Hill repeats for distance runners: the evidence

hill repeats
hill training
uphill intervals
running economy
interval training
evidence

Hill repeats for distance runners: the evidence

A runner climbing a dirt track through a green beech forest

Hills work, and the evidence says the format matters less than you think

Hill repeats have a long history in distance running. Coaches have used them for decades as running-specific strength work, a way to get a hard aerobic session in with less pounding, and a route to better economy. The research behind them is smaller than that reputation suggests, but it does exist, and it mostly points in a reassuring direction: well-trained runners who add hard uphill intervals tend to get a little faster, and the exact format seems to matter less than doing them at all.

Before getting into the studies, some perspective. For nearly all runners, the biggest gains come from running more, consistently, over months and years. Hill repeats are a second-order refinement on top of that. They can help at the margin, especially for runners who already have a solid base, and they work best when the weekly running behind them is steady.

Key takeaways

  • The best-known trial put 20 well-trained runners (average 5 km best of about 16:30, around 95 km a week) through six weeks of uphill intervals. 5 km time trial performance improved by about 2% on average.
  • No hill format was clearly best for 5 km performance. The shortest, steepest, hardest efforts did the most for running economy.
  • A second trial in 32 well-trained runners found hill intervals improved time to exhaustion, but flat intervals improved it more.
  • Running uphill sharply reduces impact forces compared with level running, which is part of why hills are seen as a lower-impact way to run hard.
  • All of the trials are small, short and run in well-trained adults. There is little direct evidence for beginners or for marathon performance.

What hill repeats are meant to do

The rationale for uphill intervals has three strands.

Running-specific strength. Pushing up a gradient demands more force from the hips, knees and calves on every step, in a movement pattern close to normal running. Coaches often describe hills as strength training in disguise.

A hard aerobic stimulus. A hill lets you reach high heart rates and oxygen uptake at slower speeds than you would need on the flat.

Lower impact. Gottschall and Kram (2005) measured ground reaction forces in 10 people (five men, five women) running at 3 m/s on a force treadmill set at level and at 3°, 6° and 9° slopes. At a 9° incline, the normal impact force peaks seen in level running disappeared, while the forces pushing runners forward rose by 75%. Downhill was the reverse: at a 9° decline, impact peaks rose by 54% and braking forces by 73%. The abstract does not describe the participants' running background, and the study measured forces in a lab without tracking injuries. It does support the idea that the uphill half of a hill session is mechanically gentle, and it is a reason to treat the jog back down with some respect.

What the trials found

Barnes et al. (2013): six programmes, one clear message

The most cited hill study is from Kyle Barnes and colleagues at the Auckland University of Technology. Twenty distance runners took part. They were young (average age 21), had an average 5 km best of 16.5 minutes, were running about 95 km a week, and had been training for about six years. That is a well-trained, competitive group, faster than most club runners.

The runners were randomly assigned to one of five uphill interval programmes, with three to five runners per group. Programmes ranged from very short, steep efforts (8–12 second sprints on an 18% gradient) through repeats of two to two and a half minutes on a 10% gradient, to long efforts on gentler slopes. Each runner did two uphill sessions a week for six weeks. Total running volume was held as it was: the hill sessions replaced some of each runner's normal running.

The results:

  • 5 km time trial performance improved by an average of 2.0% across all groups (90% confidence limits ±0.6%), roughly 20 seconds at that level. No single programme was clearly best.
  • The highest-intensity programme was clearly best for running economy, improving it by 2.4% (±1.4%), and for every neuromuscular measure, including jump performance.
  • Other aerobic measures did best near the middle of the intensity range.

The authors concluded that, until more data arrive, runners can assume any form of high-intensity uphill interval training will benefit 5 km time trial performance.

The design has real limits. There was no control group doing flat intervals or no intervals, so the study cannot show that hills beat other hard sessions, or that some of the improvement was not a training or testing effect. With three to five runners per group, the comparisons between programmes are uncertain. And because total volume was kept the same, the study says nothing about what happens when hills are added on top of a runner's usual week.

Hill programmes like these, written as fixed sets and gradients, are generic by design. kaizen is hyper-personalised: it accounts for the overall intensity of the running you actually do, including pace and elevation, to set your training target for the week, and it recalculates as you run. A climb-heavy session counts for what it cost you.

Ferley et al. (2013, 2014): hills against flat intervals

Derek Ferley and colleagues ran the comparison Barnes could not. Thirty-two well-trained distance runners (average age 27, average VO2max of 60.9 ml/kg/min) were assigned to uphill intervals (12 runners), flat intervals (12) or a control group continuing their normal training (8). Before the study, the runners were doing moderate-intensity running three or four times a week and no high-intensity intervals.

For six weeks, both interval groups did two interval sessions and two continuous runs a week, all on a treadmill. The hill group ran 10 to 14 repeats of 30 seconds on a 10% grade at the speed associated with their VO2max. The flat group ran four to six longer repeats at the same speed on a 1% grade.

Time to exhaustion at VO2max speed rose by 68 seconds in the hill group, 138 seconds in the flat group and 52 seconds in the control group. The improvement was significant in both interval groups, and the flat intervals produced the bigger gain. In the follow-up paper on economy, every group improved oxygen cost and blood lactate at submaximal speeds, control group included. Both interval groups also improved the share of their VO2max they could use at lactate threshold. Neither hill nor flat intervals improved muscle power, which the authors took as a sign that incline treadmill intervals are not enough on their own to act as strength training.

Two points stand out for interpretation. The control group improved on many measures too, so some of the gains reflect training through a six-week block and repeated testing. And the abstracts do not report whether total weekly training time in the interval groups matched what the control group did, so differences in overall volume could account for part of the effect.

How hills fit into a week

What follows is kaizen's interpretation of the evidence above.

For well-trained runners with a consistent base, short, hard hill repeats look like a reasonable way to add intensity, with some economy and neuromuscular benefit and relatively low impact on the way up. Based on Barnes, the specific format is not worth agonising over.

For recreational runners building towards more volume, the evidence is thinner. None of these trials studied beginners or slower runners. Hill sessions are hard, and a hard session that leaves you too tired to run the rest of the week costs more than it gives.

For marathoners, no trial here measured marathon performance. The outcomes were 5 km time trials, time to exhaustion and lab markers.

Generic plans often write hills in as a fixed weekly session, set without reference to the rest of your running. kaizen calculates your fitness from your recent running and past races, and because elevation feeds into the intensity of each run, a hilly week shows up in your load. You choose whether that load comes from structured repeats or a rolling long run.

For more on the physiology, see our posts on running economy and VO2max intervals.

Sources and uncertainty

  • Barnes et al. 2013, International Journal of Sports Physiology and Performance. 20 well-trained runners (5 km best around 16:30, about 95 km a week, about six years of training); five uphill programmes for six weeks; three to five per group; no control group.
  • Ferley et al. 2013 and 2014, Journal of Strength and Conditioning Research. 32 well-trained runners (VO2max around 61 ml/kg/min); uphill, flat and control groups; treadmill-based, six weeks.
  • Gottschall and Kram 2005, Journal of Biomechanics. Ground reaction forces in 10 people at 3 m/s on slopes up to 9°; running background not reported in the abstract; no injury data.

Every trial here is small, short and run in well-trained adults, so the findings may not carry over to novices, slower recreational runners or longer races. No study tracked runners for more than six weeks, and none measured injuries.

The bigger picture

Hill repeats seem to help well-trained runners a little, in almost any form, and the gains in these studies came on top of years of steady training and, in the Barnes group, about 95 km a week. That base did most of the work before the first hill rep. The choice of gradient and rep length is a refinement. Running more, week after week, is what builds the fitness that hill sessions sharpen.

Where kaizen fits

If you are spending months chasing seconds, kaizen calculates your fitness from your recent running and past races, then sets a weekly training target your goal requires, built from the pace and elevation of the running you actually do. It recalculates as each run arrives, so a week of hill repeats or hilly routes counts for what it was. You pick the sessions. See your fitness in kaizen and build the consistent weeks that make every hill worth climbing.