Should you run a marathon with negative splits?

A slightly negative or even split is the optimal race strategy on a flat course. On a hilly one, even effort matters more than even pace, and the split will follow the terrain.

A negative split means running the second half faster than the first. On flat courses, almost every well-executed marathon is close to even or slightly negative, and almost every disaster is heavily positive. The physiological reasoning is straightforward: starting above sustainable pace spends glycogen faster and accumulates fatigue that you cannot repay later.

The practical version is more modest than the ideal. Aiming for a genuinely negative split often produces a first half so conservative that the time is lost rather than banked. Most experienced marathoners aim for even effort with a first half a shade slower than the second, which lands close to even overall.

On a course with meaningful elevation, the calculus changes. The correct target is even effort, not even pace, and if the climbing is front-loaded your second half will be faster with no change in effort — while a course that saves its hills for the end will produce a positive split that is not a pacing error at all. That is exactly what a grade-adjusted split table tells you before the race.

The other reason to start conservatively is congestion. In a large field the opening kilometres are slow whether you like them or not, and fighting that costs far more energy than the seconds it recovers.

Glossary: GAP (Grade-Adjusted Pace)