Negative splits: the pacing strategy that wins races

Two runners line up with identical fitness. One charges out hard, banks time early, and then slowly unravels over the closing miles. The other holds back, runs each section at the same controlled effort, and reels in the first runner near the finish. The second runner almost always wins. The difference is not talent or training. It is pacing.

Pacing strategy describes how you distribute your energy across a race, and it is one of the most studied questions in endurance science. This article explains what negative, even and positive splits are, what the research says about which works best, why going out too fast is so punishing, how the brain regulates pace before you ever feel tired, and how to build a pacing plan you can actually execute.

Negative, even and positive splits

A “split” is simply your time for a section of the race. Comparing the first half with the second half gives you the three classic pacing patterns.

A negative split means the second half is faster than the first. An even split means both halves take roughly the same time. A positive split means the second half is slower, usually because the opening was too aggressive. Abbiss and Laursen (2008), in their landmark review of competitive pacing, catalogued these alongside all out, parabolic and variable patterns, and drew a clear line: very short, sprint duration efforts reward an all out start, but once an event lasts longer than about two minutes, performance improves when effort is distributed more evenly.

That two minute threshold is the key. Almost every race a distance runner cares about, from the 5 km to the marathon, sits firmly in the territory where even or slightly negative pacing is the winning approach.

The evidence for even and negative pacing

The cleanest experimental evidence comes from controlled time trials. Foster and colleagues (1993) had well trained cyclists ride a series of trials in which the pace of the first half was deliberately constrained, from very slow to very fast. Even pacing produced the best times, and even modest deviations from it carried a measurable cost. A fast start, in particular, defeated its own purpose.

The pattern holds at the very top of the sport. Tucker, Lambert and Noakes (2006) analysed the lap by lap structure of men’s world records in track athletics and found that, in the 5000 m and 10,000 m, the middle kilometres were run at an essentially even pace, bookended by a quick opening and a fast finish. World records are not set with reckless fast starts; they are set with disciplined even effort and a closing surge.

In the marathon, Hanley (2016) studied pacing and “packing” across Olympic and World Championship races and found that medallists of both sexes settled into even pace early and held it, while slower finishers fell off the lead group around halfway. Notably, women slowed less than men and were more likely to run a genuine negative split. Angus (2014), examining two marathon world records split by split, likewise found the fastest performances clustered around even effort rather than a fast start strategy.

Why going out too fast is the costly mistake

If even pacing is so clearly superior, why do so many runners still go out too fast? The answer is partly physiological and partly psychological. At the start you are fresh, adrenaline is high, and goal pace feels absurdly easy. Banking a cushion of time feels like the smart, safe move. It is the opposite.

A fast start has a steep metabolic price. Grivas (2025), reviewing the physiology and psychology of negative splitting, describes how a conservative early pace spares muscle glycogen, blunts the early rise in blood lactate, and limits cardiovascular drift and the climb in core temperature. Running the first half too hard does the reverse. You deplete glycogen prematurely, accumulate lactate and metabolic byproducts early, and arrive at the closing stages with neither the fuel nor the freshness to hold pace. The result is the dreaded “slow death” of the positive split, a grinding deceleration that often costs far more time than the fast start ever banked.

This is not a rare error. March and colleagues (2011) analysed hundreds of marathon runners and found that age, sex and finishing time all predicted how well someone paced: older, female and faster runners held even pace best, while younger, male and slower runners were the most likely to fade badly in the second half. Across mass participation fields, the positive split is the default amateur mistake, not the exception.

Your brain is pacing you before you feel tired

Pacing is not just a matter of willpower or fuel. It is actively governed by the brain. Tucker (2009) set out the case for anticipatory regulation, sometimes called teleoanticipation: from the very first stride, the brain sets your effort with reference to how far you still have to go. It continuously compares a subconscious template of how hard you should feel at this point against the actual feedback streaming in from your muscles, heart and lungs, and it adjusts how much muscle it is willing to recruit to protect you from blowing up before the finish.

This explains a great deal about pacing in real races. The familiar end spurt, that final acceleration in the last few hundred metres, happens because the brain senses the finish is near and judges it safe to release the reserve it has been protecting. It also explains why a fast start is so dangerous. Push the early pace beyond what the template expects, and the regulator responds by dialling back muscle recruitment later, whether you consciously want it to or not. You cannot simply override the slowdown with grit. The sensible move is to set out at a pace the system will sustain.

A practical pacing plan

Translating the science into a race plan is straightforward. The aim for nearly every runner is even pacing with a slight negative split, achieved by holding back at the start.

Pacing accurately also depends on having a realistic goal pace in the first place. A plan built around a target you cannot sustain will always end in a positive split. Learn how accurate race time predictors really are before you commit to a number, and to arrive on the start line fresh by getting your taper right. On longer races, fading in the heat is often a fuelling and fluid problem as much as a pacing one, so plan your hydration strategy alongside your splits.

The honest summary

Across controlled trials, world records and mass participation fields, the message barely wavers. Even pacing with a slight negative split is the strategy that wins races, and a fast start is the most common and most expensive way to lose one. A truly large negative split is usually a sign that the first half was simply too slow, so the realistic target for almost everyone is even effort with something held back for the finish.

None of this requires a faster engine. It requires the discipline to run the opening miles slower than your instincts demand, trusting that the reward arrives at the end. The runner who masters that, more often than not, is the one crossing the line first.

Frequently asked questions

Related reading: how accurate are race time predictors, really?.