Most marathon runners have a complicated relationship with the gym. They know they should go. They've read the articles. They've seen the Instagram posts of elite runners doing single-leg squats and box jumps. And yet, when it comes to choosing between an extra easy run and a strength session, the run almost always wins.

The reasons are familiar. Strength training feels like it belongs to a different sport. The gym is intimidating if you don't know what you're doing. And there's a persistent worry, rarely spoken aloud but always present, that lifting heavy will add bulk, slow you down, or leave your legs too tired for the sessions that actually matter.

The science says otherwise. Not vaguely, not in the "it probably helps" way that most training advice is delivered. The evidence that strength training improves marathon performance is robust, specific, and increasingly difficult to ignore. And the mechanism isn't what most runners think it is.

It's not about getting stronger (not exactly)

The most common framing of strength training for runners is injury prevention. Strengthen your glutes, stabilise your hips, protect your knees. There's truth in this, but the evidence is more nuanced than the headlines suggest. Retrospective studies consistently find that muscle weakness, particularly around the hips, is a characteristic of injured runners. But prospective studies, the ones that follow runners forward in time to see who gets hurt, have struggled to show that strength training alone reliably prevents injuries, especially when the training is unsupervised. A 2025 scoping review in Translational Sports Medicine noted that injury rates among runners haven't decreased over the last several decades, despite strength training becoming more popular.

This doesn't mean strength training is irrelevant to injury risk. It means the injury prevention framing is too narrow. The real performance case for strength training is about something more specific and more measurable: running economy.

Running economy is the oxygen cost of running at a given pace. Two runners with identical VO2max values can perform very differently in a marathon if one of them uses less oxygen at race pace. It's the efficiency of the machine, not the size of the engine. And strength training, particularly heavy resistance work and plyometrics, consistently improves it.

A 2024 systematic review and meta-analysis in Sports Medicine examined the effect of strength training programmes on running economy across middle- and long-distance runners. The findings were clear: strength training with loads at or above 80% of one-repetition maximum improved running economy, with typical improvements in the range of 2-4%. For a three-hour marathoner, that translates to roughly two to four minutes. Not from running more. Not from running harder. From making each stride cheaper.

The mechanism is primarily neuromuscular. Heavy strength training improves the rate of force development and the stiffness of the muscle-tendon unit. This means each ground contact produces more elastic energy return and wastes less energy on unnecessary muscle activation. You're not building bigger muscles. You're building muscles that fire more efficiently and store energy more effectively with every stride.

The durability effect

There's a second benefit that's only recently been quantified, and it may matter even more than the economy improvement in fresh conditions.

A 2025 randomised controlled trial published in Medicine and Science in Sports and Exercise examined what the researchers called "running economy durability," the ability to maintain running efficiency as fatigue accumulates. After ten weeks of concurrent strength and endurance training, the strength group maintained better running economy after 90 minutes of running and showed a 35% improvement in time to exhaustion at high intensity. The control group, which only ran, saw an 8% decline over the same period.

This is the finding that should change how marathon runners think about the gym. Marathon performance isn't determined by how efficient you are at kilometre five. It's determined by how efficient you are at kilometre thirty-five, when your muscles are damaged, your glycogen is depleted, and your form is deteriorating. Strength training doesn't just make you more economical. It makes you more resistant to the loss of economy that accumulates over the course of a race.

If you've ever watched a marathon and seen the contrast between the runners who hold their form in the final kilometres and those who are visibly falling apart, you're watching the durability effect in action. Some of that is aerobic fitness. Some of it is pacing discipline. And some of it, the research now suggests, is the structural resilience that comes from consistent strength work.

What to actually do in the gym

This is where most articles either get too vague ("do some squats and lunges") or too complicated (a twelve-exercise circuit with Swiss balls and resistance bands). The research points to three categories of movement that matter for runners, and none of them require complicated equipment.

Heavy compound movements. Squats, deadlifts, and their single-leg variations (Bulgarian split squats, single-leg Romanian deadlifts). These are the movements that drive the neuromuscular adaptations behind running economy improvements. The key word is heavy: the research showing economy benefits used loads at or above 80% of one-repetition maximum, which means sets of three to six repetitions, not sets of fifteen with light dumbbells. If the weight doesn't challenge you within six reps, it's not heavy enough to drive the adaptation.

Single-leg work. Running is a single-leg activity. You're never on both feet at the same time. Single-leg squats, step-ups, and single-leg calf raises address the stability and force production demands that are specific to the running stride. They also expose asymmetries, which are common in runners and worth identifying before they become problems.

Plyometrics. Box jumps, bounding, depth jumps, and single-leg hops. These train the stretch-shortening cycle, the rapid stretch-and-contract pattern that makes tendons behave like springs. The meta-analysis data suggests plyometric training may be particularly effective for improving economy at lower speeds, which is exactly where most recreational marathoners race. But plyometrics carry a higher injury risk than other strength modalities and should be introduced gradually, especially for runners without a strength training background.

What you don't need: bosu balls, cable machines, or a forty-five-minute core circuit. Core stability matters, but a few minutes of planks and dead bugs at the end of a session covers it. The time in the gym is better spent on the movements that have the strongest evidence base.

Two to three sessions per week during your base phase, dropping to one to two during your build phase, is the range supported by the research. Each session can be as short as thirty to forty minutes. You don't need to live in the gym. You need to be consistent and lift heavy enough to drive adaptation.

Periodising strength alongside your running

This is the practical core, and where most runners get it wrong. They either do the same strength routine year-round regardless of where they are in their training block, or they do strength work sporadically and drop it the moment running volume increases. Both approaches leave value on the table.

The principle is the same one we outlined in our article on training load management: one lever at a time. You don't increase running volume and introduce heavy strength work in the same week. You sequence them so that each new stressor is absorbed before the next one is added.

Base phase (12+ weeks out from race day). This is when you build strength. Two to three sessions per week, working with heavy loads (80%+ 1RM), progressing gradually over six to eight weeks. Running volume is moderate, intensity is low, and your body has the capacity to absorb both stressors. This is also when you introduce plyometrics if you haven't done them before, starting with low volumes (two to three sets of five to eight reps) and building from there.

Build phase (8-12 weeks out). As running intensity increases, strength training shifts from building to maintaining. Drop to one to two sessions per week. Keep the loads heavy but reduce the volume: fewer sets, fewer exercises. The goal is to preserve the neuromuscular adaptations you built in the base phase without creating additional fatigue that competes with your quality running sessions. This is where the sequencing of sessions matters. A heavy squat session the day before a threshold run is a poor combination. A maintenance strength session the day after a hard run, when you'd be doing easy running or resting anyway, is a better fit.

Taper phase (2-3 weeks out). Reduce strength training to one short session per week or drop it entirely. The adaptations you've built over the preceding months won't disappear in two weeks. What you're protecting now is freshness. As we discussed in a previous article on HRV-guided training, your HRV should be rising during the taper as your body super-compensates. Adding unnecessary strength sessions during this window works against that process.

Post-race. Give yourself at least a week, ideally two, before returning to structured strength work. Your muscles need to repair. Your connective tissue needs to recover. When you do return, start at reduced loads and build back up over two to three weeks rather than jumping straight back to where you left off.

The interference effect (and why it's manageable)

If you've heard of the "interference effect," you might be worried that strength training and endurance training cancel each other out. The concern has a scientific basis: molecular signalling pathways for endurance adaptations (AMPK) and hypertrophy adaptations (mTOR) can compete when activated simultaneously. This was first identified by Hickson in 1980 and has been studied extensively since.

The good news for runners is that the interference effect is largely a problem for athletes trying to maximise muscle size alongside endurance. If your goal is neuromuscular efficiency, not hypertrophy, the interference is minimal. The meta-analyses consistently show that concurrent strength and endurance training improves running economy, VO2max, and time-trial performance without compromising endurance gains. The 2024 Sports Medicine review confirmed this across multiple study designs.

There are practical ways to minimise interference further. Separate strength and key running sessions by at least six hours when possible, or place them on different days. Prioritise the session that matters more: if today's primary goal is a threshold run, do it first and fresh. If it's a strength day, run easy in the morning and lift in the afternoon. And keep the strength work focused on neural adaptations (heavy, low-rep) rather than hypertrophy protocols (moderate weight, high-rep), which generate more muscle damage and more interference with running recovery.

Why this is hard to manage yourself

The principles above are straightforward. The daily execution, as with most training decisions, is not.

The challenge is that strength training adds a second system of fatigue on top of your running. A heavy squat session on Monday affects your legs on Tuesday, which affects your interval session on Wednesday, which affects your recovery on Thursday. The interactions ripple through the week in ways that aren't always obvious, and the cumulative effect only becomes apparent when something starts to hurt or your running performance dips for reasons you can't quite identify.

This is the interference effect in practice, not the molecular signalling, but the scheduling. When should you lift relative to your hard runs? How much volume can you handle this week given that your HRV has been trending down? Should you keep your maintenance session this week or skip it because the long run took more out of you than expected? Is the fatigue you're feeling in your legs from the squats or from the tempo run, and does the answer change what you do tomorrow?

These are the kinds of multi-signal decisions we described in our article on how AI adjusts your training in real time. An adaptive coaching system doesn't just manage your running load. It manages the interaction between your running and your strength work, adjusting one when the other creates more fatigue than expected, shifting sessions when recovery signals suggest you need more time, and ensuring the strength work supports your running rather than competing with it.

Most static plans treat strength training as a separate track, a PDF of exercises pinned to the fridge alongside your running schedule. But the two systems interact constantly, and managing that interaction well is the difference between strength training that makes you faster and strength training that just makes you tired.

The bottom line

The research is clear: marathon runners who do structured, heavy strength training run more efficiently, hold their form longer under fatigue, and arrive at the later kilometres of a race in better shape than those who only run. The improvement isn't dramatic on any single day. It compounds over months of consistent work and shows up when it matters most, in the final third of a race when everyone around you is slowing down.

The movements are simple. The periodisation is logical. The time commitment is modest. And the evidence base, after decades of study across multiple meta-analyses, is about as strong as it gets in exercise science.

Your legs are your engine. Strength training makes that engine more efficient. And efficiency, over 42.2 kilometres, is everything.

Zepho is adaptive coaching for serious runners. It's live and available now on the App Store.