Ankle Mobility Exercises: The Restriction That Travels Upward

Ankle mobility exercises are the most-filmed item in the entire movement category, and the reason is simple: the ankle test is a two-second reel. Knee to the wall, heel down, measure. Someone fails it on camera, the comment section fills with people trying it against their own kitchen wall, and the clip travels. The test deserves the attention. What usually follows it does not.

What follows is almost always the same calf stretch, held for thirty seconds, repeated for a week, abandoned. The ankle is treated as a hinge that needs loosening. It behaves more like a gate: it decides how much of the leg above it can be used, and it responds to being tested, then loaded — not to being pulled on. This guide sets out what the research actually shows about dorsiflexion, what the number does and does not predict, and how to train range that stays.

Ankle mobility exercises work best when they follow a test. Measure weight-bearing dorsiflexion with a knee-to-wall lunge, note whether the block feels like a pinch at the front of the joint or a stretch at the back, then train the range you gain under load. Untested stretching changes little.

Ankle mobility exercises start with a test, not a stretch

The test that made this topic travel is also the correct place to begin. Stand facing a wall. Point the toes straight at it. Drive the knee forward over the middle of the foot until it touches, keeping the heel flat on the floor, and find the furthest foot position at which contact still happens. Record the toe-to-wall distance in centimeters, or the angle of the shin against vertical. This is the weight-bearing lunge test, and it is the version of the ankle worth knowing, because it loads the joint the way standing, walking and landing do.

The distinction is not academic. In a study of forty physically active adults, participants were grouped twice — once by passive dorsiflexion measured lying down, once by the weight-bearing lunge. Only the weight-bearing grouping tracked with anything meaningful: those with more dorsiflexion under load produced greater knee-flexion and ankle displacement and greater peak knee flexion during overhead and single-leg squats. The passive measurement separated nobody [2].

Reference points exist. A recent clinical framework for restoring dorsiflexion in athletes treats values above 9 to 10 cm, or 40 to 42 degrees of shin inclination, as adequate for the demands of walking, stair descent, sprinting and cutting, and flags side-to-side differences beyond 1.5 cm or 4.7 degrees as clinically relevant asymmetry [6]. Two ankles on the same body are the cleanest comparison available to you.

How much dorsiflexion is enough?

A 2025 simulation study modeled the joint-angle combinations that make a given squat depth geometrically possible. Working from anthropometric data, it put the minimum dorsiflexion behind a quarter-depth squat at roughly 8 degrees, half depth at about 26 degrees, and a deep squat at around 35 degrees [4]. Those figures explain the felt experience precisely: the deeper the position, the narrower the set of ankle angles that permit it.

• Shallow work asks little. Around 8 degrees covers a quarter-depth squat, which is why a restricted ankle can hide for years in partial-range training.
• Depth compounds the demand. Half depth roughly triples the requirement, and a deep squat asks for more than four times the shallow figure.
• The options narrow with depth. The same model found the number of viable joint-angle combinations dropped sharply at depth — fewer ways to solve the position, less room to compensate.

The same study is also the strongest argument against reading too much into your own number. Because hips, knees and ankles can trade contribution, many different joint-angle combinations produce identical depth, and the variability around the estimates was large enough that the authors concluded ankle mobility alone is an unreliable predictor of how deep any individual will go [4]. Ranges are population geometry. They are not a verdict on you.

Why the restriction travels upward

The ankle is the first joint to negotiate with the ground. When it runs out of range, the demand does not disappear — it is redistributed upward. The knee drifts inward, the pelvis tips, the torso pitches forward, the heel leaves the floor. Nothing about this is dysfunction. It is a body solving a task with the range it has.

The pattern is documented. A meta-analysis pooling 17 studies found reduced ankle dorsiflexion associated with dynamic knee valgus, with a standardized mean difference of −0.65, and the association was markedly stronger when dorsiflexion had been measured in weight-bearing rather than lying down [1]. That is the same signal the squat study produced, at population scale.

Landing tells a similar story, with more caution attached. A systematic review of six studies concluded that restricted dorsiflexion alters lower-extremity landing mechanics, while noting that the specific variables affected differed between studies and that individual compensation strategies were not well accounted for [3]. And in 113 volleyball players examined with ultrasound, reduced ankle dorsiflexion was the only measured variable associated with patellar tendon abnormality — a cross-sectional finding, and the authors framed it as a shock-absorption hypothesis rather than a proven chain [8].

So the honest version of the headline: ankle restriction reliably shows up as altered mechanics further up the leg, and those altered mechanics are the ones associated with problems elsewhere. Association is not causation, and no one has shown that restoring dorsiflexion removes a knee complaint. What the evidence does support is narrower and still useful — the ankle changes how the leg above it moves. Related reading: knee pain usually starts above and below the knee.

Where the block actually sits

Two people fail the wall test identically and need opposite work. The difference is what they feel at the end of the movement. A hard, blunt pinch at the front of the ankle behaves like a joint reaching its limit. A drawn-out stretch through the back of the calf and heel behaves like tissue running out of length. The clinical framework for dorsiflexion restriction maps these zones explicitly, and it assigns different interventions to each [6].

One further split is worth knowing. Test with the knee straight and you are largely reading the two-joint calf muscle that crosses the knee. Test with the knee bent, as the wall test does, and that muscle is slackened, so what remains is the deeper calf and the joint itself. Same ankle, two different answers — which is exactly why a general stretch aimed at everything often moves nothing. See also stiff or tight? they are not the same thing.

What actually changes the number

Manual joint work has the clearest short-term evidence. A meta-analysis of nine randomized trials in 364 people with chronic ankle instability found joint mobilization produced a large improvement in dorsiflexion range, a standardized mean difference of 1.02, and a moderate improvement in dynamic balance. Patient-reported function did not reach significance [5]. Read that honestly: the technique moves the range, and the range alone does not carry the function with it.

Stretching works, at doses most people never approach. The athlete framework specifies high-intensity chronic stretching above 1,200 seconds per week — that is twenty minutes of accumulated stretch time, spread across five to seven sessions, sustained for at least five weeks [6]. Consistency is the active ingredient. Two long sessions on a Sunday do not add up to the same thing.

The combination outperforms either part. In a randomized controlled trial, thirty female athletes with restricted weight-bearing dorsiflexion completed eight weeks of a program pairing joint mobilization and soft-tissue work with progressive loaded heel-raise strengthening, three sessions a week. Dorsiflexion improved by roughly 7 degrees, dynamic balance by 7 cm, and joint-repositioning error dropped by about 3 degrees [7]. Range, balance and position sense moved together — because the program trained all three.

How to train ankle mobility: test, then load

The sequence below is a structure, not a prescription. It follows the order the evidence supports: measure, locate, open, load, retest.

1. Test both ankles with the knee-to-wall lunge and write the numbers down. Left and right, in centimeters.
2. Locate the block. Front pinch or back stretch — the answer decides everything that follows.
3. Open the position. Loaded rocking with the knee tracking over the toes for the joint; long accumulated calf stretch time for the tissue.
4. Load the new range immediately. Slow heel raises through full range, deep split-squat holds, controlled knee-over-toe step-downs.
5. Retest weekly, same conditions. Compare to your own previous number, never to a reel.

• FOCUS: DORSIFLEXION
• TEST: KNEE TO WALL
• RETEST: WEEKLY

Note what is missing from that list: a fixed number to reach. The target is your other ankle, then last month's version of this one. Note also where the work belongs. An ankle trained in isolation twice a week competes with the eight hours a day it spends in a shoe on flat ground. The 12-Week Movement Protocol places ankle work inside full-body flows, where the joint is loaded in the positions that actually ask something of it — the range is built and used in the same session.

Five workouts per week, about 20 minutes each, with the ankle inside the pattern rather than beside it. Tested. Loaded. Kept.

Ankle mobility earns its reputation for a reason that has nothing to do with the ankle. It is the first joint in the chain, and the research is consistent that when it is short of range under load, the leg above it moves differently — more inward knee travel, a shallower and more forward-pitched squat, altered landing mechanics. What the research does not support is the leap that usually follows: that fixing the ankle resolves the knee. The evidence is associative, and the honest position is to say so.

What is defensible is the method. Test in weight-bearing, because the passive version predicts nothing. Locate the restriction, because a pinch and a stretch need different work. Then load whatever range you open, because the body keeps the capacity it uses. Test. Load. Keep.