How to Pace an Ultramarathon

Pace by effort, and use the other two as cross-checks. Minutes-per-mile is close to useless on trail, because grade and terrain swing it all over the place. Heart rate is fine early on, but it drifts upward (cardiac drift) and lags on short climbs, so it starts lying to you late in the day. Most experienced ultrarunners go by rate of perceived exertion (RPE). You hold an easy, conversational effort, somewhere around RPE 3 to 4 out of 10, for the first several hours, then you let the terrain and how you feel set the pace from there. Keep the watch and the heart rate in front of you as a sanity check, but do not let them run the show.

Start slower than feels right. A good rule is that the first hour or two should feel almost too easy, an effort you know you could hold all day, around RPE 3 out of 10. And the reason matters. Research on long ultras finds that more even pacing lines up with faster finishes, and the fastest runners start at lower relative intensities than the slower ones. Going out too hot does damage you cannot take back. It raises your core temperature, it shuts down your gut so you absorb fewer carbs, and it trashes your legs early, and all of that piles up over the back half. In an ultra a fast start does not cost you a slightly slower finish. It often costs you the whole race.

Grade-adjusted pace is the flat-ground pace that would cost you the same effort as your actual pace on a hill. Uphill, your GAP is faster than your real pace, because the climb is doing extra work for you. Downhill, your GAP is slower than your real pace. And your trail pace looks slow for two reasons. Climbing burns way more energy per mile, so an honest effort gives you a much slower clock, and GAP only accounts for the gradient, not the terrain. Rocks, roots, mud, and stream crossings all add cost that GAP ignores, so real trail pace often runs 10 to 20 percent slower than GAP on technical ground. Treat GAP as a lower bound and add a buffer.

Hike once running stops paying off, which for most people is when a climb forces you below roughly 18 to 19 minutes per mile. On grades of about 4 to 15 percent, if you are grinding at that pace or slower, dropping to a strong power-hike costs you less energy for almost the same speed. Above 15 percent, just about everyone should hike. The trick is to make the switch before your heart rate spikes, not after, and to hike like you mean it: hands driving on the thighs, quick cadence, eyes up the trail. Power-hiking is not quitting. It is the efficient gear for steep ground, and it banks energy you will want for the runnable miles.

The best ratio is the one you can hold for the whole race, and you should walk from the start, before the race forces it on you. Common starting points are 25 minutes running to 5 walking on runnable courses, or a frequent run 2 minutes / walk 1 minute for time-on-feet beginners (which averages near 13:00 per mile for a lot of runners). On rolling or mountain courses, let the terrain set the ratio: power-hike every real climb and run the flats and gentle descents. Walking is not failure. It saves your legs, it lets you eat and drink, and very often it gets you a faster overall time than trying to run the whole thing.

Aim for even effort. On the clock that usually shows up as an even-to-slightly-positive split, not a true negative split. The fatigue, the heat, and all the climbing in an ultra make a real negative split rare, and chasing one tempts you to bank time early, which is the classic blow-up. Studies of long mountain ultras find the steadiest pacers finish fastest. So the goal is simple: feel like you are holding the same controlled effort at mile 70 that you held at mile 10, and still be passing people in the last few hours because you did not overspend at the start.

On the climbs, hold the effort and let the pace fall. Power-hike the steep pitches instead of forcing a run, and keep your breathing easy enough to talk in short sentences. The mistake is chasing a pace number uphill, which spikes your effort and burns matches you are going to want later. On the descents the opposite trap gets you. The aerobic effort feels easy, so people bomb the downhills and shred their quads. Eccentric braking on steep descents (steeper than about -10 percent) actually costs more than flat running and causes muscle damage that can stick around for days, so run the early descents controlled and light. The runners still standing at mile 80 are the ones who protected their quads early.

Build it backward from the cutoffs, not forward from your dream finish. Take the official cutoff at each major aid station, then work out the moving pace and run-walk plan you need to reach each one with a buffer, accounting for the elevation, the terrain, the heat, and the time you will actually spend stopped (often 1 to 3 hours total over a 100). The early checkpoint cutoffs are the ones that catch people, because climbing, heat, and the night all slow you down late, so you want margin in the bank early without sprinting. A course-aware projection that loads the vert into each segment beats a flat per-mile average, which the mountains will quietly tear apart.