to seven men.

At about 34km, Wilson Kipsang of Kenya went to the front, and returned the lead group to its "missile" formation, albeit for a short time.  Kipsang is the race's intriguing character - a marathon debutant, his half-marathon best of 58:52 places him 10 th on the all-time list.  Whenever a really fast half-marathon athlete steps up, we get excited.  Kipsang moving to the front is either a sign that he feels great and is confident, or its a sign of his inexperience, for which he'll pay later.

40 km - 2:00:09 (last 5km 14:48, with a projected time of 2:06:45)

With 5km to go, the first decisive move of the race was made, and five men were cut to two by an attack by Alfred Kering of Kenya.  He was followed closely by Tadese Tola of Ethiopia.  Kipsang hung on gamely in third, but the gap opened quickly and by 40km, he was well back.  The damage in this segment is reflected by the 5km time - 14:45, very fast at the back end of the race, particularly when you look at it alongside the 15:31 of the previous 5km.  As expected, a very aggressive end to the race!

These men are running well above themselves - Tola, for example, was a 2:15:45 man before today, and is running over 8 minutes faster right now.  Alfred Kering has a few top 10 finishes in the Marathon Majors, incuding a 5th in Berlin and a 6th in Rotterdam, but his best time is "only" 2:09, which means he too is in unchartered territory here.

Tola hit the front with 3.7 km to go.  It wasn't an explosion, but it was effective as he gradually opened up a stride, then two, then three, and with 2km to go, it was a ten second lead.  That is never decisive, but it seems unlikely that Kering with return.

Finish line - 2:06:40

Tola sprinted across the line, looking very strong indeed.  An improvement of over 9 minutes a remarkable sub-plot to the race, as Tola reclaims the Paris title for Ethiopia (Kenya and Ethiopia have practically alternated wins in Paris since about 2003).

Second went to Kering, who hung on, eventually finishing in 2:07:10, thirty seconds back.  In third, only 2 seconds behind, was Wilson Kipsang, followed by a host of athletes running big PBs, as one would expect in a race where that many men were still together with only 7 km to run.  In all, eight athletes ran 2:09 or faster, which is not quite as impressive as last year, where six men went sub 2:07 and eleven broke 2:09. 

The women's race - Bayisa sets course record

The women's race was highlighted by a new course record, as Ethiopia's Atsede Bayisa defended her 2009 title in a new PB and Paris record of 2:22:02.  The race itself was a non-entity - Bayisa was always in control, and the real highlight was provided by the strong performance of Christelle Daunay, who broke the national record with an exceptionally strong finish and time of 2:24:22.

So the times were fast, as we've come to expect of marathon running.  There were PBs all over the place, in both the men's and women's fields.  It suggests that Paris is a great feeder for the big marathons, and that athletes on the fringe can make their breakthroughs in the French capital.



DrTim said...

Great stuff. 2:06 to start the season is pretty ominous.

Rotterdam is running but not being televised on Universal Sports till 12pm ET ... that's like 5 hours away.


Ross, you go any other feeds?


Ross Tucker and Jonathan Dugas said...

Hi Tim

Yup, good start.

All I have on Rotterdam is this:

And then here is a live feed, but my internet provider is too slow to give me decent quality:


I'm trying to find more


Ross Tucker and Jonathan Dugas said...

Hi again Tim

I see the race has just ended - 2:04:46ish.

So it was quick, very quick, but from what I can gather, it slowed quite a lot in the second half.

They were on course for a 2:03:40 at the 20km point, and then the next 10km were covered in 30:18, dropping them back a good deal.

The final 12 or so seem to have been covered in 2:56 per km, which means it picked up.

Reading that thread on letsrun, it seems there were some major shifts in pace, but that the really fast (low 2:04) time was lost between 20 and 30km.

Be good to see the race later, though I fear with this internet speed, I won't be!


DrTim said...

Thanks Ross.

Will watch with a heavy heart at 5pm UK time as I posted my withdrawal form for London yesterday. Quite sad but with strained hamstring and left plantar ligament I didn't really have a choice. Will watch my sub-3 hour body fade away or bloat up as the case may be :)

Any analysis you've done on the masters or golf in general that may convince my wife it's a sport? Her definition, no sweat = no sport!


DrTim said...
This comment has been removed by the author.
DrTim said...

Actually, sorry that article is bollocks :)

What I'd be curious to know is if there is any scientific data out there on the physical control / motor skills required to control a 1.5m club moving at 200 kph to then hit a ball to 1000 feet to within a few feet of where you were aiming.

But we can save that for another day ...

Gene said...

A full replay is supposed to be at http://www.universalsports.com/video/assetid=4dd634cc-3817-4f36-8390-9d33eaced692.html#watch+full+race+replay, but the video says "coverage will begin shortly" (6pm MT).

Seb said...

If you want individual split times, you can search by name or bib #