How to run negative splits (and why most people don't)
Most recreational runners positive split their races. You feel great at the start, adrenaline carries you through the first few miles at a pace your body can’t sustain, and then physics catches up. The second half takes far longer than the first, and you finish slower than you should have.
A negative split is the opposite. Your second half is faster than your first. It requires deliberate restraint early and confidence that the speed will come later. Almost every distance running world record has been set with even or negative splits, and there’s good reason for that.
The physiology of why positive splits are so costly
Running faster than your sustainable pace doesn’t just feel harder. It burns through your fuel supply disproportionately.
Your body has two main fuel sources during endurance running: fat and glycogen (stored carbohydrates). At easy paces, you burn a higher percentage of fat, which you have essentially unlimited stores of. As intensity rises, you shift toward glycogen, which you have maybe 90 minutes worth of at moderate effort.
The relationship isn’t linear. Going from 8:00/mile to 7:30/mile doesn’t just increase glycogen burn by a proportional amount. It increases it disproportionately, because you’re crossing metabolic thresholds where your body relies more and more heavily on carbohydrate oxidation. Those first few miles at a pace that’s “only” 30 seconds too fast can chew through glycogen you’ll desperately need in the final miles.
There’s also a lactate accumulation issue. Running above your threshold pace early in a race floods your muscles with hydrogen ions and other metabolic byproducts. Your body can clear these, but it takes time and energy that’s now unavailable for forward motion. Starting conservatively keeps you below that threshold longer, preserving your ability to actually push the pace later when it counts.
The math is straightforward. If you run the first half of a half marathon in 52 minutes (7:55/mile pace) and the second half in 59 minutes (9:00/mile pace), your finish time is 1:51. If you run the first half in 55 minutes (8:24/mile pace) and the second half in 54 minutes (8:14/mile pace), your finish time is 1:49. The second scenario is two minutes faster despite feeling slower at the start. You ran negative splits and your overall time improved because you didn’t blow up.
Why it’s psychologically brutal
Knowing this doesn’t make it easy. At the start of a race, you’re rested, tapered, caffeinated, and surrounded by excited people. Your body feels fantastic. Every instinct says to go.
Running 10-15 seconds per mile slower than your goal pace when you feel like you could easily run 20 seconds faster takes real discipline. It feels like leaving money on the table. Runners around you are pulling ahead and some part of your brain screams that you’re falling behind.
The other psychological trap is that you can’t feel glycogen depletion coming. You feel fine until you suddenly don’t. There’s no gradual warning system. One mile you’re running strong, the next mile your legs are concrete. By the time you feel the bonk, it’s too late to do anything about it.
A lot of runners understand negative splitting intellectually but can’t execute it because the emotional pull of the start line is too strong. It takes practice in training, not just reading about it.
How to actually do it
Use heart rate, not pace, for the first half
This is the single most useful strategy I’ve found. Instead of looking at pace for the first half of a race, I watch my heart rate. I know from training that my half marathon effort sits around 160-165 bpm. For the first 5-6 miles, I keep my heart rate at or below 158, regardless of what the pace readout says.
This works because heart rate doesn’t lie about effort the way pace does. On a cool day with a tailwind, you might be running faster than goal pace at the right effort. On a hot day or a hilly course, goal pace might be above the right effort. Heart rate adjusts for all of that.
The catch is that cardiac drift means your heart rate will naturally rise over the course of a long race even at the same effort. So what you’re really doing is banking cardiovascular headroom early and spending it later.
Start 10-15 seconds per mile slower than goal pace
If you’re not using heart rate, a simple rule: run the first two miles 10-15 seconds per mile slower than your goal finish pace. This will feel absurdly easy. That’s the point.
By mile three or four, you can settle into goal pace. At the halfway point, if you feel controlled and your breathing is manageable, you can start to push slightly below goal pace. In the final 2-3 miles, you release whatever’s left.
Pick a landmark to start building from
Before the race, decide on a specific point where you’ll start picking up the pace. For a half marathon, I use mile 8. For a 10K, I use the 4-mile mark. Having a concrete plan removes the guesswork and the temptation to accelerate too early.
When I reach that landmark, I don’t dramatically speed up. I increase effort by maybe 5% and let the pace follow naturally. Then at mile 10 (or 5 in a 10K), I increase again. The acceleration is gradual, not a sudden gear change.
Practice in training
I do negative split long runs about once a month. The format is simple: run the first half at an easy conversational pace, then pick it up to moderate effort for the second half. Last month I did 14 miles with the first 7 at 9:15 pace and the second 7 at 8:25. The total time was faster than if I’d tried to hold 8:45 the whole way, because that pace would have been too hard for the first half and I’d have slowed in the final miles.
Tempo runs work for this too. Instead of running 4 miles at the same tempo pace, run the first mile slightly below tempo and the last mile slightly above it. You learn what it feels like to accelerate when you’re already working.
Why this matters even more in triathlon
If you race triathlons, negative splitting the run leg is close to essential.
You’re starting the run after swimming and cycling. Your legs are pre-fatigued, your glycogen stores are partially depleted, and your heart rate is already elevated from the bike. The temptation to run at your standalone 10K pace off the bike is enormous, especially in a sprint triathlon where the run is short enough that you think you can gut it out.
You can’t. Or at least, you’ll pay for it. The bike-to-run transition changes your physiology enough that you need to recalibrate.
My approach now is to run the first mile off the bike about 20-30 seconds per mile slower than I would in a standalone race. It takes about a mile for my legs to find their running rhythm after cycling, and once they do, I can start pushing. The first mile always feels terrible regardless of pace, so running it conservatively costs almost nothing and saves a lot.
The ego problem
The hardest part of negative splitting is watching people pass you early and trusting the plan. It requires a kind of confidence that only comes from having done it successfully before.
Start with a training run. Do a 6-mile negative split run this week. Run the first 3 miles easy, like actually easy, then pick it up for the last 3. Feel what it’s like to finish fast. Then try it in a low-priority race where the result doesn’t matter.
Once you’ve crossed a finish line feeling strong while the people who passed you at mile 2 are walking, you won’t want to race any other way.
