Physiology of recovery..?

5s recovery. 15…30. 5mins recovery, 10mins, 20mins. 4hrs-12hrs recovery.

What’s happening to our bodies in these periods between hard work and are they all really recovery?

In swimming and some mean TR workouts, recovery intervals are measured in seconds - surely no recovery is happening here at all. Even ATP reserves take a minute or two to rebuild.

Long recovery intervals can often be the opposite, what are we gaining from ages spinning easy power before the next bout?

And finally the time between workouts or even sleep - at what point is our body in a state of repairing those damaged fibres and so on?


1 Like

a lot of questions, and for the recovery interval question alone the answers are dependent on the interval format.

Way back when I was road racing, I was solo off the front in a criterium. This was definitely not a regular occurrence, as I’m pretty small and more oriented towards anaerobic power than time trialing. It was a bit of a unique setup.

The course had a slight climb to start, then a right turn with a steeper climb, then right turn and a descent, then a few more turns before the finishing straight. I was surprisingly holding my own. With a few laps to go, my legs started to falter up the short, steeper climb.

Instinctively, I did not panic, and just rested up on that descent. It was just a few seconds, probably less than 10. But when I started back up, my legs were OK. Burning, but OK.

Basically, there are circumstances where even a 5-10s recovery between intervals can help.

So, I left you guys on a cliffhanger. The chasers totally failed to coordinate. This is like a men’s cat 4 road race. I don’t think there were any organized teams who had more than one rider in the chasing group. A handful (3?) of them caught me just before the finishing straight, only for me to outsprint them.


This podcast covers the subject pretty well.

Basically, it depends on the goal of the interval.

Yes… recovery is happening even in the “seconds” of recovery. You can take an exercise physiology course to really get into it… but oversimplifying things, in even in short recovery intervals Oxygen is being transported to muscles, breathing rate and heart rate decreases. There a lot of chemistry happening at the cellular level. Long recovery intervals, for example 1:1 in VO2 max or long threshold, allows you to no longer be in oxygen debt, lactate management, etc. Needed recovery varies per individual. Some recover faster than others.

N+1 but when training post collegiate I was shocked how fast intervals were run and how little recovery we were given. A Kenyan group was in town and on the track at the same time. I saw them do 6x800 all sub 2 min with less than 30 second 50 meter jog. Insane.


As I frequently tell students, fatigue (as in, failure to maintain the expected or required force or power output, OR any reduction in the force-generating capacity of muscle, regardless of present demands) is always multifactorial.

It therefore logically follows that reversal of fatigue (i.e., recovery) is also always multifactorial.

[Note that this is why modeling W’bal using a mono-exponential function doesn’t work very well, and why I was playing around with multiexponential functions to model dynamic changes in FRC 5-10 y ago now.]

Beyond that, I would have to write a treatise to cover all of the mechanisms/scenarios. One important thing to emphasize, though, is that since lactate doesn’t cause fatigue, the rate at which you clear it is irrelevant.


If no recovery is happening. Try Spanish Needle and skip those 15s recovery’s and do 5mins at 150%.

What exactly is happening on cell level, no idea. But recovery is happening.

1 Like

IIRC, “Spanish Needle” is 15 s on/off microintervals?

If so, perhaps what’s more important to completing the workout is what isn’t happening during the work periods.

But, your point is correct, “recovery” of muscle (reestablishment of cellular homeostasis) starts essentially instantaneously once the signal to contract ceases, and some aspects (e.g., restoration of resting membrane potential, resynthesis of high energy phosphate stores) proceed on millisecond-to-second time scales.

OTOH, others (e.g., normalization of muscle pH, replenishment of glycogen) take minutes to hours to complete.

Like I said, recovery is complicated. One thing is clear, however: recoverability (which I consider a distinct, albeit not uniquely determined, contractile property of muscle - alongside strength, speed, power, and fatigability, which are all different things with their own non-overlapping determinants) is trainable (although not necessarily independently).

(Speaking of fatigue and recovery, here’s our latest:

The influence of acute dietary nitrate supplementation on skeletal muscle fatigue and recovery in older women - PubMed)


So recovery is trainable too?

By that, I mean that the more you cause your body to recover for short periods of time from intense efforts, it gets better at it?

Does doing a lot of, say, 40/20 intervals regularly & frequently train the body so that it can do more recovery (whatever that is) in 20 seconds than it would have been able to do without having been caused to do it repeatedly?

Recovery is absolutely trainable, but that’s not entirely the point of 40/20s. You can achieve the same thing by increasing endurance volume. I would argue the latter is a better way to do it long term than the former. You try to do 40/20s for more than a few weeks and your gains are going to peter out. It’s about improved aerobic fitness, and that can come from a host of things… but endurance volume is king, IMO.

1 Like

Maybe, like other aspects of training, there is a balance to be struck?

Is the recovery trained by, say, 40/20s different in some way from the recovery trained by endurance volume?

Is the recovery that the body starts at the end of higher intensity interval different in some way from the recover that its starts under endurance loading?

Absolutely, recoverability is trainable…I can’t really think of any mechanism or method of training it specifically, however. I therefore find myself somewhat regretting bringing it up. :slight_smile:


The smart coaches I listen to when learning say “recovery is an aerobic process”, so essentially any training that improves aerobic fitness will necessarily improve your ability to recover quickly between efforts. I see this in my athletes, where one marker of improved fitness/performance is more rapid HR response between efforts. End a 20 minute threshold interval and HR drops rapidly down to a recovery number? Good!

But I don’t train 40/20s to improve recovery. I train 40/20s to train the ability to go hard again and again and again. You’re training multiple things there - mental toughness, anaerobic capacity, etc., and yeah, probably your body’s ability to recover too. But again, the gains you’re going to see from 40/20s are going to peter out pretty rapidly (2-6weeks) whereas riding endurance volume you can do until the day you die, and you’re “training recovery” by doing so.

But you’ve got Dr. Coggan in here essentially telling you this is too complex to explain physiologically on a forum, so I’m sure as hell not going to try, lol!

1 Like

Could one argue that recoverability and repeatability are two sides of the same coin? IE, if person A can repeat the same interval/rest pattern more frequently than person B, wouldn’t it imply that person A is recovering better between intervals (assuming mental toughness is equal)?

Does anyone know of research around optimal recovery modalities? In TR, we more or less get a recovery week with easy z2 rides and potentially less volume. Is that optimal?

Would it be better (or equivalent) to take 3 days off the bike completely, for example?

Back in the day on the farm, we would do a (running) workout called “The Michigan.” We would start off with a 2k @ marathon pace on grass or the wood chip trail next to the track, then immediately go into 1600 at 5k/10k pace on the track, then back to tempo. The workout would cut down a lap each cycle continuous with no breaks (the “recovery” was the tempo at marathon pace). Other workouts such as 24 x 400 (sets of 6) would have a 200 jog recovery with a rolling start that usually was less than 45 seconds. Mile repeats had 3 minute recovery…
[Note: This was for the distance group 5k to marathon; 800/1500 group obviously doing something different.]

It took a while to adjust to not just the intensity but “training” your body to recover while still moving at a reasonable pace. Like you said the key was to run relaxed during the tempo and to get your heart rate down!

1 Like

They are certainly related, for sure.

However, I can also envision how they might differ - for example, one individual might have higher glycogen stores than another, and therefore be able to do more repeats of a particular effort before depletion in a subset of motor units forces them to stop.

One issue to consider is that even terms like “strength” or “fatigue” can mean different things to different people, even different exercise scientists. As I said before, I consider “recoverability” to be a separate property of muscle, but in fact the term is rarely used in the scientific literature - “repeatability” even less often (if ever). That makes conversations like this one a bit challenging, or at least more challenging than they might be if such terms were clearly defined.


I don’t like TRs recovery week structure at all. I think it’s too much work and not enough recovery.

My general guidelines (specifics obviously vary by athlete, volume, training history, etc):

-Reduce ride time by 50% or more (for high volume people)
-At least two full rest days, often more.

For higher volume athletes, I like 3 full days off, then a recovery spin and build back into riding starting with 1 hr up to a long ride of about half of normal long ride duration on the weekend.

Then I usually end a recovery period with openers and a short duration PDC test.


I.e., train for performance and let your physiology sort itself out.


Yep, I caveat most of my guidance here with “I’m not a physiologist.” :innocent:

1 Like