Mountain Ultra Leg Strength: What Actually Holds Up On The Descent

Nobody drops out of a mountain ultra because their legs were not strong enough on the climb. They drop because somewhere on the third big descent the quads stop absorbing, the stride turns into a controlled fall, and every step starts costing more than it should.

Mountain ultras are eccentric events. You spend hours asking your quads to work as brakes, and braking is what tears muscle. That is the whole problem in one sentence. Load above capacity equals damage, and on a descent the load arrives whether your capacity is ready or not. Gravity does not negotiate.

I keep a master guide on this for my own training and for the runners I coach. It is on version six. What follows is that guide rebuilt as an article, with every claim in it run back through the literature. Most of it held up better than I expected. Two things did not, and I have flagged both, including one that moved a movement up my own ranking and pushed another one down.

The hierarchy

If you have twenty minutes a week, you start at the top and work down. Nothing below the fourth line matters much if the first line is missing.

Heavy lifting makes you faster without making you bigger

This is the one runners resist hardest, and it is the best supported thing in the guide. Systematic reviews put the running economy improvement from heavy strength work at roughly 2 to 8% depending on the athlete and the protocol, with the larger gains going to runners who were not already strong.

Economy is the whole game in an ultra. A 4% improvement means every step costs less oxygen, and across fourteen hours that compounds into something enormous.

The mechanism is neural, not structural. Heavy lifting expands what Steve Magness calls the recruitable fiber pool. Late in a race, as your usual fibers give up, your nervous system reaches for fibers that are still fresh. That is what keeps you tall when tired instead of collapsing into a shuffle.

One correction here. Version six claimed that endurance signaling actively blocks the mTOR pathway, so runners cannot gain mass. The molecular interference is real acutely, but the long term human data does not support that clean a story. The actual reasons heavy lifting does not bulk up ultra runners are simpler: the rep volume is low, the energy cost of your training is enormous, and the programming is built around neural work rather than hypertrophy. Right conclusion, wrong reason.

The descent is the event

Here is what a descent does to a quadriceps.

Fifteen minutes. Not fifteen hours. One steep quarter hour costs better than a fifth of your quad strength, and it is still gone two days later. Now picture that at hour nine of a hundred kilometer race with four thousand metres of descent behind you.

The good news is that this adapts, and it adapts fast. It is called the repeated bout effect. One hard eccentric session confers real protection on the next one, and that protection lasts weeks, not days. Studies have shown the damage response significantly blunted when the same downhill bout was repeated two and three weeks later.

And this is where version six had a gap I want to name. The guide loaded up on slow eccentric split squats and step-downs, which are genuinely useful, and then quietly assumed those would buy the repeated bout effect for race day. The clearest evidence for that protection comes from downhill running itself, not from gym eccentrics.

So the fix is simple. Keep the slow eccentrics. Add actual downhill running. A dedicated descent session every two to three weeks, on terrain that resembles your race, with the last hard one landing three to five weeks out so you get the protection without carrying the damage to the start line.

What I got wrong about isometrics

Version six ranked yielding isometrics third and credited them with stiffening tendon and relieving pain. I have moved them to fourth, and I want to explain why rather than quietly edit the table.

The enthusiasm traces back to a 2015 study that found dramatic, immediate pain relief in patellar tendinopathy after isometric holds, with pain dropping from 7 out of 10 to near zero and staying down for 45 minutes. That result launched a thousand rehab protocols, mine included.

It has not replicated cleanly. Later trials and a pooled analysis found isometric exercise no better than isotonic for pain relief, and one crossover study found the effect small and not sustained at 45 minutes. The honest read now is that contraction mode is probably not the important variable.

I still program them, and I still like them, for a completely different reason. Isometrics let you load a joint hard with no eccentric damage and almost no soreness, which makes them the one strength input you can keep during race week without paying for it. That is a real and useful property. It is just not magic, and it does not outrank loading the soleus.

The soleus is the engine nobody trains

Look at that number again. Six point seven times body weight at an easy pace, climbing toward eight point seven as you speed up. Nothing else in the leg comes close.

Most runners train calves standing with a straight knee, which recruits the gastrocnemius. Bend the knee to ninety degrees and the gastroc goes slack, which forces the soleus to carry the load by itself. That is why the seated raise matters and the standing one does not replace it.

When the soleus runs out of capacity late in a race, you feel it as calves that will not fire, a stride that flattens out, and pace decay that no amount of will fixes. It is also the structure sitting underneath most Achilles problems, which is why the Achilles deep-dive on Form & Function spends as much time on loading the calf as it does on the tendon itself.

Plyometrics are a multiplier, not a standalone

Version six put plyometrics at number eight and described them as teaching the body to behave like a bouncy ball instead of a tomato, which is Magness's phrase and a good one. The ranking turns out to be roughly right, for a more interesting reason than I gave.

On their own, the pooled effect of plyometric training on running economy is close to trivial. Combined with heavy resistance training, the effect becomes large. Duration matters too: under six weeks shows nothing, while programs beyond seven weeks and fifteen sessions show real gains.

So plyometrics are not a strength substitute and they are not optional garnish. They are a multiplier that only works when line one of the hierarchy is already in place and you stick with them long enough.

Dose stays conservative. Twenty to forty foot contacts, once or twice a week, when fresh. Learn to absorb force before you produce it. Skip the deep depth jumps entirely, because those eccentric loads were designed for jumping specialists, not for someone who also runs a hundred miles a week.

The Magness hill circuit

This is the bridge between the gym and the mountain. Sprint a moderate to steep hill for 35 seconds. Stop and immediately perform 10 to 12 explosive reps, squat jumps or pogo hops. Then without rest, launch into a second 35 second surge on pre-fatigued legs. Walk down and rest fully, three to four minutes, then repeat. Three to five sets total.

Adjust for grade. Baseline is 30 to 35 seconds on flat or slight incline, and for every 1% of slope above 5%, subtract a second. A 15% wall becomes a 20 second effort.

What the best in the world actually do

None of the athletes below train the way a textbook says to. What they share is not a program. It is that the specific demand of their event shows up somewhere in their week, every week.

The common thread is the one at the top of this article. Every one of them loads the specific tissue their event punishes, on a schedule they can actually hold. That is the whole job.

The week

Add the downhill running session on its own two to three week rotation, separate from these lifting days. It is a running session, not a gym session, and it needs its own recovery.

None of this replaces the aerobic work underneath it. Strength raises the ceiling, but the layers that carry you through a long day get built in the easy and threshold zones, which I laid out in the training pyramid. And if your knees are the thing that complains on descents rather than your quads generally, three kinds of knee pain sorts out which one you are dealing with.

The taper

The evidence check

I put every load-bearing claim in the guide back under the light. This is the part I would want to read if someone handed me their training protocol.

When loading is not enough

Sometimes the tissue is too irritated to accept the load that would fix it, and you end up in a loop where every attempt to build capacity flares the thing you are trying to build.

That is the point where I bring in the regenerative side of the practice at Form & Function Chiropractic and Rehab in Brooklyn: focused shockwave for deep, focal tendon and bone stress, radial pressure wave for broader muscular work, and EMTT alongside it, with BioFlex laser when the tissue needs calming rather than stimulating.

None of that replaces the strength work. It shortens the runway so the strength work becomes possible. The capacity side still gets built through mobility and joint work, a proper running assessment, and where the engine itself is the limiter, metabolic testing.

The exercise index

Heavy trap bar deadlift. Feet hip width inside the bar, hinge at the hips, chest up, spine neutral. Drive the floor away through the midfoot. Stop one or two reps short of failure. Always.

Slow single-leg split squat or step-down. Stand on a 6 to 8 inch box. Lower on a strict four to five second count, barely touch the heel down, do not bounce, then drive up in one second.

Heavy seated soleus raise. Knees bent to ninety degrees, toes on a block, heavy load across the lower thighs. Drop the heels below the block for full stretch, then drive up and hold two seconds at the top.

Spanish squat hold. Heavy band behind the knees pulling them forward. Sit back with an upright torso, knees at 70 to 90 degrees, shins driving vertical against the band. Hold 45 to 60 seconds.

Nordic hamstring curl. Ankles secured, hips extended, torso straight. Lower on a four to five second count as far as you can control, catch yourself, push up lightly, and pull back with the hamstrings.

Tibialis anterior raise. Back against a wall, heels about twelve inches out. Pull the toes toward the shins as high as they go, pause one second, lower slowly. If your shins are already talking to you, read shin splints versus stress fracture before you add load.

Box drop and pogo hops. Step off a 12 to 18 inch box onto two feet and spring up immediately with minimal ground contact. Land like a spring with stiff ankles, knees tracking over the toes, never collapsing inward.

Single-leg wall sit. Back flat against the wall, thigh parallel to the floor, one foot lifted. Press through the midfoot of the grounded leg, hips square, 30 to 45 seconds per side.

Book your evaluation

Every number in this guide is a population average until someone measures you. If your quads are giving out on descents, or your calves are the thing that always goes late, the useful question is not which exercise to add. It is which capacity is short and why.

If you would rather have this built into an actual training block instead of bolted onto one, that is what run coaching is for.

That is what the ADAPT method exists to answer, and it is the same equation underneath all of it. Load above capacity equals pain. You can lower the load or raise the capacity, and only one of those gets you up the mountain faster.

Book your evaluation and we will find out which one you are short on.

References

1. Blagrove RC, Howatson G, Hayes PR. Effects of Strength Training on the Physiological Determinants of Middle- and Long-Distance Running Performance: A Systematic Review. Sports Medicine, 2018.
2. Llanos-Lagos C, et al. The Effect of Strength Training Methods on Middle-Distance and Long-Distance Runners' Athletic Performance: A Systematic Review with Meta-analysis. Sports Medicine, 2024.
3. Heavy Resistance Training Versus Plyometric Training for Improving Running Economy and Running Time Trial Performance: A Systematic Review and Meta-analysis. 2022.
4. Effect of Strength Training Programs in Middle- and Long-Distance Runners' Economy at Different Running Speeds: A Systematic Review with Meta-analysis. 2024.
5. Vernillo G, et al. Downhill Running: What Are The Effects and How Can We Adapt? A Narrative Review. Sports Medicine, 2020.
6. Downhill running increases markers of muscle damage and impairs maximal voluntary force production. European Journal of Applied Physiology, 2023.
7. The repeated bout effect influences lower-extremity biomechanics during a 30-min downhill run. 2022.
8. A single bout of downhill running attenuates subsequent level running-induced fatigue. 2020.
9. Dorn TW, Schache AG, Pandy MG. Muscular strategy shift in human running: dependence of running speed on hip and ankle muscle performance. Journal of Experimental Biology, 2012.
10. Rio E, et al. Do isometric and isotonic exercise programs reduce pain in athletes with patellar tendinopathy in-season? A randomised clinical trial. Journal of Science and Medicine in Sport, 2015.
11. Isometric exercise and pain in patellar tendinopathy: A randomized crossover trial. 2019.
12. van Dyk N, Behan FP, Whiteley R. Including the Nordic hamstring exercise in injury prevention programmes halves the rate of hamstring injuries. British Journal of Sports Medicine, 2019. See also the 2021 methodological reappraisal.
13. Using Nutrition and Molecular Biology to Maximize Concurrent Training. Gatorade Sports Science Institute.
14. Tapering and Peaking. National Strength and Conditioning Association.

Circuit structure and the recruitable fiber pool framing draw on Steve Magness's work in The Science of Running.