Professor Marathon: What the London Marathon results teach us about pacing

About 3 miles into the 2012 London Marathon I came across a girl called Kayla who I had run most of a half-marathon with about 2 months previously. I asked her what her race plan was. Her answer was interesting:

“I plan to do the first half in 9 minute miles, then step up to 8.30 minutes miles until the last 10k and then try to give it a big finish.”

I remember thinking at the time what a brave plan it was and I would be impressed if she did it. We split up at mile 13 as she picked up the pace and I stopped for 10 seconds to high-five my family and friends standing at the side of the road.

Being older and more conservative, I planned to try to run the whole race at exactly the same pace – 9 minutes a mile. As things turned out neither of us succeeded in our aim that day, but interestingly I caught and overtook Kayla at 23 miles. She had completely blown up and was running at little more than a shuffle. I would love to tell you that I strode past her confidently but in truth I was just shuffling faster than she was.

This story is a great illustration of the perils of marathon pacing and is worth a closer look. I am not a coach so I am going to use some facts from the results of last year’s London Marathon to illustrate a point.

If you divide the marathon in two and measure your splits as two half marathons then you have three possible outcomes. You can run the first half of the marathon faster than the second – this is a positive split. You can run the second half faster than the first – a negative split or do them in the same time – a neutral split.

Kayla’s race plan would have given her a negative split. It is what everyone dreams of. Use the first half marathon to find your feet and then roar home during the second half. The negative split is the Holy Grail – even for the elites. But when you are thinking about your marathon pace, let history and experience be your guide. Statistically a negative split is very difficult to achieve.

In an effort to gain some insight into marathon splits, I have laid my hands on a spreadsheet showing the results for all 36,665 finishers of last year’s London Marathon. I have analysed the results of everyone who finished in between 3 hours and 5 hours. In total that was 24,927 people.

First let’s look at how common a negative split was. Of these just 1,151 managed to run the second half faster than the first. The other 23,776 ran the second half slower. So just one in 20 people managed a negative split.

In the London Marathon as a whole 1,383 runners ran a negative split compared to 35,282 runners who ran the second half slower than the first.

Second let’s look at the uglier side of that statistic and analyse the number of people in the same group who hit the wall, many of whom may well have done this by trying to run a negative split. I have counted anyone who took more than 30 minutes longer to run the second half than the first as having hit the wall. To put this in perspective, Kayla in her completely exhausted state, ran the second half just under 10 minutes slower than the first so would not have hit the wall according to my definition. This will give you an idea of how dramatically slower a 30 minute difference is. 3,314 poor runners in the 3-5 hour bracket fell victim to the wall, meaning that three times more people hit the wall than ran a negative split. This might not be just because they got their pacing wrong – poor nutrition could also cause this – but pacing will surely have contributed.

This just confirms what I experienced which is that the second half of the marathon is much harder than the first. In particular the last 6-8 miles is tough. Someone once said that a marathon is a 20-mile prelude to the hardest 10k you will ever run and that is about right.

If you are running a Spring marathon, before you decide on your pace strategy and contemplate attempting a negative split don’t listen to me listen to yourself. Ask yourself this question and answer it honestly: “How am I going to manage what 35,282 people in last year’s London Marathon, including all of the top runners, couldn’t?”

Whatever your race plan – good luck?


13 responses to “Professor Marathon: What the London Marathon results teach us about pacing

  1. There is a huge banner at Mile 20 of the Dallas Marathon that says START. That’s when the marathon truly begins. Negative splits are very difficult to pull off in long distance races, as you proved. Most people are lucky just to keep an even pace all the way to the end.


  2. Ah I do love a good statistical analysis. Sadly I wasn’t included in your numbers having finished in 6.28, walking the last 10miles with my running buddy who’d strained his calf. Many people said I should have left him, I know I did the right thing by staying with him.

    I know a lot of people who ran marathons last year and only one that kept anywhere near to their pace plan. It’s such a long way and so many things can happen. Negative splitting is indeed the Holy Grail.


  3. It was my first marathon so I was on for a PB whatever my time. Going back next year to smash it and would like a consistent-ish pace all the way round. Realistically will probably be slower in the second half but would like to not crash and burn!


  4. Absolutely true! Second half is harder and the last 10 km is when the real race begins, both physically and mentally. Negative split is a great thing to have, but just finishing your marathon strong is a wonderful achievement!


  5. Goodness, as a first marathoner having completed 20 miles last wkd en route to the Ldn marathon I was feeling very optimistic till I read this!

    Some really comprehensive statistical analysis you’ve done there Peter, interesting stuff! I was thinking it very unlikely I would manage a negative split but though I might be able to maintain a neutral if I run slower than I like in the first half. I guess the question is: should you run deliberately slow in the first half in an attempt to make a negative split or are the odds so stacked against you being able to do this that you’re actually losing unnecessary time when you’re at your freshest?


    • Thank you for your comment Emma. My last marathon (London 2012) was my first and so far only marathon so I speak with limited experience. However if I had my time again I would ignore the positive/negative split concept as it is really a false measure. Rather than break the race into two halves I would break it down into 6 or 7 segments and assume I will get slower in each one. But the first one will be the slowest because of the sheer number of people around you. Don’t worry about the distance, if you can run 20 miles you can run 26!! Apart from that, drink a lot, take on fuel regularly and enjoy the wonderful crowds. They are the real highlight of the day. You will have a day that you will remember forever. Good luck. I wish I was doing it again!!!


  6. Pingback: Marathon Data, Part 1: A Matter of Time(s) | spreadsheetjournalism

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s