Strength training for runners

Runners are told to lift, and the advice usually stops there. Do some strength work. Two sessions a week. Body weight is fine to start with. None of that is wrong, and it is vague in exactly the place where the research is specific.

Two systematic reviews published in 2024 pooled the training studies on middle distance and long distance runners and sorted them by what kind of strength training was actually done. What came back is narrower than the advice. The method matters, and the methods are not interchangeable.

Heavy is the part that works

The first review covers 38 studies and 894 runners, from moderately trained club runners up to highly trained ones. Training at or above eighty per cent of a one repetition maximum improved running performance, and combining heavy lifting with jump work improved it more. Plyometric training on its own did not reach significance for performance¹.

The second review asked the same question of running economy, which is the energy you spend holding a given pace. It covers 31 studies and 652 runners, and the picture is the same. Heavy loads improved economy. Combined methods improved it more. Training with submaximal loads, and training with isometric holds, did not improve it at all².

So the fork in the road is not whether to lift. It is whether to lift heavily enough for it to count. Three sets of fifteen with a light dumbbell is submaximal load training, and that is the arm of the evidence that did not move.

The strength work that changes your running is the strength work that looks like strength training.

Why the light version is so common

Three sets of fifteen is not where the evidence points, and it is easy to see why it is where people land. Running is endurance, so the strength work ought to be endurance too, and a heavy barbell looks like it belongs to a different sport and a different body.

The reasoning is neat and the evidence does not support it. Your running already trains your ability to repeat a submaximal effort for an hour, and it trains it far better than a dumbbell will. What running does not train is the ability to produce a lot of force quickly, which is the thing the heavy work adds and the thing that shows up in economy.

There is also a practical reason. Lifting light is easy to fit anywhere and easy to do badly with no consequence. Lifting heavily needs a warm-up, a rack, some rest between sets and a record of what you did last time. It is a real session rather than fifteen minutes at the end of a run, and that is most of why it gets skipped.

What it does not do

It does not raise your maximal oxygen uptake. Across every method those reviews examined, the effect on maximal oxygen uptake was trivial, and the same was true of velocity at maximal oxygen uptake and of the maximum metabolic steady state¹.

That is worth saying plainly, because it locates the benefit. Lifting is not a second way of training your aerobic engine. It changes what you get out of the engine you have. The same pace costs you less, and the last few kilometers hold together better.

The injury argument, honestly

Strength training also carries the strongest injury prevention evidence of any exercise intervention that has been tested. A meta-analysis of randomised trials across sports found strength training programs cut injuries to under a third of the rate in the control groups, a risk ratio of 0.315³.

Three caveats belong in the same breath. The strength training estimate rests on four of the twenty-five trials in that review rather than on all of them³. The analysis pooled several sports rather than running alone. And it is now more than ten years old. Read it as a strong signal about lifting in general rather than as a number for your season.

How to fit it around the running

1. Two sessions a week sits comfortably inside the evidence. The programs in the performance review ran from six to forty weeks at one to four sessions a week¹, so two is neither the minimum anybody has tested nor the maximum.
2. Load is the variable to protect. The threshold in both reviews is eighty per cent of a one repetition maximum or above¹². Everything else about the session can flex around your mileage. That number is the part that cannot.
3. Keep it going through the season. Six weeks was the shortest program in the review¹, and most ran far longer. A block of lifting in January and nothing after March is not what was studied.
4. Let the running be the priority it is. Nothing above says lifting should cost you a key session. It says that when you do lift, lifting lightly is the version that has not been shown to work.

The bookkeeping is the hard part

A runner's lifting has one job. Get heavier, slowly, without eating the legs you need on Saturday.

Both of the difficult parts are admin rather than effort. Knowing what eighty per cent of your maximum is on each lift, as your maximum moves. And knowing whether last week's session was one you actually recovered from, which is a question about what you lifted rather than about how you felt on the drive home.

Neither is hard. Both are tedious, and being tedious is why most runners' strength work quietly turns into three sets of fifteen with a light dumbbell by March.

Reckoner writes each week from the sets you logged the week before, so the load moves when your own reps say it should and holds when they say it should not. It is a strength coach and nothing more than that. It does not know about your mileage, and the running is still yours to plan around it.