Negative Split Marathon

Should you attempt a negative split marathon?

Is the negative split the best way to run a marathon? The data suggests not. It might be far better to accept some slowdown during the second half.

What is a negative split marathon?

Simply, a negative split marathon means running the second half faster than the first. It is commonly recommended on coaching and running websites to attempt to run the second half of a marathon faster than the first.

According to an article on Runners World ‘Anyone can and should run negative splits’. The article continues, ‘If you conserve your resources during the early part of a run, they’ll be available to you at the end’.

Is running a negative split the best approach to the marathon?

If we are to believe what the majority of marathon advice says then yes, the most effective way to tackle the marathon is by aiming to run the second half faster than the first; the negative split. This advice comes from coaches all the way down to even beginner runners who have picked up their ‘knowledge’ from the web.

Often the advice is justified on the basis of world records, for example, Eliud Kipchoge’s astounding performance at the 2018 Berlin Marathon. His 2.01.39 was achieved with a 1st half of 1.01.06 and a 2nd half of 1.00.06; a negative split of 33 seconds. Coincidentally, that 33-second negative split is exactly the same as Dennis Kimetto’s previous world record of  2.02.57 set at Berlin 4 years before.

That argument supported by world records in favour of a negative split marathon does not however stand up to scrutiny after Eliud Kipchoge’s record-breaking run at the 2022 Berlin Marathon. His time of 2.1.09 took 30 seconds off his previous world record.
The half-way split was interesting, the first half he ran in 59.51, the second half in 1.01.18. A positive split of 1.27.
Indeed his pace during the first half was near metronomic: varying between 14.09-14.14 per 5k. Compare that to his 5k splits between 30km and 40km: 14.30 and 14:43. A significant difference of  6 seconds per km (10 secs per mile) especially considering that 30-35km are downhill.

Nevertheless, it’s quite common knowledge amongst runners that the negative split is the way to go.

There is, of course, plenty of ‘knowledge’ on the web that has become so quoted and repeated that it’s accepted. For example, it takes 21 days to form new habits; alligators living in New York sewers; etc, etc.

But, what do the results of marathons tell us?

Let’s look at some figures.

First London Marathon 2017.

There were 31 elite men who finished the race, how many ran a negative split?

Zero. On average, the elite men ran the second half 6m 6s slower than the first.

Granted the London marathon does have a slightly downhill start, but it’s negligible; the winner, Daniel Wanjiru ran his first 5k 28 seconds faster than his second.

One of the most surprising results was achieved by Josh Griffiths. He wasn’t even in the elite field, yet he came 13th and was the first British runner home.
According to the newspapers, at half way ‘he started passing athletes in the men’s elite race’. For the first half he ran an average of 3m 9s, for the second half he ran 3m 13s; a positive split of 1m 29s. His pace for kms 35-40 was 3m 17s. So, even though he was passing other highly talented and experienced runners, he was slowing down.

Of the top 1000 finishers in the men’s race, 88 ran a negative split — less than 9%. More than 90% of the fastest runners in the race ran the second half slower than the first.

What about the women?

There were 26 elite women who finished the race, none of these ran a negative split.

The winner, Mary Keitany, ran a positive split of 3m 13s — it was a world record for a women-only race. So a negative split is certainly not a requirement for a world record.

Fukuoka Marathon

Fukuoka marathon is a very high quality race. It does not have a large field; the qualifying time is 2h 35m.
The course is flat and fast with 2 slight climbs in the first half of about 5 metres. All runners who run in this race are highly accomplished marathon runners. 

In 2016, the top 100 finishers ran under 2h 30m.
Of those, 3 runners ran a negative split. On average, the top 100 finishers ran the second half 3m 7s slower than the first half. 97% of the top 100 runners achieved their result without running a negative split.

Dubai Marathon

Dubai is another flat, fast marathon. The field is not as large as other big city marathons, but it does attract some very fast runners, probably due to the offer of $1 million for a world record.

In 2016, of the top 100 runners, 14 ran negative splits. The average difference between 1st and 2nd half times was 8m 14s.

London Marathon 2023

If there was ever an argument to recommend the negative split then surely Kelvin Kiptum’s extraordinary run proves the point. Kiptum’s time was just 16 seconds outside the world record and a course record for London. He ran the first half in 1:01:40 and the second in a remarkable 59:45.

Obviously, he won the race, so should you try to emulate him to run your best time ever? If we look a little deeper into the results we see that his run was exceptional not just because of the record-breaking time. Of the 24 elite male finishers, there were just 2 who ran negative splits: winner Kiptum and British runner Phil Sesemann who placed 8th and ran the second half 3 seconds quicker than the first. So, a little less than 10% of the elite men ran a negative split.

What about the women’s London Marathon results?

Another extraordinary performance by the winner, Sifan Hassan. The commentators had written her off in the early part of the race when she was clearly suffering from injury and dropping off the lead group. But she later fought back and finally out-kicked Alemu Megertu to cross the line in 2:18:33. A stunning come-back run, but even though she ran the fastest second half, she did not run a negative split. Her second half was 1:13 slower than her first. Not surprisingly, she wasn’t alone in recording a positive split. All of the elite women ran the second half slower than the first.

If just 2 of the elite men and women runners at London in 2023 ran negative splits, is it a realistic goal?

Should you attempt to run a negative split marathon?

That’s what most coaches and books advise. Even pundits on the TV, themselves often experienced marathon runners, will say that it’s the best way to run a marathon. Why? The results clearly show it isn’t.

Maybe it’s the notion that to run an effective marathon one must keep the pace high till the end. For sure, there is a lot of sense in the saying ‘pinch a minute from the first half and you’ll pay more than double that in the second’. Most of us who have run more than a few marathons will know what it’s like to overcook the first half. Often dramatic slowdowns face runners who set off even just a little too quick. However, a slow down in the second half does not necessarily indicate a falling apart.

Running a negative split is so unusual — amongst all standards of runner — that aiming for it almost sets us up for failure.

Why jeopardise all that training and effort in trying to achieve something that is not only highly unlikely, but also arguably of no benefit? Slowing down in a marathon, when you have not planned for it, can be soul-destroying, and that’s the last thing you want when undertaking such a mentally and physically demanding event as the marathon.

It’s entirely natural to slow down during a marathon

As we run a marathon the body undergoes huge changes. These changes affect our ability to keep moving at a constant speed. We are likely to be dehydrating, heating up and of course, our joints, muscles, and the whole body begin to hurt more and more. Not only does it feel like it’s getting harder to maintain a pace, it actually requires more physical effort. The heart is required to beat faster just to keep a constant pace.

Don’t underestimate the psychological effect

Anyone who has run a bad marathon knows what it’s like when things stop going according to plan. Once the head drops in a marathon, what’s left of the task can seem overwhelming. One of the keys to successful marathon running is maintaining a positive mindset throughout the race.

There’s plenty we can do to help with this. Firstly: training. If we’ve put the training in and have completed successful long training runs then this provides good mental fuel for a race effort. Long runs produce ups and downs. One minute we can be thinking how wonderful we feel, how easy this long-distance running is. The next, everything is turned on it’s head. For no apparent reason our task can seem impossibly hard. Then again (sometimes) we feel great again. These are the normal highs and lows of marathon training and racing. The more often we experience them, the more often we are able to deal with them. What usually floors us is when we feel unexpectedly bad, or when it becomes clear that our goal is not going to be achieved. Once we reach this negative state, the whole exercise can seem pointless, even jeopardising our finisher’s medal.

As we’ve already seen, a negative split is quite elusive (and not even necessary) and if our goal is dependent on it, we risk the demoralising effect of a marathon effort falling apart.

Now, think for a moment that you are surrounded by runners who are experiencing this disillusionment, they are comparatively going backwards whilst you are thinking, ‘hey, this is what happens, it’s part of my plan and I’m doing just fine’. Who would you rather be?

What marathon split should you be targeting?

It should be clear by now that may be a negative split it not worth chasing. But what are the alternatives? We should expect a slowdown. But, what we don’t want is a massive slowdown. This indicates a thoroughly miserable run. A run in which the first half of the marathon has left us so depleted that running the second half (or at least the last few miles) becomes a trail of survival rather than one of decent running. Large positive splits are usually the result of poor training, injury, and/or a misjudgement of pace. However, many of the fastest times ever run and the vast majority of marathon finishes feature a positive split. the very best runners appear to run their marathons with a small positive split; by small I mean single-figure percentage.

Expect a slowdown. It is highly likely that’s what you will get, and it does not determine the success or otherwise of your marathon result. However, we are all different and what works for me or you might not work for others. Nevertheless, we’ve seen that a positive split works for many, many people, even world record breakers. I think it’s time to ditch the idea that a negative split is the holy grail of marathon running.

Further reading

An alternative and excellent view specifically regarding negative splits at Chicago and New York marathons is on the fellrnr website, there’s lots of data and this little nugget in the conclusions ‘Most runners have their best performance with a slight positive split’. : Are negative splits faster in the marathon? An analysis

There is also some very detailed writing on the subject by ultra runner Stuart Mills. Whilst it seems rather extreme, this man is a very successful distance athlete and also, significantly, a sports scientist and principal lecturer at Brighton University. His article Run as fast as you can, while you can. is extremely thought-provoking and well worth reading.

What are your experiences?

Let me know in the comments section. How have your marathon runs panned out with regards to pacing and outcome?