Paula Radcliffe is one of the greatest marathon runners of all time. At the start of her career, she was very good but not dominant. An analysis of her progress as she came onto the world stage revealed a significant increase in running economy with a significant decrease in V̇O2max as she rose from a solid junior athlete to an international star. So, while her maximal aerobic ability declined, her running economy improved and her endurance performance took her to the top of the sport. Similarly, one of the earliest longitudinal studies in running was on the miler Steve Scott; his improved performance was due to improved running economy without a change in V̇O2max. So, while maximal aerobic capacity is important for endurance runners, running economy matters more.

By 1980, it was recognized by Conley and colleagues that "among highly trained and experienced runners of comparable ability and similar V̇O2max, running economy accounts for a large and significant amount of the variation observed in performance on a 10 km race.” In 1985, the legendary coach and scientist, Jack Daniels published an article that was a call-to-action on developing a better understanding of running economy as an independent and relevant endurance performance metric, noting the numerous examples of runners with better economy beating runners with higher V̇O2max values. Over then subsequent decades, many studies concluded that running economy is a critical factor in endurance running performance. In 2006, a study of elite distance runners from Eritrea and Spain, with matched V̇O2max values, concluded that the reason for the superiority of the runners from the African nation was primarily their significantly greater running economy. 

How can you improve your running economy? There are three training approaches I recommend: heavy resistance, plyometrics (including sprinting), and mileage.


Since the 1980s, studies from many research groups have repeatedly demonstrated the positive effects of heavy resistance training on endurance capacity and performance. For example, Hickson and colleagues reported that a heavy strength training program improved indices of endurance performance. In a 1995 study of female collegiate distance runners, a strength training program improved running economy by about 4%. In that study the runners progressed from 2-3 sets of 10 repetition maximum (RM) to 2-3 sets of 5 RM over the course of 10 weeks. There were no changes in body mass or fat measurements, V̇O2max, or limb circumferences. Indeed, it was was recognized long ago that there's a difference between strength and size and that it’s most desirable to improve strength without increasing size for runners. This is easily accomplished following current guidelines, which I outline below. In 2008, Støren and colleagues reported that maximal strength training improved time-to-exhaustion at maximum aerobic speed and running economy at 70% V̇O2max, which is the approximate average intensity for many ultra-marathons.

Meta-analyses are studies of studies - they compile all the previous unique studies on a subject with the purpose of leveraging the power of numerous studies to draw generalizable and broadly reliable conclusions. A 2006 meta-analysis reported that “maximal-force training led to greater improvements than other intensities. Subgroup analyses also revealed that beneficial effects on performance were consistent irrespective of the athletes’ level.” According to Dr. Keith Baar, my guest on episode 58, one set of very heavy weight lifting moved slowly and to fatigue in about 8 repetitions is sufficient. You don’t need to worry about gaining muscle mass on this sort of program. It’s weight lifting volume (lots of reps and sets) that causes gains in muscle mass while one or two sets with near maximum effort develops strength without bulking up, as shown in this study.

Fast forward to 2018 and multiple meta analyses (e.g., here and here) found that strength training improves economy of motion for endurance athletes from many sports, including running - both moderately and highly trained distance runners - and that the effect is most prominent for maximal-force training and less, but still significant, for maximal-power training. What this means is that a slow, maximum effort is the most effective approach. Again, avid listeners of SOUP may recall Dr. Keith Baar in episode 58 explaining that even a single set is effective and sufficient to strengthen the connective tissue linkages within muscle. Stronger linkages from the muscle cell proteins, like dystrophin, to the extracellular collagen-based matrix results in more effective tension development throughout the muscle. Slow, heavy resistance does this while reducing stiffness in the muscle end of the tendon. When we lift heavy, there is a huge amount of stress between the fibers of the muscle, and this causes the muscle to reinforce the connections between fibers making the extracellular matrix (muscle connective tissue) in the muscle stiffer. At the same time, similar shear forces in the tendon cause the collagen fascicles to slide relative to each other breaking crosslinks and decreasing stiffness at the muscle end of the tendon. The result is a decrease in tendon stiffness, but an increase in muscle stiffness. We think that this decreases muscle injury while maintaining very good energy return.


If you’ve not been doing weight training, then start with lighter resistance and don’t go all the way to failure until you see how your body is responding and how sore you get. We don’t want excessive soreness, especially when starting out. My athletes who haven’t been doing resistance training might spend the first 2-3 weeks doing one set of squats with 20 repetitions 3-5 days per week using a resistance that causes them to just begin feeling a light leg burn by the end of the set.

I suggest doing brief, heavy-resistance workouts twice per week early in your season as you develop endurance then once per week mid-season as you begin focusing on economy, and none in the 6-8 weeks before a goal race. For example, early season it’s 2 days per week doing one set each of two legged squats, single leg squat or leg press, and standing calf raises each for 8 repetitions to near failure. That’s a total of 4 sets per workout, easily done in just a few minutes. More is not better. Don’t overdo heavy resistance training. The workouts can be done any time of day. Before a run, after a run, or at another time of day are all okay.

It’s my hypothesis, that heavy resistance training will be more important for athletes with a history of muscle pulls and for those who have or do run mostly flat terrain, especially road and track runners. Running down steep hills, like in trail running with a lot of vertical change, requires a lot of eccentric muscle contraction, which places a lot of stress on intramuscular linkages and is likely to enhance intramuscular tension capacity with long term adaptations similar heavy resistance training. So, if you’re a trail runner already running hard downhills, heavy weight lifting may be less useful than if you’re a road marathon runner.


While heavy weight lifting strengthens muscle and reduces stiffness at the muscle end of tendons, plyometrics increase tendon stiffness for improved energy return. Plyometrics are jumping and bounding exercises. As you begin to emphasize economy in your running workouts, heavy-resistance workouts should be replaced with plyometrics. This is why my athletes do heavy resistance training during the endurance-building early-season and transition to plyometric exercises in the economy-building mid-season.

The first phase of plyometrics should be introducing basic movements 3-4 days per week such as skipping, jumping jacks or jumping rope, and short segments of easy bounding where you run and bounce to increase flight time just a little for 5-10 strides. After you can comfortably accumulate a few of minutes of each of these in 15-30 second bouts, the program can progress to running-specific plyometrics, sprinting and uphill running.

Sprints and uphill running bouts of 15-30 seconds have been shown by several research groups to significantly improve running economy, even in well-trained endurance and ultra-endurance runners. Indeed, explosive running really is the most running-specific form of plyometrics.

The idea of repeated sprints may strike fear in those who recall running repeated sprints in sports like soccer back in your glory days. But fear not. Those sprints were often run with short rest periods such that you were nearly broken by the end of numerous sets of several repetitions and accumulating fatigue. That is not what we need for developing running economy. In fact, we want complete recovery between repetitions. That’s because the stimulus we’re after is promoting structural, neuromuscular, and non-metabolic (non-enzymatic) intracellular adaptations. We specifically are not trying to wear down over the course of repetitions but to maintain peak power and activate as many muscle fibers in the active skeletal muscles as possible. So, recovery between repetitions needs to be complete. Three to four times the bout duration is fine. Short sprints should have big rests (and this).

The mechanisms by which sprints improve running economy have been investigated. One of my favorite studies in 2018 was published from the lab of the legendary exercise physiologist Jens Bangsbo in Copenhagen, Denmark. They discovered that 30-second maximal sprints altered the expression of specific proteins in both slow-twitch and fast-twitch motor units. A motor unit is one neuron from the spinal cord and the group of skeletal muscle cells that it controls; episode 59 with Dr. Roger Enoka goes into more depth on motor units if you’re interested. Slower-twitch and faster-twitch motor units refers to the groups of muscle cells you use for running easy and slow up to powerful and fast, respectively. The proteins that the Bangsbo lab found altered by sprint training are ones that influence the efficiency of cellular energy production, tension development, and force transmission. They further found that sprint training improved 10K running performance due to adaptations in both slow- and fast-twitch motor units and that running economy at 60% of maximal steady state, which is similar to the pace for ultra-marathons, was improved through these adaptations in slow-twitch motor units.

How can you include sprints in your economy training? On two non-consecutive days per week, run 5-10 sprints of 15-30 seconds with complete recovery between. Recovery can be walking or easy jogging. The Bangsbo lab used a recovery duration of 2.5 minutes between 30 second sprints - that’s 5 times the bout duration. Many runners will be fully recovered in less time. What’s important to appreciate is that shortening the rest period does not help you, but it may hurt the quality of your sprints if you cut it too short.

Here are some key resources on sprints for improving running economy:
*15-20 second up steep hill; Barnes lab study and review paperBarnes & Kilding SOUP episode
*30 second shuttle runs; Millet lab study  Millet SOUP episode
*10-40s sprints all out with >5x recovery (e.g., 10x30sec all out twice per week with 2.5 min walking recovery) Bangsbo lab study

The athletes I coach run 15-30 second sprints during economy building phases. Are these on flats, inclines, or declines? Is there a difference among 15, 20, 25, and 30 seconds? The studies show that the differences probably don’t matter much in terms of efficacy for improving economy. You want to be careful of injury risk. Hard sprints downhill, for example, are a higher injury risk than the same sprints uphill. You also want to be aware of meeting the goal, which is to sprint. So, if you choose an incline so steep or a duration so long that you begin fatiguing and slow down, then it’s too much. It’s very important that you can complete all of the intervals without slowing down or fatiguing significantly. Your legs should not be feeling heavy at the end of uphill sprints; if they are, it’s too steep or you’ve gone too long. You should finish these workouts feeling tired but a bit energized and you should always stop while still being able to do a few more before fatigue would manifest. The rule of thumb I like to use is 30 seconds flat or slight decline (1-2%) to 15 seconds at 15% incline. Basically, you start with the notion of 30 seconds flat or at a slight decline and then subtract 1 second for every 1% rise in incline. 30 seconds for zero, 25 seconds for 5% incline, 20 seconds for 10% incline, 15 seconds for 15% incline. But, you don’t have to be too precise with this; it’s a general approach. Let experience guide an empirical decision for the time and incline relations you choose.

If an athlete asks me for one thing they can do to improve running economy, my answer is running hills. Running fast uphill some days and running hard downhill on other days is, in my opinion, the best single approach. You get the stimulus for improved tendon tension on rapid uphill sprints and the increased intramuscular strain for development of intramuscular connective tissue in the eccentric loading of downhill running. Running stairs probably provides a similar benefit if you don't have hills.


We started this topic thinking about Paula Radcliffe’s rise to prominence. While her V̇O2max was declining and her running economy improving, her overall volume increased substantially as well. It’s thought that the more running you can do, the more economy you’ll develop. Of course you have to stay free of injury and there will be a limit on how much improved economy you can get from increasing volume. For many years, top running coaches have emphasized the importance of steadily increasing running volume at sustainable efforts no higher than marathon pace for improving endurance running performance.

In research on junior and elite national level endurance athletes across several sports in work from several scientists including Dr. Stephen Seiler, my guest in episode 71, improved performances over seasons and years track with increases in volume of easy training. We’ll have an episode on zones of intensity in a future episode but easy is zones 1 and 2 in a 5-zone model and zone 1 in a 3-zone model. It's one of the reasons I say that ‘recovery runs’ don’t exist. While it’s true that you may go on an easy run following a hard day of training, and therefore think of it as a day in which you're recovering, to me that’s a net recovery day because there’s more recovery going on than breakdown. The easy run you go on is stimulating improvements in your endurance capacity regardless of what you did the day before. It may be a run you do while the net for the day is recovery but the run itself is its own stress.

While we’re on the subject of building volume at easy paces, it’s important to note that you must maintain your overall volume of easy running as you gradually add economy training. In studies where volume of easy running was reduced with the addition of high intensity training, subjects got faster for short duration running but lost important features of the capacity for endurance. So, add heavy resistance training and/or plyometrics and sprints into the training while at least maintaining your volume of easy running.

A final note on developing overall economy for ultra-marathons. There may be a lot of hiking in your longer races. Hiking is different from running. Walking is different from hiking. A lot of running does not make you a good hiker. And, walking around at work and throughout your daily life does not make you a good hiker. Hike fast in training if hiking is part of your races, especially in the final 4-6 weeks before a long ultra-marathon. You don’t just want your running to be faster, you also want your hiking to be faster. I’ve been out on trails with veteran long-distance backpackers and it’s astonishing how fast and efficiently they hike, faster than many runner's jogging pace. For most people who don’t do a lot of hiking, it’s easier to cut minutes off your hiking speeds than it is to cut seconds off your running speeds.

That brings our summary of key features in developing running economy to a close. We previously explored development of endurance as part of our dive into the plan to develop stamina (script of that episode here). In the next post on training your body, we’ll explore the basics of interval training to enhance stamina once you have a foundation of endurance and economy.

Remember to keep it all in perspective. We run because we love running. If diving into these details detracts from your joy or adds negative psychological stress to your running, then just go out and run. Fun always comes first.

Some additional studies that may be of interest:

Eur J Appl Physiol. 2003 Mar;89(1):1-7. The effect of plyometric training on distance running performance. Spurrs RW1, Murphy AJ, Watsford ML. PMID:12627298

J Strength Cond Res. 2006 Nov;20(4):947-54. Short-term plyometric training improves running economy in highly trained middle and long distance runners. Saunders PU1, Telford RD, Pyne DB, Peltola EM, Cunningham RB, Gore CJ, Hawley JA. PMID:17149987

J Strength Cond Res. 2003 Feb;17(1):60-7. Improvement in running economy after 6 weeks of plyometric training. Turner AM1, Owings M, Schwane JA. PMID:12580657

J . Appl. Physiol. 86(5):1527–1533, 1999. Explosive-strength training improves 5-km running time by improving running economy and muscle power. Paavolainen et al.

"CONCLUSION: Explosive training and heavy weight training are effective concurrent training methods aiming to improve RE within a few weeks. However, long-term training programs seem to be necessary when the largest possible improvement in RE is desired.”

"Running economy (RE) was measured in 20 of the studies and generally showed improvements (2-8%) compared to a control group, although this was not always the case. Time trial (TT) performance (1.5-10 km) and anaerobic speed qualities also tended to improve following ST. Other parameters [maximal oxygen uptake ([Formula: see text]), velocity at [Formula: see text], blood lactate, body composition] were typically unaffected by ST.”

An important factor in minimizing the energetic cost of running includes "low thigh antagonist-agonist muscular coactivation”. Strides, sprints, hill surges are intended to reduce coactivation by improved neuromuscular recruitment patterns. They may also improve other factors: "alignment of the ground reaction force and leg axis during propulsion” and "low activation of lower limb muscles during propulsion”. Plyometrics and heavy strength training may result in "greater leg stiffness”, another important factor.