Ankle Mobility for Squatting: How to Screen Yours and Fix the Common Restrictions

Heels peeling off the floor at the bottom, knees folding in, the bar drifting out over your toes: the usual suspect is a stiff ankle, and the usual fix is a random calf stretch that changes nothing. Ankle mobility is measurable. Two minutes with a tape measure tells you whether the ankle is your limiter, how far off you are, and which side, and from there you fix the restriction you have instead of the one a video assumed you have.

Why the ankle decides how your squat looks

To sit into a full squat with your heels down, the shin has to travel forward over the foot, and that travel is ankle dorsiflexion: a heels-down deep squat needs about 35 degrees of it, where walking gets by on roughly 10 (Murofushi et al. 2025). An ankle that handles daily life without complaint can still stop your squat at parallel.

When the shin cannot travel, the body finds depth somewhere else. Researchers restricted dorsiflexion in 30 healthy adults with a 12-degree wedge under the forefoot and the squat changed on cue: less knee flexion, more knee valgus, a large jump in medial knee displacement, less quad activity and more soleus, the same pattern seen in people with patellofemoral pain (Macrum et al. 2012). The numbers on depth point the same way. In 101 healthy adults, bent-knee ankle dorsiflexion plus hip flexion range explained 43.5% of the variance in squat depth in men, and in women, straight-knee dorsiflexion and dorsiflexor strength explained 32.4% (Kim et al. 2015), so hips and technique matter too. The ankle is the biggest single piece.

Outside the rack too. In 35 active adults, more dorsiflexion range went with more knee bend and lower ground reaction forces on landing (Fong et al. 2011), which is why a stiff ankle also turns up in plyometric work and in running knee pain.

The knee-to-wall test: two minutes, a tape measure, a wall

The weight-bearing lunge test, knee-to-wall to most lifters, is the standard clinical screen for ankle dorsiflexion. It holds up. The original reliability study reported coefficients of 0.97 to 0.99 (Bennell et al. 1998), a novice with a tape measure carries a measurement error of 0.4-0.6 cm (Konor et al. 2012), and it predicts how you squat where a passive ankle stretch on a table does not: in 40 active adults, a limited knee-to-wall score meant less knee flexion and less ankle travel in the squat, while non-weight-bearing dorsiflexion predicted nothing (Dill et al. 2014).

How to do it:

• Set up. Barefoot. Face a wall with one foot pointing straight at it and a tape measure on the floor running out from the wall.
• Drive the knee to the wall over the second toe, heel flat, and if it touches, slide the foot back 1 cm and go again.
• Stop where the heel lifts or the knee no longer reaches. Your score is the distance from the wall to the tip of the big toe at the last position where the knee touched with the heel down.
• Do both sides. Three attempts each. Take the best and write both numbers down.

Two reference points. Clinicians generally treat 9-10 cm, or 40-42 degrees of shin angle, as the normal mark, and the degrees version is what you get from an inclinometer app instead of a tape (Tourillon et al. 2025). And in 71 healthy men asked to sit in a heels-down deep squat, the knee-to-wall distance that best separated the 55 who could from the 16 who could not was about 11 cm: 10.75 cm on the right with 80% sensitivity and 81% specificity, 11.25 cm on the left, with the group that could averaging 12.7-12.9 cm and the group that could not averaging 8.1-8.6 cm (Kasuyama et al. 2009). Under 11 cm, a full heels-down squat is a coin flip. Under 8, the ankle is the first thing to look at.

A side-to-side gap above 1.5 cm counts as a clinically relevant asymmetry (Tourillon et al. 2025), and a two-legged squat hides it: you shift onto the mobile side, one knee caves, and the bar drifts without you feeling it. That is why you test both.

No tape measure? Drop into a relaxed heels-down squat, stay there, and grade yourself the way 92 adults were graded in one study: those who sat comfortably averaged 23 degrees of bent-knee passive dorsiflexion, those who needed effort to hold it averaged 16, and those who fell backwards averaged 9 (Kim JY et al. 2025, Squatting Posture Grading System for Screening of Limited Ankle Dorsiflexion, Annals of Rehabilitation Medicine).

What your score means

Read each side against the table. It is built from the norms and cut-offs above and it reads per side, so a lifter at 12 cm right and 7 cm left reads two rows plus the asymmetry row. Two thresholds are in play. Clinicians call 9-10 cm a normal ankle; a heels-down deep squat needs closer to 11. The table reads against the squat, so borderline here means borderline for depth, not abnormal. The 11 cm cut-off comes from 71 young men, so read it as a guide rather than a line, and read your own retests against your own first number.

Find your restriction: muscle or joint

Stretching a joint block does nothing, and mobilizing a joint that already glides wastes your evening. Two checks sort it.

• Bent knee against straight knee. Stretch the calf with the knee locked, then with the knee bent, and notice the difference: markedly freer with the knee bent means the gastrocnemius, which crosses the knee, is the tight tissue, while little change means the limiter sits deeper, in the soleus or the joint. This is the check that matters for squatting. The knee is bent at the bottom, the gastrocnemius goes slack, and the soleus carries the stretch, and ultrasound work agrees: adults who could not hold a heels-down deep squat had a soleus that had not lengthened by 40 degrees of dorsiflexion, where the deep-squatters' had (Murofushi et al. 2025).
• Where you feel it. Hold the end of the knee-to-wall lunge. A pull in the back of the calf or the Achilles is soft tissue, and a pinch or a hard stop at the front of the ankle is the joint, the talus not gliding back under the tibia. Old sprains live here. Lost dorsiflexion after a sprain is common enough that restoring it is a standard rehab target (Terada et al. 2013).

Front pinch: joint work first. Calf pull: soft tissue first. Both: both, joint first in the session.

Fix the soft-tissue restriction

Stretching works on the calf. The effect is smaller and slower than most lifters expect, so the dose has to be real.

• The acute effect is 2-3 degrees. Pooled trials of static calf stretching added about 2 degrees of dorsiflexion for total stretch time under 15 minutes and about 3 degrees for 15-30 minutes (Radford et al. 2006), which is enough to change how a session feels and not enough to change your squat on its own.
• The chronic effect is about 5 degrees. Across 20 controlled trials, static stretching programmes improved dorsiflexion by 5.2 degrees on average (95% CI 4.4-6.0), PNF by 4.3, while ballistic stretching did not reach significance (Medeiros and Martini 2018). Five degrees is a little over a centimetre at the wall, which moves you a band at the boundary and not from the bottom of one.
• The dose in the research is more than 1,200 seconds of stretching a week for at least 5 weeks (Tourillon et al. 2025). That is 20 minutes a week. Four days of 5 minutes, straight-knee and bent-knee, with the bent-knee soleus stretch taking the larger share.
• Foam rolling buys a window. Three 30-second passes on the calf improved dorsiflexion for at least 20 minutes on that leg and about 10 minutes on the other (Kelly and Beardsley 2016), though it did not separate from the control group, so treat the window as a warm-up trick, not a fix. Two weeks of 3 x 60 seconds per leg, six days a week, added about a centimetre to the knee-to-wall score, most of which was still there a week after stopping (Seever et al. 2022). Roll first. Then stretch or mobilize, then squat inside the window.
• Load the new range. A stretch gives you range. Strength in it comes from a bent-knee calf raise on a step, with a 3-second lowering into a full stretch at the bottom, 3 sets of 8-10, twice a week, which teaches the soleus to be strong there. After a sprain, static stretching combined with a home exercise routine produced the largest dorsiflexion gains of any intervention reviewed, an effect size of 1.06 (Terada et al. 2013).

Fix the joint restriction

Front of the ankle pinches? Work on the talus and leave the calf alone for now.

• Banded knee-to-wall lunges. Loop a heavy band around the front of the ankle at the crease, anchor it behind you at floor height, and do your knee-to-wall lunge against the pull: 4 sets of 12-15 slow reps per side with a 2-second pause at the end. The band pushes the talus back as the shin comes forward, the direction it stopped gliding, which makes this the self-applied version of the mobilization-with-movement physios use. A meta-analysis of ankle-sprain trials found it added 0.9 cm to the knee-to-wall test over controls (95% CI 0.4-1.4, 144 participants), with a large pooled effect on range of motion (ElMeligie et al. 2025); the authors note 0.9 cm sits below the 1.3-1.5 cm they consider meaningful, so mobilization alone will rarely move you a whole band. The stretching and loading above finish the job.
• Dose it like a clinic would. 60 glides per session, 4 sets of 15, 2-3 sessions a week, and up to double that if the stiffness is old (Tourillon et al. 2025). Retest at the wall straight after.
• Know when to hand it over. Three weeks of consistent work that moves nothing, a block that feels bony rather than stiff, or an ankle that swells: that is a physio's job, and if the history includes a fracture, read returning to training after injury before you force range.
• One exception to all of this. Pain at the back of the heel where the Achilles meets the bone changes the advice: end-range dorsiflexion under load compresses an insertional Achilles tendinopathy against the heel bone, so the deep stretching and banded work here make it worse. Follow that protocol first and come back to the ankle screen once the heel is quiet.

Squat while you fix it

You do not stop squatting for 6 weeks because your ankle scored 6 cm. You change the geometry.

• Raise the heel. Weightlifting shoes sit at roughly 20-22 mm under the heel, and they cut the dorsiflexion a given depth demands: in 32 lifters, lifting shoes reduced ankle flexion and increased knee flexion compared with trainers, with a more upright trunk (Legg et al. 2017), and a review of 14 studies found wedges under the heels do the same job (Pangan and Leineweber 2021). No shoes? A 2.5 kg plate under each heel is the free version, and a plate calculator tells you which plates are spare once the bar is loaded.
• Pick the variation that fits the ankle you have. Front squats and goblet squats ask for more forward shin travel. They wait. A box squat to a height you reach with heels down keeps the pattern loaded while the range arrives, and a wider stance with the toes out asks the ankle for less.
• Do the ankle work first. Roll, stretch, banded lunges, then squat inside the 20-minute window, with loads lighter and the tempo controlled while the new depth is unfamiliar. Work off a percentage of a known one-rep max rather than what felt heavy last week, since a deeper squat is a different lift; an RPE calculator keeps the load honest when the movement changes under you, and a warm-up set calculator gets you to a working weight with the range already grooved.
• Retest every 4 weeks, same wall, same time of day. The smallest change that beats measurement error is 1.1-1.5 cm with a tape measure in the novice-rater study (Konor et al. 2012) and 1.6-1.9 cm across the systematic review of the test (Powden et al. 2015), so under that is noise, and two retests in a row that clear it is progress. Keep the heel as long as you need it; plenty of strong lifters squat in heels for their whole career.

Common mistakes

• Stretching the straight-knee calf only. The squat bottoms out with a bent knee, where the soleus limits you, so the bent-knee version takes the larger share of your stretching time.
• Testing with the heel drifting up. Let it peel and you hand yourself 2-3 cm of score you do not have, then go looking for the problem in your hips.
• Chasing a number you do not need. A wide-stance low-bar squatter needs far less dorsiflexion than a weightlifter catching a snatch, and if your squat hits depth with heels down and knees tracking, stop testing. Go train.
• Ignoring the short side. A 1.5 cm gap is where the bar drift and the one caving knee come from. Work only the short side until it closes.
• Mobilizing through pain. A pinch that sharpens from session to session is a reason to stop and get it looked at.

How Movement Rebels fits

Your knee-to-wall score lives in your head unless you put it somewhere the coach reads. Write it into "Notes for your coach" in the Movement Rebels app: "knee-to-wall 6 cm left, 10 cm right, squatting in heels until it closes." The coach reads that note alongside your logged lifts and PRs, so it knows the 6 cm ankle and the 140 kg back squat belong to the same person. Ask it in chat to add a 10-minute ankle block before each squat day and it puts them in your calendar. If you also run, the pace and cadence coming in from Garmin or Apple Health tell it what your ankles are already absorbing that week, so your weekly plan accounts for both the strength work for runners and the mobility work instead of stacking them blind. And if your HRV and resting heart rate say you slept badly, ask the coach to move the heavy squat. Better that than pushing through on a stiff ankle and a flat nervous system.

Measure both sides tonight. Write the numbers down. In 4 weeks, measure again at the same wall.