The Ledger Never Balanced
Twenty five years of clinical reasoning, resting on a transaction nobody has ever seen.
Your player tore the distal end, had it repaired, got through a clean return to play, and then strained the same hamstring further north. You sat in the review meeting and said the word compensation. Everybody nodded. Everybody in that room has said it too.
So have I, for twenty five years. I have sat across a table from physiotherapists I would trust with my own family and we have said it to each other, and we were not being lazy. It is the most sensible reading available. If one limb cannot do the job, something takes it up. If one part of a muscle is scarred and shortened, another part carries more. Of course it does.
But look at what that word actually claims. It is an accounting statement. Load left one place and arrived in another. That is a debit and a credit, and in twenty five years I have never once seen both sides of the entry written down. We have the credit, an injury that turned up somewhere new, and we have filled in the debit from memory. Then we balanced the books and signed off the rehab.
I am not telling you the reasoning is wrong. I would make the same call tomorrow morning, and so would you, and we would both be right to.
What I am telling you is what happens to a good hypothesis when it gets repeated for twenty five years with nothing measured underneath it. It quietly stops being a hypothesis. Somebody takes the question mark off, and two generations of practitioners later nobody remembers it was ever there. That is the enemy in this piece. Not the clinicians. Repetition.
What The Ledger Actually Shows
It is worth knowing precisely where the evidence sits, because it is a great deal better than nothing and a long way short of proof.
Schuermans and colleagues used muscle functional MRI and found that footballers with a history of hamstring injury recruit biceps femoris and semitendinosus in a different pattern to uninjured players.1 Avrillon and colleagues got closest to an actual measurement. In seventeen elite athletes with a previous biceps femoris injury, the injured muscle contributed less to total knee flexion torque than the same muscle on the uninjured side, and semimembranosus picked up the difference.2 That is a debit and a credit, both written down. It was measured in knee flexion on a bench rather than in a running athlete, and the spread between individuals was wider than the average shift, but it is the nearest thing we have to seeing the transaction. Silder and colleagues showed the structural version, atrophy in the injured biceps femoris long head alongside enlargement of the muscle sitting next to it, still there long after return to play.3 Fyfe and colleagues set out the mechanism most people have in mind when they say the word, neuromuscular inhibition after injury leaving the athlete with deficits that persist into return to sport.4
Read that list again. Three of the four are observations that a difference exists. Avrillon is the exception, and it is a bench measurement of a bench task. Not one of them shows load moving from one structure to another in a running athlete, which is the only condition any of us are actually being paid to care about. The credit is documented. The debit is mostly still guessed at.
And the systematic review will not let you round it up. Presland and colleagues pulled the muscle activity work together and found the evidence moderate at best under controlled strength testing, and conflicting once the athlete runs.5
Then look at what running meant in that review. Surface EMG amplitude, mostly biceps femoris through late swing, during treadmill and overground sprinting. Not contact time. Not vertical force. Not step length, and not where the foot lands relative to the centre of mass.
A more recent meta-analysis went far wider, sixty three studies and 1,796 participants across nine different lower limb injuries, and for hamstring injury during running it reports biceps femoris activation no different at any phase of gait.6 Then look at what that rests on. Two studies. The rest of the hamstring running work in it, lateral hamstring, medial hamstring, gluteus maximus, gluteus medius, rectus femoris, is single studies of eighteen to twenty people each. The authors say it themselves in their limitations: no study in the entire review exceeded sixty eight participants. And the review excluded strength and joint torque outcomes by design, so mechanics were never on the table in the first place.
Running was examined with an electrode, on twenty people at a time, and nobody looked at the mechanics.
Then there is the one that should make you uncomfortable, because it points the other way entirely. Edouard, Morin, Brughelli and colleagues showed that as sprinters fatigue, the gluteals take up more of the horizontal force job, and they read it as the hamstring being protected.7 If compensation can protect an athlete, then the word on its own tells you nothing about whether you are looking at a problem or at the thing keeping him on the park.
Inter-limb is no tidier, and it is where the hole shows up most plainly of all. Fox and colleagues meta-analysed inter-limb asymmetry against athletic performance and found a statistically significant but weak association with sprint times, an r of 0.2 across nine studies and 211 athletes.8 Now read what was actually measured there. The asymmetry came from unilateral jump testing. The outcome was a sprint time. Nobody measured asymmetry in the running itself. Even the meta-analysis is standing in the gym looking out at the track.
What Actually Changes On Monday
Not much, and that is the honest answer.
You do not stop reasoning about compensation. Drop it and you have nothing to replace it with, and the clinical reasoning gets worse, not better. You keep the word. What you change is the status of the sentence.
Say what you measured. Say what you inferred. Keep them in separate columns and never let the second one drift across into the first, because that drift is the whole problem and it happens quietly, over years, in meeting rooms full of people who all agree with each other.
Then go and get a second signal from somewhere the first one cannot see. Not because it closes the ledger. It does not. Because two independent observations pointing at the same limb is a different quality of argument to one observation and a story, and it is the best any of us can honestly do right now.
Somebody will measure this properly one day, at the muscle level, in a running athlete, and see the load actually shift. That will not be me. When it happens, half of what we have all been saying for twenty five years will turn out to be right, and we still will not know which half we got lucky on.
Until then, keep the books straight.
Say compensation if you like. Then say what you measured.
Put the inference in its own column and label it as one.
When somebody tells you the other limb took it up, ask them how they know.
Be honest when the answer is that it stands to reason.
And we hypothesise.
1. Schuermans J, Van Tiggelen D, Danneels L, Witvrouw E. Biceps femoris and semitendinosus, teammates or competitors? New insights into hamstring injury mechanisms in male football players: a muscle functional MRI study. British Journal of Sports Medicine. 2014;48:1599-1606.
2. Avrillon S, Hug F, Guilhem G. Bilateral differences in hamstring coordination in previously injured elite athletes. Journal of Applied Physiology. 2020;128(3):688-697.
3. Silder A, Heiderscheit BC, Thelen DG, Enright T, Tuite MJ. MR observations of long-term musculotendon remodeling following a hamstring strain injury. Skeletal Radiology. 2008;37(12):1101-1109.
4. Fyfe JJ, Opar DA, Williams MD, Shield AJ. The role of neuromuscular inhibition in hamstring strain injury recurrence. Journal of Electromyography and Kinesiology. 2013;23(3):523-530.
5. Presland JD, Timmins RG, Maniar N, Tofari PJ, Kidgell DJ, Shield AJ, Dickson J, Opar DA. Muscle activity and activation in previously strain-injured lower limbs: a systematic review. Sports Medicine. 2021;51(11):2311-2327.
6. Andringa-Bate J, Almusallam M, Coventry M, Murphy MC, Rio E, Pizzari T. Neuromotor function following lower-limb muscle or tendon injury: systematic review and meta-analysis. Sports Medicine and Health Science. 2026;8(4):390-406.
7. Edouard P, Mendiguchia J, Lahti J, Arnal PJ, Gimenez P, Jiménez-Reyes P, Brughelli M, Samozino P, Morin JB. Sprint acceleration mechanics in fatigue conditions: compensatory role of gluteal muscles in horizontal force production and potential protection of hamstring muscles. Frontiers in Physiology. 2018;9:1706.
8. Fox KT, Pearson LT, Hicks KM. The effect of lower inter-limb asymmetries on athletic performance: a systematic review and meta-analysis. PLOS ONE. 2023;18(6):e0286942.
Take Three Pieces Of It Free
One of them is an audit that makes you separate what you measured from what you assumed, on your own systems. Same two columns as above, pointed at your work instead of your athlete. No cost, no card.
Send me the three