The science of running in super spikes: New data on Nike, Adidas, and PUMA track shoes

Super spikes are track spikes for middle and long distance events that use the two key innovations in shoe tech that have revolutionized marathon shoes: “super foam” with high deformation and excellent energy return, and carbon fiber plates in the midsole.

I have a longer and more detailed article in the works that takes a deep dive into the biomechanics of super shoes more generally, but I wanted to get this post out since it’s track season and people want to run fast!

The impetus for this post was a fascinating new preprint on the benefits of super spikes from a multi-site collaboration involving research labs in Spain, the US, and Canada. The focus of the study is middle distance performance in super spikes versus traditional spikes.

Why it’s hard to tell if super spikes work

With marathon super shoes, it’s relatively easy to see if super shoes “work”—you get a group of runners, then have them run on a treadmill at marathon pace and monitor their oxygen consumption both in regular shoes and super shoes.

If oxygen consumption is lower on average when the runners are wearing super shoes, then you know that the super shoe is improving their running economy. This is exactly the experimental design that the original “4%” study used (and indeed, the last author of this new preprint is the first author on the original Vaporfly paper).^[1]

With track spikes, you have two problems: first, you can’t wear them on a treadmill. Second, you aren’t really interested in whether they work at marathon pace—you want to know if they work at 800m, mile, 5k, or 10k pace.

Unfortunately, you can’t test running energetics at paces faster than about half marathon pace: because you are relying on anaerobic energy production, oxygen consumption is no longer a good proxy for the energetic cost of running.

Testing the performance benefits of super spikes

This preprint is the first study I’ve seen that’s tackled the problem of super spike performance in a rigorous and systematic way.

The experimental design is both clever and elegantly simple: they recruited 12 middle distance runners and had them do 200 meter repeats at race pace (either 800m or 1500m effort, depending on their specialty) both in super spikes and traditional spikes.

To avoid the placebo effect, the researchers used a lightweight fabric sleeve to cover the upper on the spikes. The runners also were not shown their splits—they had to go by effort alone. The experiment was repeated again on a separate day.

With this experimental design, the results are incredibly easy to interpret—if super spikes work, you’ll run the same effort but hit a faster pace for your 200 meter repeats.

To validate the experimental method, the preprint first tested the effects of adding 200 grams of extra mass to a traditional spike (the PUMA EvoSpeed Distance 9).

This extra mass on the spike slowed the runners’ splits by 1.2%, which is exactly the slowdown you’d expect based on treadmill energetics studies conducted at slower speeds.

Previous energetics research done on treadmills has established that 100 grams of extra mass increases the energetic cost of running by 1%, and because of how energetic cost, speed, and air resistance interact, this translates to 0.6% slower per 100 g at 30-second 200m pace.

So, the fact that this protocol produced a perfect agreement with the known effects of extra shoe mass is strong evidence that it can reliably detect small but real changes in running speed.

Results: Nike vs. PUMA vs. Adidas super spikes for middle distance running

After validating the experimental design with the added-mass experiment, the researchers turned their attention to assessing whether current super spikes offer a real advantage compared to traditional spikes. This set of experiments compared four shoes:

• PUMA EvoSpeed Distance 9 (the traditional spikes)
• Nike ZoomX Dragonfly
• Adidas adizero Avanti TYO
• PUMA evoSPEED Long Distance Nitro Elite+

Here’s how each of the three super spikes stacked up compared to the traditional spikes:

Shoe	Speed increase	Statistically significant?	Midsole construction
Nike ZoomX Dragonfly	1.6%	Yes	Carbon fiber plate
Adidas adizero Avanti TYO	0.8%	No	Fiberglass rods
PUMA evoSPEED Long Distance Nitro Elite+	2.1%	Yes	Plastic plate

PUMA has an unreleased prototype that might be twice as good as Nike Dragonfly spikes

There was one additional experiment in the preprint that caught my attention—it was very much like the experiment above on Nike, Adidas, and PUMA spikes, except all three super shoes being tested were unreleased PUMA prototypes.

Here are the results from this final experiment:

Shoe	Speed increase	Statistically significant?	Intended event	Carbon fiber plate?
PUMA Prototype 1	1.8%	Yes	Middle distance	Yes
PUMA Prototype 2	1.1%	Yes	Long distance	No
PUMA Prototype 3	3.1%	Yes	Middle distance	Yes

No currently released PUMA shoe matches the design features of Prototype 3: it has a 19 mm thick Nitro foam midsole and a carbon fiber plate. To date, all of PUMA’s commercially available spikes only have a pebax (plastic) midsole plate.

This is a pretty exciting result, but we shouldn’t be too eager to compare these results directly to the Nike Dragonfly, since this experiment took place on a different track and involved different runners, so it’s not a 1:1 comparison. Still, given that PUMA’s commercially available evoSPEED Nitro Elite+ outperformed the Dragonfly in the study on commercially available shoes, that’s one piece of evidence that might strengthen the case that this unreleased PUMA shoe could be the best super spike yet.^[2]

PUMA, like all shoe companies, is bound by the World Athletics rules regarding open-market availability: any shoes that a PUMA runner wears at the Olympics must be available on the open market at least one month before the Olympics begin this summer.^[3]

So, if PUMA releases a middle or long-distance spike with a carbon fiber plate, you can be almost guaranteed that it’s a descendant of this third super spike prototype. Since PUMA has already announced a new sprint spike, I expect that a distance spike will follow.

For their part, Nike has already announced the Dragonfly 2, which launches on May 1st. Only more research will tell whether the Dragonfly 2 offers an improvement over its predecessor, or over PUMA’s prototypes.

Why super spikes work

One nice feature of this preprint is that the runners wore sensors on their waistband that measured cadence and stride length.

This simple measurement revealed that virtually all of the improvements from super spikes are from covering more ground per step, while maintaining the same cadence.

From this finding, it’s reasonable to infer that super spikes allow you to apply more force to the ground for the same effort level. As for why they allow you to do that, the answer is still unclear.

My own suspicion is that much of the benefit is because of the super foam. Super foam helps you run faster for two reasons:

1. Super foam dissipates less energy as heat, so your muscles do not need to do as much mechanical work to achieve a given change in your mechanical energy [1].
2. Super foam deforms (squishes down) more, which allows your muscles to operate at a more efficient place on the force-length and force-velocity curves. This means you can produce the same amount of muscular force with less energy expenditure.

Recent simulation studies using digital models of the musculoskeletal system suggest that both mechanisms are important for super shoe energy savings: better energy return and more deformation in the foam interact synergistically to decrease the energy cost of running [2,3].

The importance of carbon fiber plates in super spikes

If it’s all about the foam, why do super shoes have carbon fiber plates? The best explanation I’ve heard comes from Rodger Kram at the University of Colorado, who says that the plate allows you to apply force to more super foam at the same time—kind of like how snowshoes let you apply force to a bigger area of snow at once.

This plate/foam interaction is the only explanation I’ve heard that explains the most interesting puzzle in super shoe research: if you take a pair of Vaporflys, then use a coping saw to cut the carbon fiber plate into several pieces (leaving the foam mostly intact), these lobotomized super shoes still reduce the cost of running by the same amount as intact shoes [4]!

The findings from this preprint do suggest that the carbon fiber plate is doing something, for two reasons:

First, the Adidas super spike has fiberglass rods, not a carbon fiber plate, and it underperformed compared to the Nike Dragonfly, which does have a carbon fiber plate. Fiberglass rods might store bending energy, but they probably wouldn't have a strong foam/plate interaction effect. 

Second, the PUMA prototype which featured a carbon fiber plate outperformed a nearly identical prototype which had no carbon fiber plate. This suggests the plate is doing something important, though it’s not clear if that’s energy storage or a foam/plate interaction. 

More research on the biomechanics of super spikes is needed to determine what role energy storage in the plate itself contributes to improved performance in super spikes.

I am skeptical of the “bending the plate stores energy” explanation because the actual amount of energy stored in the plate is tiny, and the only real biomechanical effect is seen at the toe joint —not exactly a prime mover in running [5].^[4]

But maybe the plate alters how your body can apply force to the ground, which has cascading effects on muscle energy expenditure. Only careful musculoskeletal modeling studies can tell.

In any case, PUMA’s current super spike uses a plastic (pebax) plate—based on this research I’d be very surprised if they don’t move to carbon fiber for their newer shoes.

Individual differences in super spike response

Even if one super spike or super shoe is better than another on average, it doesn’t guarantee that everyone will respond to that shoe the same way.

Much of the research on individual responses to super shoes is not very reliable, because to get a valid estimate of a given runner’s actual individual response to a given shoe, you need to take multiple measurements in each pair of shoes [6].

Though this new preprint was mostly focused on average responses, it did look at individual responses a bit—albeit only in the unreleased PUMA prototypes. The results suggested that about three-quarters of people had the same relative response across multiple days. In other words, if you ran faster in Prototype 1 vs. the traditional shoe in the first workout, there was a ~75% chance you’d also run faster in that same prototype for the second workout.

Unfortunately, we’re still very far from identifying the reasons why one individual runner might benefit more from a given super shoe or super spike, largely because small studies like this one can only reliably identify differences within an individual (e.g. whether one shoe works better than another), not differences across an individual (e.g. what biomechanical features of gait are predictive of a super shoe’s benefit).^[5]

Rules regarding super spikes: are super shoes and super spikes legal on the track?

World Athletics has a number of regulations regarding super spikes and super shoes, but unless you are a world-caliber elite runner, the rules do not apply to you. However, the World Athletics rules do have some bearing on the shoe market more broadly. Here are the relevant details for super shoes and super spikes.

All super shoes and super spikes must be available on the open market

Elite athletes are not allowed to compete in “custom” prototype shoes. Any shoe worn at the Olympics or World Championships must be available to the general public on the open market for at least one month before the competition.

This is great news for the rest of us, because it basically guarantees that every major competition cycle will bring the latest shoe tech to market. You can see this in spring 2024 with the cascade of new track and road racing shoes from all of the major players. 

Track super spike are limited to stack heights of 20 mm or less

Under new World Athletics guidelines, no track shoe can have a thicker stack height (i.e. midsole thickness) greater than 20 mm, or about 0.75 inches. This rule is only enforced at elite, professional-level events like the Olympics. The World Athletics rules specifically state that this stack height limit does not apply to club, high school, or college competitions.

Even the NCAA has no restrictions on shoe thickness (except for high jump), and at major college meets you’ll routinely see 10k runners wearing marathon super shoes that are much thicker than 20 mm.

Since almost every “normal” running shoe is thicker than 20 mm, there is no possible way this rule would ever be enacted at the high school level—plenty of JV runners don’t own spikes of any kind, so they’re routinely competing in normal running shoes with stack heights over 20 mm.

The stack height limit does matter, though, because it means that all the best super spikes will have a stack height of 20 mm or less. All the PUMA prototypes in the preprint from above, for example, had 18 or 19 mm stack heights.

To be clear, 20 mm is quite thin: even lightweight old-school style racing flats like the Nike Streakfly, Saucony Sinister, and the New Balance FuelCell SuperComp Pacer have stack heights well over 20 mm.

In practice, this rule means there’s going to be two categories of racing shoes to choose from for long-distance track races: super spikes with a stack height of 20 mm or less, and marathon super shoes with stack heights of up to 40 mm. If your feet get beat up in thin spikes and racing flats, you should strongly consider marathon shoes, especially for the 10,000 meters.^[6]

I think this rule is understandable, but also bad news for injury-prone 10k runners: doing 10,000m on the track in spikes can really wreck your feet, ankles, calves, and Achilles tendons. Meanwhile, a 10k in super shoes is much easier on your legs, so it's unfortunate that track athletes who run the 10k don't have the option of wearing something less destructive, especially if they're racing a lot--for example, an early-season 10k, then conference, regionals, and nationals.

Recap

Compared to a traditional spike, a top of the line super spike from Nike or PUMA can help you run 1-2% faster in middle distance events. That’s equal to about 1–2 seconds in an 800 or 3–6 seconds in a mile. Though this study did not test performance in long-distance events, it’s quite likely that similar benefits extend up to the 3k, 3200m, 5k, and 10k.

As of right now, Nike and PUMA’s best super spikes outperform Adidas’ spikes by 0.8-1.3%, though not by a statistically significant margin. PUMA appears to have an unreleased super spike that could be twice as much of an improvement on traditional spikes as Nike’s Dragonfly, though Nike is also debuting an updated Dragonfly this summer which might also be an improvement over the original iteration.

If I were racing a spring track season, I would try on the Nike ZoomX Dragonfly and the PUMA evoSPEED Long Distance Nitro Elite+ and go with whichever shoe felt more comfortable. If I had a really big competition later in May, June, or July, I might try to score a pair of Dragonfly 2s or, if they are released in time, PUMA’s as-of-yet unnamed carbon fiber spike.

If you only dabble on the track, or if you strictly run the 5k and 10k, a marathon super shoe is a perfectly good replacement for super spikes, and in fact might even be better: because marathon shoes only have to comply with the more relaxed 40 mm stack height limit, they can pack much more super foam in between your foot and the ground.

Note to the reader—I don’t have any special insider info on the shoe industry. I just poke around on Google Scholar a lot more than the average runner! I don’t have any conflicts of interest here: there are no affiliate links for shoes in this article and I have never done research for any shoe company.

Find out about the latest running research

If you enjoyed this article, sign up for my email newsletter! I have more articles in the works on the latest research findings on performance, injuries, and more. My newsletter is the best way to find out when I release new content.

Support my work

If you want to learn more about training for middle distance races, you should check out my two books!

I have a brand-new book out titled Marathon Excellence for Everyone, which is your comprehensive guide to modern marathon training. If you or your athletes are going to move up to the marathon some day, you will absolutely want to check it out.

If you're more focused on 800m, 1500/1mi, 3k/3200m, 5k, or 10k, check out Modern Training and Physiology - it's a simple primer on both the physiology and the practical aspects of training for shorter distances. It's a great resource both for runners and their coaches.

Footnotes

---

^[1] The lead author on this new preprint works for PUMA, which explains why the researchers were able to get access to the unreleased PUMA prototypes.

^[2] Technically, the difference between the Nitro Elite+ and the Dragonfly is only a numerical one; the difference was not statistically significant. As the Adidas results show, differences of <1% are hard to detect with this type of experiment.

^[3] As the mad scramble to buy Nike Alphafly 3s shows, “available on the market” does not guarantee you’ll actually be able to get your hands on a pair! Shoe companies are not actually required to restock the shoe if it sells out, per World Athletics rules, though any flagrant abuse of this provisio will likely result in a rule change.

^[4]  One of the most remarkable things about super shoes, in my opinion at least, is how little your gait mechanics change. The ground reaction forces and the joint torques at your knees and ankles are basically identical in a traditional trainer vs. a super shoe. To me, this supports the muscle energetics idea: you basically use the same whole-body movement pattern, but with better muscular efficiency.

^[5] The technical reason here is that repeated-measures studies like this one gain a lot of statistical power by letting the individual act as his or her own control group—any individual idiosyncrasies or variations effectively cancel out, leaving only the variation due to the different shoes. Comparisons on things that vary across people, like say cadence or vertical oscillation or body weight, need much bigger sample sizes to overcome the greater “noise” from other aspects of individual variation.

^[6] Racing flats like the New Balance FuelCell SuperComp Pacer and the Saucony Sinister appear to comply with World Athletics’ interim rules from 2022, which allowed up to a 25 mm stack height for events longer than 800 meters; I suspect some shoe companies will come out with 20 mm compliant racing flats reminiscent of the most extreme of the minimalist days, e.g. the New Balance RC5000, which had only 16–17 mm of foam.