Hello and welcome to the Endurance Lab newsletter.

In last week’s newsletter, I talked about my training philosophy and dropped a podcast covering it in more detail. I’ll keep the intro quick this time, but if you haven’t listened already, go check it out. It’s pretty in-depth and covers how I think about training from a broader perspective.

Over the next few newsletters, I want to get into the specifics of my theory e.g., what I’m trying to develop, which workouts I use, and why I might change something like rep length or recovery depending on where we are in the season.

We’re starting with running economy.

I’ve covered this before, but I want to get more practical this time. What can you actually do to improve it? And when you write a workout, how do you know whether you’re giving yourself a good opportunity to develop it?

Strength training and running economy (RE)

First and foremost, I’d start is strength training and plyometrics. The data is simple and consistent that this improves RE. You can add these to your training without having to completely restructure your running workouts.  [1]

For context, running economy describes how much energy you use to run at a particular speed. Typically we measure this as an amount of oxygen and then normalize that to a particular distance. Think about it like sub max heart rate. If you can run at any given pace with less oxygen, you’re more efficient.  [1]

That brings us to a study by Dr. Michele Zanini and colleagues, published in 2025. The researchers had well-trained male runners complete 10 weeks of either endurance training alone or endurance training with added strength and plyometric work. They then assessed running economy throughout a 90-minute run. [2]

This was the routine roughly. It’s really well done to be honest. [2]

The interesting finding was that the benefit became apparent later in the run. The runners who added strength training improved their ability to preserve running economy as the run progressed. [2]

Blue line is the strength group

In essence, they became more durable. This is the physiological resilience idea I’ve discussed before, applied specifically to running economy. If you only assessed someone when they were fresh, you could miss an adaptation that matters when they’ve already been running for an hour. [2]

What does the broader research show?

Of course, that’s one study. What happens when you look across the literature?

A 2024 meta-analysis by Llanos-Lagos and colleagues compared high-load resistance training, submaximal-load training, plyometrics, isometrics, and combinations of methods. [1]

The high-load group had a beneficial effect size of about 0.27, while combined methods had an effect size of about 0.43. The authors described these as small and moderate improvements, respectively. [1]

Combined methods on RE [1]

Before we go further, I want to explain what those numbers mean because I think this is where people can get a little lost.

A quick detour on effect size

Imagine a room of people. One group adds Pop-Tarts to their usual diet, and the other keeps eating as usual. You measure their body weight before and after six months.

Let’s assume both groups start at an average of 150 pounds. At the end, the Pop-Tart group averages 155 pounds, while the control group still averages 150.

The difference in average change is five pounds.

A standardized effect size expresses that difference relative to the variability in the measurements. In a simple example, you divide the five-pound difference by a pooled standard deviation.

Standard deviation describes how spread out the values are. If everyone weighs exactly 150 pounds, it’s zero. If three people weigh 151, 152, and 153 pounds, their sample standard deviation is one pound.

So, if our five-pound difference came with a pooled standard deviation of 10 pounds, the standardized effect size would be 0.5. If the standard deviation were 20 pounds, the effect size would be 0.25.

This is the part to get: Same five-pound difference. Different standardized effect size. This is why a well controlled study will have far larger statistical effects than a poorly controlled one.

That’s why I want to know the actual change alongside the effect size. A value of 0.27 doesn’t mean running economy improved by 0.27%, and it doesn’t tell you directly how many seconds someone might gain in a race.

There’s also a lot that can contribute to variability in a training study. People start at different fitness levels, respond differently, and don’t necessarily complete every session in exactly the same way, and measurement error adds another layer. You can reduce some of that variability by having everyone complete every condition, so each person serves as their own control, but that gets difficult with training interventions. For example, if someone completes 10 weeks of strength training first, how do you know those adaptations aren’t influencing their results when they move into the next condition? A reviewer would ask that, and they’d have a point.

This is why training studies often compare separate groups. It makes sense for the question, but it also means you’re comparing different people.

None of this means a small effect size is secretly a huge benefit. It means you need to look at the actual changes, the uncertainty around them, and how the study was designed before deciding how useful the result is.

Why I still read the individual studies

Back to the meta-analysis.

Submaximal-load training didn’t show a statistically clear pooled benefit, and that analysis included only three studies. Plyometrics had a near-zero overall estimate, although the researchers found a benefit when looking specifically at lower running speeds. [1]

One study that caught my attention was Lundstrom’s 2017 paper. [3]

At first glance, its contribution to the forest plot could make you think the plyometric group became less economical. But the comparison was against a core-training group, and both groups were also training for a marathon. Both groups improved running economy. The core group improved numerically more, without a statistically significant difference between groups, and that’s a very different result from saying plyometrics made runners less economical. My point is that the comparison matters. You’re asking whether adding plyometrics produced a greater improvement than adding core training during that marathon program.  [3]

This is also where sensitivity analyses can be helpful. You repeat the analysis after removing a particular study and see how much the overall estimate changes. If the conclusion changes substantially, that tells you the result depends heavily on which studies are included. You still need a defensible reason for excluding a study.

I’ve done a meta-analysis myself, on resistance training and arterial stiffness. It has nothing to do with running, but it gave me an appreciation for how many decisions go into one of these papers. You read a ton of studies, establish consistent methods, and try to combine results from protocols that are often quite different. Then most readers, understandably, look at the forest plot and the conclusion, and I do that too. But when a result seems surprising, I think it’s worth opening the individual paper and checking what the researchers actually compared.  [4]

What I’d have you do

Taking all of this together, I think heavy strength training and appropriately progressed plyometric work deserve a place in a runner’s program. If it’s up to me, I’d generally lean toward heavier loads and lower reps. Something like 3 × 5 or 5 × 5, depending on your experience, the exercise, and what else you’re doing that week.

I’d use movements such as squats, lunges, or rear-foot-elevated split squats, alongside calf work. Those are examples of how I might build the program rather than a claim that one particular exercise or rep scheme is uniquely best. I like to use both bilaterial and unilaterial movements and alternate them week to week or session to session. I typically group them together and prioritize some form of triple extension in the posterior chain during a power and strength movement.

I also like including some elastic work. This means bouncing or hopping.

For example, a 2023 study by Engeroff and colleagues used a simple, progressive daily hopping program in amateur runners. After six weeks, running economy improved at 12 and 14 km/h, without a clear improvement in peak oxygen uptake. [5]

You don’t necessarily need a complicated menu of jumps to get something useful out of this type of training. I’d start with an amount you can perform well and recover from, then progress it alongside the rest of your training.

How this carries over to running workouts

All right, this newsletter is already getting long, but I want to briefly connect this to how I write running workouts. Consider 8 × 400 m at mile pace with one minute of recovery. That’s a demanding session. As the reps accumulate, you’re asking the athlete to keep producing the pace with progressively more fatigue. Now take the same eight reps at the same pace and give them three or four minutes of recovery. The distance and pace haven’t changed, but the experience of running the workout has. With longer recovery, the athlete has a better opportunity to approach each rep feeling recovered and run the pace in a controlled manner. If my priority that day is practicing the speed and coordination of running at mile pace, I might choose the longer recovery. I want the athlete to get repeated exposures to that pace while they’re still moving well.

I wouldn’t claim that this exact comparison has been shown to produce a greater improvement in running economy. I also wouldn’t say that three or four minutes guarantees complete recovery, or that a runner who looks smoother necessarily has a lower oxygen cost. Running economy needs to be measured. But I do think recovery deserves more attention when we’re deciding what a workout is supposed to accomplish. Perfect practice makes perfect is how I tend to think about these sessions. I want enough recovery that the athlete can keep practicing what I’m asking them to do.

As we move closer to racing, I also want them to sustain that pace as fatigue develops. Shorter recovery can be useful there, and it creates a different demand. Some coaches argue that fatigue itself can help develop economy because the body learns to complete the task more efficiently. I see the reasoning, and I think learning to maintain your movement under fatigue is important. Where I get cautious is assuming that completing the workout tells us exactly which quality improved. An athlete might hit every 400 comfortably in training and still struggle halfway through a mile race. There are several possible explanations: pacing, aerobic fitness, fatigue resistance, or simply relying on recovery between reps that they won’t get during the race.

I wouldn’t diagnose poor running economy from that alone. What I would do is look at the sessions we’ve been giving them. Have we provided opportunities to practice race pace while reasonably fresh? Have we developed the ability to sustain it as fatigue builds? Are they recovering enough to benefit from the work? Those questions influence whether I use shorter reps, longer reps, or more recovery.

For now, I’d start with strength training, add a manageable amount of plyometric work, and be deliberate about the recovery in your running sessions.

 That’s all for now… cheers mates. Dr. E

References

1. Llanos-Lagos C, Ramirez-Campillo R, Moran J, Sáez de Villarreal E. Effect of strength training programs in middle- and long-distance runners’ economy at different running speeds: A systematic review with meta-analysis. Sports Medicine. 2024;54(4):895–932. https://doi.org/10.1007/s40279-023-01978-y

2. Zanini M, Folland JP, Wu H, Blagrove RC. Strength training improves running economy durability and fatigued high-intensity performance in well-trained male runners: A randomized control trial. Medicine & Science in Sports & Exercise. 2025;57(7):1546–1558. https://doi.org/10.1249/MSS.0000000000003685

3. Lundstrom CJ, Betker MR, Ingraham SJ. Effects of plyometric and explosive speed training on recreational marathoners. Journal of Sports Science. 2017;5:1–13. https://doi.org/10.17265/2332-7839/2017.01.001

4. Evans W, Willey Q, Hanson ED, Stoner L. Effects of resistance training on arterial stiffness in persons at risk for cardiovascular disease: A meta-analysis. Sports Medicine. 2018;48(12):2785–2795. https://doi.org/10.1007/s40279-018-1001-6

5. Engeroff T, Kalo K, Merrifield R, Groneberg D, Wilke J. Progressive daily hopping exercise improves running economy in amateur runners: A randomized and controlled trial. Scientific Reports. 2023;13:4167. https://doi.org/10.1038/s41598-023-30798-3

Book a free coaching consultation call if you are interested in joining the team, here: https://www.endurancelaboratory.com