Ankle dorsiflexion is the movement that brings the top of the foot towards the shin: the one you make, without thinking, every time you go into a squat, go down stairs or absorb a landing. It is also one of the movements I most often find limited in the clinic, after badly recovered sprains, periods of immobilisation, or simply years of heeled shoes and desk life.
Here you will find how to measure it, what the studies really say about how much can be gained, and exercises with doses and progressions. I will make the honest point straight away: the documented gains exist but amount to a few degrees, and the link with pain or injury is much more nuanced than it is usually presented.
The wall test: measure before you train
The wall test (knee-to-wall, called the weight-bearing lunge test in the literature) requires only a wall and a tape measure:
- Stand barefoot in front of a wall, with the toes of the foot being tested about 10 cm from the skirting board (this is only a starting point, not a threshold).
- Keeping your heel firmly on the ground, bend your knee trying to touch the wall.
- If you touch it easily, move your foot 1-2 cm further away and try again. If you cannot touch it, move closer.
- Your score is the maximum toe-to-wall distance at which your knee can still touch the wall with your heel on the ground.
It is a test that holds up: a systematic review of 12 studies found good reliability both between different examiners (ICC 0.80-0.99) and for the same examiner repeating the measurement (ICC 0.65-0.99). And a radiographic study of 20 participants verified that, at the end of the range, 91.8% of the movement really happens at the tibiotalar joint: so you are measuring the ankle, not the foot giving way.
The number that matters is the measurement error
This is the part that is usually left out. The same review reports a minimal detectable change of about 1.9 cm (4.7°) for the same examiner and 1.6 cm (4.6°) between different examiners. In plain terms: if on retest you gain 1 cm, you have not gained anything demonstrable, because it sits within the noise of the instrument. Only a change of more than about 2 cm can be read as a real change.
The same applies to comparing right and left. An international study of 899 healthy adults (Canada, Spain, Iran) found no significant difference between the two limbs (p = 0.172, d = 0.05): on average, in healthy people, the two ankles are equivalent, so an asymmetry is worth watching, but only if it exceeds the measurement error. Below that threshold you are interpreting your tape measure, not your ankle. And the idea that asymmetry causes compensations remains a clinical hypothesis, not a proven fact.
There is no validated cut-off
The "10 cm" that circulates everywhere as the line between normal and limited is a convenient convention, not a validated value. The same study of 899 adults shows why it cannot be one: the result falls progressively with age (the sharpest drop starts from the 60-69 age band), and men have slightly higher values than women. The authors do not publish a single threshold, but separate percentile bands for each combination of age and sex. It makes no sense to compare a fifty-year-old with the good reference for a twenty-year-old.
The useful reference, therefore, is you: record the starting value of both ankles and repeat the test every 2-3 weeks, knowing that below a 2 cm difference nothing measurable has happened.
What can really be gained
The meta-analyses agree on the direction and scale down the size of the effect.
On static calf stretching, a meta-analysis in the British Journal of Sports Medicine of 5 randomised trials found average increases of 2.07° (up to 15 minutes of total stretching), 3.03° (15-30 minutes) and 2.49° (over 30 minutes), and the authors explicitly conclude that the increase is small, statistically significant, but of uncertain clinical relevance. A more recent meta-analysis of 20 studies in healthy subjects reports somewhat more generous figures for work sustained over time: +5.17° with static stretching (95% CI: 4.39-5.95) and +4.32° with proprioceptive neuromuscular facilitation (95% CI: 1.59-7.04), while ballistic stretching produced no significant effect (+3.77°, 95% CI: from -0.03 to 7.56).
Compare these numbers with the measurement error discussed earlier: average gains between 2° and 5° are in the same order of magnitude as the 4.7° that the test cannot distinguish from noise. That does not mean the work is useless. It means it has to be done for weeks, and that miracles in three sessions do not exist.
A third systematic review of 23 studies confirms that the evidence supports static stretching alone or combined with other therapies, and that for most other conservative interventions there are no studies of sufficient quality. Finally, after a sprain, a review of 9 studies found the largest effect for static stretching included in a home programme, measured two weeks after the acute trauma (Cohen's d 1.06; 95% CI: 0.12-2.42, so with a very wide uncertainty), while mobilisations with movement in people with recurrent sprains showed small effects (d between 0.14 and 0.39). The authors' conclusion is the most important one: first you understand what is limiting the ankle, then you choose the technique.
Why it can matter: squats and landings
In 30 healthy subjects, one study simulated stiffness of the plantar flexors with a 12° wedge under the forefoot during a two-legged squat. In the wedge condition there was less knee flexion, more valgus and more medial displacement of the knee (a very large effect on the latter), less activation of the vastus lateralis and vastus medialis obliquus, and more activation of the soleus. The authors note that these alterations resemble those seen in people with patellofemoral pain: they resemble them, they do not cause them. It was a simulated, acute constraint in people with no injury history. There is also a detail that calls for caution in generalising: with the wedge, the measured dorsiflexion angle actually increased, because the wedge tilts the foot. The model reproduces the effect of a short calf on the starting position, not an ankle that locks.
On landings, a systematic review of 6 studies concludes that reduced dorsiflexion alters lower limb mechanics, but the studies do not agree on which variables change; the authors speak of a possible predisposition to injury as a hypothesis, not as an established finding.
And the direct link with injuries? A prospective study of 241 male sports science students followed for 1 to 3 years saw 44 subjects (18%) suffer an inversion sprain: among the factors associated with risk is reduced dorsiflexion range, but alongside running speed, cardiorespiratory endurance, balance, strength of the dorsiflexors, coordination and muscle reaction times. It is one piece of a mosaic of seven, measured on a young male population: using it as "little dorsiflexion means you will sprain your ankle" is a stretch.
The exercises
1. Wall mobilisation with oscillations
It replicates the test: in the same position, at a distance where the knee almost touches the wall, move the knee forward to the limit, hold for 2-3 seconds and return. Slow movement, heel always on the ground.
- Correct execution: the knee travels over the 2nd-3rd toe, not inwards; you should feel a deep tension in front of the ankle or behind the calf, not a sharp pinch.
- Typical error: lifting the heel or letting the arch of the foot collapse inwards. If that happens, you are too far from the wall: reduce the distance.
2. Lunge with knee overload
Front foot on a low step, hands on a weight or on the knee itself: push the knee forward past the toes, adding a gentle pressure.
- Typical error: turning it into passive stretching held at random. Slow, controlled repetitions, moving in and out of the range, are better.
3. Calf stretches: gastrocnemius and soleus
This is the exercise with the most solid evidence of the four, and it is what makes both versions useful: stretching with the knee straight (gastrocnemius) and with the knee bent (soleus). The second, often forgotten, is the one that matters when the knee is flexed, that is, in the squat. For the difference between the two, see the dedicated piece on soleus stretching.
4. Assisted deep squat
Holding a handle or a stable piece of furniture, lower yourself into a full squat with your heels on the ground and stay there for 20-30 seconds, breathing, gently shifting your weight from one ankle to the other. This is clinical practice more than literature: the idea is to use in movement the range you are training, but I have no meta-analysis to cite in support.
Sets and repetitions
This scheme reflects how I set the work in the clinic: the meta-analyses confirm the effectiveness of static stretching sustained over time, but they do not validate one specific combination of sets and frequency.
| Exercise | Sets | Reps/Duration | Frequency |
|---|---|---|---|
| Wall mobilisation | 3 per side | 12-15 oscillations | 5-6 days/week |
| Lunge with knee overload | 3 per side | 8-10 slow | 3-4 days/week |
| Calves (gastrocnemius + soleus) | 2+2 per side | 30-40 seconds | 5-6 days/week |
| Assisted deep squat | 3 | 20-30 seconds | 3-4 days/week |
The total is about 10-12 minutes of work. Expect weeks, not days: the studies that document gains work on sustained protocols, and your criterion of success is not the feeling but the test, that is, more than 2 cm compared with the starting point.
Progressions
When the wall test improves measurably and the basic exercises become easy: add weight (a kettlebell or dumbbell on the knee in the lunge), move to the deep squat without support, and finally bring the movement into dynamic actions: controlled slow descents from a step, walking lunges. If your goal is the barbell squat, the final test is squat technique with heels firmly planted on the ground.
When to stop and get checked
There are situations in which pushing for mobility is the wrong move:
- Acute, "stabbing" anterior pain in the crease of the ankle when you bring the knee forward. This is the picture described for anterior ankle impingement: pain that appears during activity with extreme dorsiflexion under load, sustained by bony or soft tissue causes, and which can develop as secondary damage from repeated trauma. Forcing it does not loosen it: it irritates it.
- Stiffness that appears after a fracture or immobilisation: a bony block or a joint outcome does not stretch out, and it needs to be assessed, with imaging if necessary, before you load it.
- Recent sprain: in the first weeks the priorities are progressive loading and stability; aggressive pushing into dorsiflexion comes later.
- Swelling, heat or night pain in the ankle: first understand the cause, then train.
- No measurable progress after 6-8 weeks of consistent work. This is the point where the review on sprains is right: if you do not know what the limiting factor is, you are guessing with the same exercise.
In all these cases the problem is not the exercise, it is that there is no diagnosis of why that ankle does not move. A physiotherapy assessment answers exactly that question: start with the questionnaire and we work out together whether your limitation is muscular, joint-related or simply a question of method.
One final note: this movement is one piece, not the whole ankle. If you want to work on the joint in all directions, plantarflexion, stability, proprioception, you will find the full picture in the article on ankle mobility.
Sources
- Powden CJ, Hoch JM, Hoch MC. Reliability and minimal detectable change of the weight-bearing lunge test: a systematic review. Manual Therapy, 2015. https://pubmed.ncbi.nlm.nih.gov/25704110/
- Smith MD, Lee D, Russell T, et al. How much does the talocrural joint contribute to ankle dorsiflexion range of motion during the weight-bearing lunge test? A cross-sectional radiographic validity study. Journal of Orthopaedic & Sports Physical Therapy, 2019. https://pubmed.ncbi.nlm.nih.gov/31337266/
- McBride S, Isabelle PL, Gómez-Carrión Á, et al. International normative values for the weight-bearing lunge test across age and sex in 899 healthy adults. Musculoskeletal Science and Practice, 2026. https://pubmed.ncbi.nlm.nih.gov/41723909/
- Radford JA, Burns J, Buchbinder R, et al. Does stretching increase ankle dorsiflexion range of motion? A systematic review. British Journal of Sports Medicine, 2006. https://pubmed.ncbi.nlm.nih.gov/16926259/
- Medeiros DM, Martini TF. Chronic effect of different types of stretching on ankle dorsiflexion range of motion: systematic review and meta-analysis. The Foot, 2018. https://pubmed.ncbi.nlm.nih.gov/29223884/
- Young R, Nix S, Wholohan A, et al. Interventions for increasing ankle joint dorsiflexion: a systematic review and meta-analysis. Journal of Foot and Ankle Research, 2013. https://pubmed.ncbi.nlm.nih.gov/24225348/
- Terada M, Pietrosimone BG, Gribble PA. Therapeutic interventions for increasing ankle dorsiflexion after ankle sprain: a systematic review. Journal of Athletic Training, 2013. https://pubmed.ncbi.nlm.nih.gov/23914912/
- Macrum E, Bell DR, Boling M, et al. Effect of limiting ankle-dorsiflexion range of motion on lower extremity kinematics and muscle-activation patterns during a squat. Journal of Sport Rehabilitation, 2012. https://pubmed.ncbi.nlm.nih.gov/22100617/
- Mason-Mackay AR, Whatman C, Reid D. The effect of reduced ankle dorsiflexion on lower extremity mechanics during landing: a systematic review. Journal of Science and Medicine in Sport, 2017. https://pubmed.ncbi.nlm.nih.gov/26117159/
- Willems TM, Witvrouw E, Delbaere K, et al. Intrinsic risk factors for inversion ankle sprains in male subjects: a prospective study. The American Journal of Sports Medicine, 2005. https://pubmed.ncbi.nlm.nih.gov/15716258/
- Hess GW. Ankle impingement syndromes: a review of etiology and related implications. Foot & Ankle Specialist, 2011. https://pubmed.ncbi.nlm.nih.gov/21926368/
