Negative split: definition, formula, example

A negative split means running the second half of a race faster than the first. The opposite, a positive split, is what most people run without meaning to: a start carried by adrenaline and a fresh crowd, then a long fade. Coaches have recommended the negative split for decades, mostly as a cure for that fade. The evidence behind it is thinner than the advice suggests, but the habit it builds, holding back early, is sound for anything from 10K upward.

How a split is measured

Cut the race in two equal distances and compare the times. Expressing the gap as a percentage of the first half lets you compare a 10K with a marathon:

The ±1% band for “even” is a practical convention, not a standard: GPS drift, wind and a single hill can move a split by that much on their own. On a hilly or point-to-point course, compare halves of similar profile, or look at the grade-adjusted pace rather than the clock.

A worked example

A runner aims for 50:00 over 10 km, an average of 5:00/km. They pass 5 km in 25:30 (5:06/km) and run the second 5 km in 24:30 (4:54/km).

• T2 − T1 = 1,470 − 1,530 = −60 s
• split = −60 / 1,530 × 100 = −3.9%

That is a clear negative split, and probably a bit too clear: finishing a minute faster on the second half of a 10K usually means the first half was conservative and some time was left out there.

Over a marathon the margins are smaller. For 3:30:00 (4:59/km average), a sensible plan is 1:45:50 at halfway (5:01/km), then 1:44:10 (4:56/km) for the second half: a split of −1.6%. Running the first half only two or three seconds per kilometre slower than the average costs 50 seconds at halfway, and buys some margin for the last 10 km, where most marathons are lost.

How to use it

The point is not to sprint the end but to avoid paying interest on a fast start. Running the first kilometres above your sustainable pace burns through carbohydrate and tolerance to fatigue that you cannot recover later in the race. A few practical rules:

• Set your target pace from a recent race, not a hope. The race predictor and the pace charts give splits per kilometre for a given finish time.
• Plan the first 3 to 5 km at 5 to 10 seconds per kilometre slower than goal pace in a half or a marathon, then settle. In a 10K, the first kilometre is the one to restrain.
• Rehearse it. A progression run, such as 12 km starting at easy pace and finishing the last 3 km at marathon pace, teaches you what restraint feels like. A long run of 26 km with the last 8 km at goal marathon pace does the same with more fatigue.
• Judge the race afterwards by its split. A positive split above 3 or 4% in a marathon almost always points to the first half, not the second.

Limits

The research does not make the negative split a law. In their review of pacing strategies, Abbiss and Laursen (2008) found that evenly distributed pace tends to suit events longer than a couple of minutes, and that in very long events trained athletes usually slow progressively, whatever their intentions. Díaz and colleagues (2018) analysed every marathon world record over fifty years: on average, the record holders ran the first half slightly faster than the second. The more recent records were the most even, which led the authors to suggest that very little speed variation, rather than a marked negative split, may be the way forward.

So the honest target for most runners is an even split with a slight negative lean, not a dramatic finish. Courses also get in the way: a race that climbs in its second half will look positive on the clock while the effort was perfectly even. On trails and in ultras, where terrain and night dominate the second half, comparing halves by time tells you little; effort, heart rate and fuelling matter more.