Why eccentric exercise can help achilles tendinopathy

Posted on October 26, 2016


A disturbing trend that appears to be coming back is emphasizing a need to load painful conditions. I’ve worked with a significant amount of 50 year olds and up. I consistently have to remind them to not go into pain when exercising. “Oh, I thought if it hurt that means I need to make it stronger.” A fancier version of this is fat prairie doggin’ its way back.


Trying to get some traction but having trouble.

One condition that always gets roped into the load-into-pain-mentality is achilles tendinopathy. (Your achilles has been hurting you for a while.) The reason for this is a study done nearly 20 years ago showing those who eccentrically strength trained their achilles improved, compared to other modalities (rest, ice, conventional physical therapy, etc.). These people, legend has it, were told to go into pain for their exercise as well. From there, others piggy backed.

From a logical perspective, it’s troubling to state things like “you better be loading this issue,” or “you need to be loading achilles tendinopathy.” Some taking this to “you need to work through pain when doing this loading.” As if going into pain is a necessity. One way it’s been phrased:

“You better be putting strength on this dysfunction.  If your achilles hurts you load it everyday with strength exercises, you push it hard, you ask it to adapt.”

“…we sell the patient on the fact that the body adapts to stressors.  That pain is normal and doesn’t mean damage.  That it would be weird not to have pain.  That we can keep participating with pain and that they will get better as they adapt.”

Sounds reminiscent of a football coach from the 70s. “No pain, no gain!” Or some hippie life counselor. “If you can see it and you can believe it, you can be it!”

The most relevant population for this condition is runners. Are runners not already loading their achilles, pushing it hard, asking it to adapt???

Microscopically –something under appreciated– it’s been found the matrix of the tendon is buckling in achilles tendinopathy.

3-D ultrastructure and collagen composition of healthy and overloaded human tendon: evidence of tenocyte and matrix buckling

When something is buckling it’s not because it needs more loading! And good luck selling how this isn’t indicative of damage.

“Perhaps the tendon is weak, that’s why it’s buckling. Thus, strength training is needed.”

Again, is running not a means to strengthen the achilles? “Indeed, six out of seven studies found a larger cross sectional area in runners compared with non-runners,”

Achilles tendon adaptation and Achilles tendinopathy in running

If we took it as true that strength training is needed, wouldn’t we at least want to wait for the tendon to calm down / reorganize before loading it? Before “pushing it hard” “every day”? We fix a damaged bridge before we throw more cars on it, don’t we? Sure, humans are not inanimate structures, but the analogy holds. Plus, if we took this tendon-is-weak-need-loading logic to its conclusion, then that would mean in something like a partially torn tendon we need to strengthen it too. Shouldn’t we let it repair first?

Or we could say, “Give the area a break from loading, then begin to load it again,” but that’s a very different statement than “load it every day, push it hard.” As is “You don’t want to completely eliminate load, but you need to significantly decrease it. Carefully increasing load along the way.”

-> I don’t believe this is poor communication, but maybe it is. I believe some have been hit by the pendulum swinging back the other way and now genuinely think “We need to put people into pain to get them out of it.” Or that going into pain is fine.

Furthermore, one of the hallmarks of achilles tendinopathy, one of the ways it is diagnosed in the hallmark study of eccentric exercise, is hypertrophy!

“All patients in both groups had degenerative changes (tendinosis) in the tendon 2 to 6 cm above the Achilles tendon insertion on the calcaneus. The diagnosis was based on pain located in the Achilles tendon for at least 3 months. In all patients, localized degenerative changes in the tendon, corresponding to the painful area, were seen on ultrasonographs. All ultrasonography examinations were done by the same radiologist, and the degenerative changes were described as local thickening of the tendon, irregular tendon structure and fiber orientation, and separated tendon fibers.”

Heavy-Load Eccentric Calf Muscle Training For the Treatment of Chronic Achilles Tendinosis

Since when do tendons, or any tissue, thicken in response to not enough loading??? Don’t things hypertrophy when they’re getting pushed hard? (Meaning the achilles has adapted to the loading placed on it. It just can’t adapt enough. Hence, pain.)

But again, some have found an improvement in their condition once embarking on an eccentric exercise protocol. What gives?


The hallmark study’s protocol

“The patients were instructed to do their eccentric exercises 2 times daily, 7 days/week, for 12 weeks. During the 12-week training regimen, running activity was allowed if it could be performed with only mild discomfort and no pain.”

“In the beginning, the loading consisted of the body weight, and the patients were standing with all their body weight on the injured leg.”

“They were only loading the calf muscle eccentrically, no following concentric loading was done. Instead, the noninjured leg was used to get back to the start position. The patients were told to go ahead with the exercise even if they experienced pain. However, they were told to stop the exercise if the pain became disabling. When they could perform the eccentric loading exercise without experiencing any minor pain or discomfort, they were instructed to increase the load by adding weight.”

That’s quite different than what we’ve heard so far. Instead it’s “some discomfort is ok, but if it becomes significantly painful, stop. With running, pain is not allowed at all. Oh, and if you want to load the area more, you can’t do so until the previous load is completely pain free!

What load actually means in these protocols

What’s a typical runner get mileage wise per week? 20 to 30 miles? Ok, and what’s a typical stride on these longer distance runs? 1.9 meters (6 feet)?

20 miles per week would be 32000 meters.

  • 32000 meters / 1.9 meters per stride = 16,842 strides

That’s 8,421 steps per foot.

The eccentric exercise protocol consisted of 3 sets of 15 reps, performed twice per day, seven days per week:

  • 3 * 15 * 2 * 7 = 630 reps per week

Nothing like loading your achilles THIRTEEN times less!

In fact, when we run we load the achilles eccentrically (when it lengthens) and concentrically (when it shortens). In the eccentric protocol, they only loaded it eccentrically. If we consider every stride is then loading the achilles twice, or every eccentric rep is only half a rep, then if you stopped running in this protocol we’re talking reducing the reps by a factor of 26!

But we haven’t truly accounted for stress on the achilles yet. Only reps.

For seated heel raises:

Some argue one needs to get to where the resistance ends up being the entire body weight, because that’s what we run with. However, that thinking is flawed because the calves are not the only thing helping propulsion and absorption when running. We have all those other leg muscles too.

Not only that, when we run we actually encounter forces 2-3 times our weight. 

-> This is our physics weight, not our bodyweight. That is, F =mass * acceleration due to gravity, not what you weigh in pounds.

180 pound person’s physics weight is (need to convert to kilograms):

Force = 82 kg * 9.8 m/s^2 = 802 newtons

Standing, walking, running, and jumping on a Force Plate

It’s not so simple as to how much we need to load our achilles during a heel raise to make it comparable to running. Luckily, we have,

The mechanics and energetics of human walking and running: a joint level perspective.


(a) and (d) are the ankle. (a) is walking; (d) is running. What pace does our average runner go at? Maybe seven minutes per mile? That would be 8.5 miles per hour, or 3.8 meters per second. The solid black line is 3.25 meters per second, and that’s as fast as these people went.


The graph is relative to power, in watts per kilogram. This is crucial. Speed is an extremely relevant detail when discussing injuries. Want somebody with achilles pain running to feel better? Slow them down.

What we’re interested in is the maximum amount of stress the ankle / calves / achilles have to deal with. After all, that’s typically what we’re worried about when coming back from an injury. Not the least stressful aspects of the activity, but the most.

10 watts per kilogram looks fair here:


The average mass of the subjects was 77 kilograms.

  • 10 watts per kilogram * 77 kilograms = 770 watts

When running at 3.25 meters per second (7.25 miles per hour), for a 77 kilogram person, the ankle on average will generate a peak of 770 watts, in one stride. Thus the ankle / achilles / calves have to be able to deal with this.

-> Initially, this number does seem a bit high. However, this paper is well cited, is the only one I find with data like this, and we’ll see even if it’s off by a factor of two, that won’t change the argument.

How much do we need to load our heel raise exercise then?

  • Power = (Force * Distance) / time
    • Power = (mass * acceleration * distance) / time

Power is how much force we generate, over what distance, divided by how quickly it gets done. Generate more force; over a greater distance; in less time => more power.

In a heel raise exercise,

  • Power = 770 watts
  • Distance = let’s say 6 inches = 0.154 meters
    • This is how far your heel goes up
    • We’re assuming starting from flat ground, rather than elevated. If you increase the distance in this case then you’ll need to increase the time.
  • Acceleration = 9.8 meters per second^2
  • Time = let’s say 1 second
    • How long it takes to go up in a heel raise exercise rep

We can then figure out what the mass in a heel raise exercise needs to be,

  • Power = (mass * acceleration * distance) / time
  • 770 watts = (mass * 9.8 m/s^2 * 0.154 meters) / 1 second
    • 770 watts = mass * 1.5
      • 770 / 1.5 = 513 kilograms

Or 1,129 pounds. (Can include your bodyweight if standing.) This is on one foot! How on Earth are we going to achieve this with heel raises? We gotta be this guy (who still isn’t on one foot)?

Even if you want to say “I think an elevated heel raise could be done in 1 second. That would double the distance, halving the mass we need.” That still puts us at more than 500 pounds, and remember, this is PER REP. Meaning to make it comparable to running we not only need this mass, we need to do it for THOUSANDS of reps.

-> Even then, you’re still not simulating the speed running entails.

This is a big reason so many NFL players pull a hamstring once holding out or coming into camp late. (See: Joey Bosa for the Chargers this year.) Because speed is a crucial, underappreciated element. Football players are rarely lacking strength when they come into camp, but they unlikely have gone at training camp / game speed much, if at all, all off-season. It’s one of those things that can’t be simulated. Meaning they’ve unlikely generated the power during the off-season that they will once going at training camp speed. And if they have, then it’s unlikely they’ve done it for the amount of reps they will in camp.

When I played high school and college ball, no matter what anybody did in the off-season, the first week of camp was hell as everybody was sore as fuck. No matter how much we ran, many of us got blisters, pulled hamstrings / hip flexors. You’d been running all off-season, but you hadn’t been running like that.

It can be hard to appreciate speed, but another way to think of this is torn achilles are virtually always from something running oriented, not from heavy calf raises.

In these eccentric studies, a person may bump the load to wearing a backpack with something inside of it -maybe 50 lbs max?- or get up to a weight machine -where many only get up to ~300 lbs.

The real reason eccentric protocols work is because they comprise a DRASTIC reduction of volume on the achilles tendon, with a possible never painful running program as an adjunct. Shocking how running with no pain can help one…run with no pain.

-> Work is a relative term. Here is a study finding 60% of eccentric exercisees still had pain five years later and 52% sought alternative treatment. More than likely those who do well with this stuff 1) reduce their running and slowly work their way back up 2) make the strength work barely, if at all, painful, where those who have issues 1) don’t heed the avoid pain advice / 2) take running off for 12 weeks but then jump back into it cannonball style.

Even if you think the physics above is all bullshit, you can’t argue with the difference in reps between running e.g. five miles and doing calf raises for a few sets of 15 reps.

-> Plus, the strength training crowd still hasn’t accounted for,

“Indeed, six out of seven studies found a larger Achilles tendon CSA in runners compared with non-runners, although sometimes this difference was only present in specific parts of the Achilles tendon or only if the Achilles tendon CSA was normalised to body weight. Other studies examining the effect of a (non-running) training programme on tendon size and stiffness also found that tendon hypertrophy following training is region specific.”

What’s the likelihood running and heel raises give the same region specific hypertophy? Muscle adapts differently whether it’s loaded quickly or slowly. Does tendon have speed specific adaptations as well? (Yes.) (Doesn’t mean strength training is futile.)

The real reason a protocol like this can be helpful is an only once pain is gone increase in activity is embarked on. Which is the real hallmark of coming back from any injury! Rather than rest for a while, then start running like you were before. That often fails. After all, running like you were before is why your achilles started hurting.

Even if going into pain were beneficial, one has to weigh the undoubted decrease in adherence / increase in misery of regularly having someone do that to themselves. (The eccentric protocol called for twice a day, every day, 12 weeks. That could be a lot of pain!) Where -ignoring things like lactic acid- the main reason I don’t have my clients exercise into pain is because being in pain sucks.

Enter your email address to follow this blog and receive notifications of new posts by email.