Best Shoes for Haglund’s Deformity: It’s the Heel Collar, Not the Cushioning
A Haglund's bump hurts when a shoe presses on it. How to test the heel collar, why a higher drop helps, and what the one randomised heel-lift trial really showed.
On this page
- The short answer
- Haglund’s, bursitis or tendinopathy? The distinction changes what a shoe can do
- Why this is not the same as ordinary Achilles tendonitis
- The four things to check, and how to check them in the shop
- The numbers, side by side
- The heel-lift question, and what the trial actually showed
- A tall heel is not a heel lift
- The four shoe modifications the literature names
- What we have not measured, and will not pretend to have
- Frequently asked questions
- What we are testing next
Testing status: nothing on this page has been worn or scored by us. The figures in the table are each brand’s own published specification, or a named third party where the brand publishes none, and the source is printed beside every one. The one property that matters most for this diagnosis — how soft the heel collar is — we have not measured on any shoe, and we say so below rather than guessing.
The short answer
A Haglund’s deformity is a bony prominence on the back of the heel bone, where the Achilles tendon attaches. The bump itself is not the problem. It becomes painful when something presses against it from outside, or when the tendon is loaded against it from inside — and both of those are footwear problems before they are anything else.
So the shoe has three jobs here, and they are not the ones most “Achilles” shoe advice talks about: keep anything firm off the back of the heel, raise the heel so the tendon is not stretched taut across the bone, and leave enough room inside to add a heel lift if you need one. A shoe that is excellent for almost every other foot can be exactly the wrong shoe for this one, because the part that hurts you is the part nobody else notices.
- You felt a sudden snap or pop at the back of the ankle and cannot push off, rise onto that toe, or walk normally. That is an Achilles rupture until proven otherwise, it is frequently missed in the first week, and no shoe on this page is the answer to it.
- The skin over the bump is broken or weeping, or you have diabetes or reduced sensation in your feet. A pressure sore over a bony prominence in a foot that cannot feel it is an urgent problem, not a blister.
- The heel is hot, red and swollen with a fever, or the pain woke you and has not settled. Infection and inflammatory arthritis both present in exactly this spot.
Haglund’s, bursitis or tendinopathy? The distinction changes what a shoe can do
Three names get used here as though they were interchangeable, and separating them tells you how much footwear is actually going to buy you.
- Haglund’s deformity is the bone — a prominence at the upper back corner of the heel bone. It is a shape, largely the one you were born with. On its own it is painless, and plenty of people have one and never know.
- Retrocalcaneal bursitis is the sac that sits between that bone and the Achilles tendon becoming inflamed. This is the part that swells, and the part the back of a shoe presses on. Footwear helps most here, and quickly.
- Insertional Achilles tendinopathy is the tendon degenerating where it attaches to the bone, sometimes with calcification inside it. Footwear helps, but as a supporting act — the treatments with trial evidence behind them are loading and shockwave.
That middle one is the mechanism behind everything else on this page. Squeeze a bursa between a bony ridge and a stiff collar ten thousand times a day and it stays swollen; take the pressure off and it settles. It is also why the swelling can go down markedly while the bump itself stays exactly where it was — you are not shrinking bone, you are letting the soft tissue over it calm down.
Why some heels get this and others never do
Worth knowing, because it takes the blame off your shoe cupboard. In 2021 Agnieszka Wnuk-Scardaccione and colleagues in Kraków compared 30 runners who had had retrocalcaneal bursitis with 30 who never had, classifying the surface shape of the heel bone on X-ray and measuring the tendon and bursa on ultrasound. A flat surface of the calcaneal tuberosity raised the odds roughly fourfold (OR 4.3), and a calcaneal slope above 25° carried odds of 2.8 against a more horizontal heel bone; a thicker Achilles tendon and a smaller bursa were both strongly associated too.
The limits are worth stating plainly: 60 people in total, all of them runners, and a case-control design comparing people who had already had the problem — so it describes who gets it rather than proving what causes it, and it cannot say whether the thicker tendon came before the bursitis or after. It studied no footwear at all. What it does support is the framing this page runs on: the shape is yours, and the shoe decides whether it hurts.
Why this is not the same as ordinary Achilles tendonitis
This distinction changes the advice completely, and it is the single most useful thing on this page. Mid-portion Achilles tendinopathy sits in the tendon itself, a few centimetres above the heel bone. Insertional tendinopathy — the kind that travels with a Haglund’s bump — sits right where the tendon meets the bone.
For mid-portion problems, the standard rehab is heel drops off the edge of a step, taking the heel below level. At the insertion, that same end-range position presses the tendon against the very bone it attaches to, which is the thing already hurting. Advice written for one can make the other worse, and a great deal of what circulates online does not distinguish between them at all. Our guide to shoes for Achilles tendonitis covers the mid-portion case; this page is the insertional one.
If you do not know which you have, that is a question for an examination rather than a shoe. The rough rule patients find useful: press with one finger. Pain in the soft tendon above the heel bone points to mid-portion; pain on the bone at the very back, often with a visible or palpable bump and a shoe rim that catches it, points to the insertion.
The four things to check, and how to check them in the shop
- 1. Press the heel collar with your thumb, at the height the bump sits. It should be soft, unseamed and yielding at that exact point. Firm plastic, a moulded logo, a stiffened rim or a seam crossing the back of the heel will all press the bump. This is the check that matters most and the one no spec sheet answers.
- 2. Find the top edge of the heel counter. Run a finger up the back of the shoe until the stiffness stops. You want that rim sitting clearly below your bump or clearly above it — never level with it. A rim that lands on the prominence concentrates every step onto the sorest square centimetre you own.
- 3. Check the heel height. The more the shoe lifts your heel, the less your ankle has to bend upward, and the less the tendon is drawn tight across the back of the bone. Drop is the published figure for this, and it is in the table below. Flat and zero-drop shoes, which suit plenty of other feet, are the hardest category here.
- 4. Lift the insole out. If it comes out, you can add a heel lift and still have room for your foot. If it is glued down, you cannot, and the shoe has quietly ruled out the one intervention with a trial behind it. Our Orthotic Volume Rating explains what to look for once the liner is out.
One more, free: look at the shoes you already own. A worn patch, a shine, or a collapsed spot on the inside of the heel collar tells you exactly where your bump has been rubbing, and how high it sits. Bring that pair with you.
The numbers, side by side
These are the eight shoes on this site with the highest published heel-to-toe drop — the measure of how much a shoe raises your heel relative to your forefoot, and the one specification that speaks directly to tension at the Achilles insertion. Every figure is the brand’s own unless the last column says otherwise.
Read down the lift column rather than the heel column and two things that are easy to confuse come apart. Three of these shoes carry a heel stack of 42–43 mm and deliver 18, 10 and 8 mm of lift; the shortest heel in the set, 35.7 mm, gives more lift than the tallest HOKA. That is the section two headings below, and it is the single most common mistake people make buying for this problem.
| Shoe | Drop | Heel stack | Weight | Who published the drop |
|---|---|---|---|---|
| Dansko XP 2.0 | 18 mm | 43 mm | — | Derived by us |
| Orthofeet Coral | 15.7 mm | — | — | The brand |
| Brooks Addiction Walker 2 | 12.2 mm | 35.7 mm | 411 g | Derived from lab stacks |
| Brooks Beast GTS 26 (Ariel GTS 26) | 12 mm | 38 mm | 354 g | Brooks |
| Brooks Adrenaline GTS 24 | 12 mm | 39 mm | 283 g | Brooks |
| Saucony Triumph 24 | 10 mm | 43 mm | 250 g | The brand |
| Brooks Ghost 18 | 10 mm | 36 mm | 289 g | The brand |
| HOKA Clifton 11 | 8 mm | 42 mm | 283 g | HOKA |
A dash means nobody publishes that figure and we have not measured it. Figures are men’s unless the review says otherwise.
Two things in that last column are worth pausing on, because they are the kind of thing a spec sheet hides. Dansko publishes no heel-to-toe drop at all — the 18 mm figure is ours, derived by subtracting the platform height from the heel height on Dansko’s own listing, and we label it that way rather than presenting it as the brand’s. Brooks publishes no drop for the Addiction Walker 2, so that figure comes from subtracting an independent lab’s measured stacks. The two highest-drop shoes we can name are both shoes whose makers never tell you the number.
The heel-lift question, and what the trial actually showed
Everything above this heading is clinical reasoning. This is the part with published evidence behind it, and the part that is still judgement.
Raising the heel is the oldest advice in this condition, and it has finally been tested in a randomised trial aimed squarely at the insertional problem: the LIFTIT trial, by Bourke, Munteanu, Garofolini, Taylor and Malliaras, published open access in the Journal of Foot and Ankle Research in 2024.
- Twenty-six people, twelve weeks, heel lifts against a convincing sham. The comparison group wore flat full-length innersoles made of the same materials and branded identically, which is a far better control than most footwear studies manage.
- The signal favoured the heel lifts on every patient-reported measure. Pain fell 3.2 points against 2.7 in the sham group, and the between-group effect sizes ran from moderate to large across pain, the VISA-A tendon score, lower-limb function and quality of life. Thirty-one per cent more of the lift group rated themselves improved overall.
- Half the participants asked for a lower lift than the one they started in. They began at 12 mm and the group finished at a mean of about 10.8 mm. If you try this and 12 mm feels like too much, you are in the majority, not doing it wrong.
- Sixty-five per cent reported an adverse event — none of them serious. Mostly new aches from a sudden change in geometry. The authors’ own recommendation for future trials is a gradual wearing-in period, which is exactly what we would tell a patient.
And here is the limit the authors state themselves, in plain terms: this was a feasibility trial, and it was not powered to evaluate efficacy. Its job was to find out whether a full trial could be run — recruitment, retention, acceptability — and to collect a preliminary signal. The authors estimate a properly powered study would need somewhere between 47 and 241 participants depending on the outcome chosen, and note their design included no measure of why heel lifts might work. So the honest reading is: the direction is encouraging, the mechanism is assumed rather than demonstrated, and anyone telling you heel lifts are proven for this condition is ahead of the evidence.
- Wnuk-Scardaccione A, Mizia E, Zawojska K, Bilski J, Wojdyła J. Surface Shape of the Calcaneal Tuberosity and the Occurrence of Retrocalcaneal Bursitis among Runners. International Journal of Environmental Research and Public Health 2021;18(6):2860. doi:10.3390/ijerph18062860
- Bourke J, Munteanu SE, Garofolini A, Taylor S, Malliaras P. Efficacy of heel lifts for insertional Achilles tendinopathy (LIFTIT): A randomised feasibility trial. Journal of Foot and Ankle Research 2024;17(4):e70025. doi:10.1002/jfa2.70025 — open access.
A tall heel is not a heel lift
This is the mistake that sends people home with the wrong shoe. Heel stack height is how much material sits under the back of your foot. Heel-to-toe drop is how much more sits under the back than under the front. Only the second one tilts the foot, and only the second one shortens the Achilles. A 43 mm heel with a 43 mm forefoot under it is a platform, not a lift, and it does nothing whatsoever for tendon tension.
Thirty-seven of the 43 shoes on this site publish a heel-to-toe drop. The median is 8 mm and the mean is 7.5 mm. Twelve shoes have a heel stack of 40 mm or more — the ones that look and feel tall in the shop. Of those twelve, exactly two have a drop above the median. Five are below it. The tallest running shoe on the list has one of the flattest figures on the list.
| The tall shoes | Heel stack | Drop | What that means here |
|---|---|---|---|
| Dansko XP 2.0 | 43 mm | 18 mm | Tall and steeply tilted — the only shoe here that is both |
| Saucony Triumph 24 | 43 mm | 10 mm | The only closed lace-up here that is tall and above the median drop |
| New Balance 860v15 | 43 mm | 8 mm | Exactly median — tall adds nothing |
| HOKA Clifton 10 · 11 | 42 mm | 8 mm | Median drop in a very tall shoe |
| ASICS Gel-Kayano 32 · GT-2000 15 | 40 mm | 8 mm | Median drop |
| Skechers Aero Burst | 42 mm | 6 mm | Below median despite the height |
| HOKA Bondi SR · NB 880v15 | 40 · 40.5 mm | 6 mm | Below median |
| HOKA Bondi 9 | 43 mm | 5 mm | Joint-tallest heel on the site, second-flattest drop in this table |
| NB Fresh Foam X More v6 | 42 mm | 4 mm | Flattest tall shoe here |
Stack and drop figures as published by each brand or measured by RunRepeat’s laboratory; every figure and its source is listed in the shoe spec database. Drop is a whole-shoe figure and does not tell you the shape of the heel collar above it.
None of this makes the tall shoes bad. It makes them irrelevant to the specific mechanism at issue here. If you are buying for tendon tension, read the drop column and ignore the height. Our heel-to-toe drop chart sets the stated figure against the measured one for every shoe where both exist, because the two disagree more often than you would expect.
The four shoe modifications the literature names
The clearest published list of conservative footwear options for rearfoot disorders comes from a 2020 narrative review in Annals of Palliative Medicine by Choo, Park and Chang. For Haglund’s deformity it names four shoe modifications, and it is worth reading them as four separate strategies rather than a single recommendation, because they conflict.
- Move the quarter trim line. The top edge of the shoe’s back is cut either above the prominence or below it, so that the rim never rests on the bump. Above means the pressure is spread over a taller, softer collar; below means there is nothing at that height at all.
- Add a heel lift. Raising the heel inside the shoe lifts the tendon insertion above the counter’s rim and shortens the Achilles slightly. This is the modification a shoe’s heel-to-toe drop already performs, which is why drop is the specification that matters most on this page.
- Pad the inside of the heel counter. A soft insert behind the heel spreads the same load over more area. It does not remove the pressure; it changes its distribution.
- Wear sandals or clogs. The simplest answer: a shoe with no closed back cannot press on the bump. The review lists it alongside the others rather than as a last resort.
Source: Choo YJ, Park CH, Chang MC. Rearfoot disorders and conservative treatment: a narrative review. Annals of Palliative Medicine 2020;9(5):3546–3552. For retrocalcaneal bursitis the same review adds a heel lift and loosening the shoe counter, and notes that trials of those are rare.
Then comes the sentence that ought to be printed on every product page that uses the word “Haglund’s”. The review states, of those four modifications, that their effects have not so far been studied, and that clinical trials evaluating the effectiveness of shoe modifications should be performed. That is the honest state of the evidence in 2026: the mechanical logic is sound and uncontroversial, the clinical experience is broad, and the trials do not exist. Anyone selling you a shoe on the strength of a study has a study you should ask to see.
The fourth of those — sandals or clogs — is the one the shoe table above cannot answer, because every shoe in it has a back. The Dansko XP 2.0 already named above is an open-backed clog and covers it for work. For the hours at home, where a closed shoe is doing the damage and bare feet are doing no favours either, there is a simpler option on this site.
Nothing to press on the bump · $59.95
OOFOS OOahh Slide Spec analysis
A slide has no back, no counter and no collar, which makes it the least complicated answer to a pressure problem: the shoe cannot touch the part that hurts. Its 20 mm heel is modest by the standards of this site, and OOFOS does not publish a forefoot figure, so we cannot give you a drop for it and will not guess one. What it is genuinely good for is the hours at home and around the house, where a closed shoe is doing the damage and bare feet are doing no favours either.
The honest caveat: this is a recovery slide, not a walking shoe. There is no heel hold, no width options — one width, whole sizes only — and no published drop. It answers exactly one of the four modifications and makes no claim on the other three.
| Published figure | Value |
|---|---|
| Heel stack | 20 mm OOFOS |
| Forefoot stack | Not published anywhere — |
| Heel-to-toe drop | Cannot be derived — |
| Back of shoe | Open — slide OOFOS |
| Widths | One width, whole sizes OOFOS |
| List price | $59.95 OOFOS |
Added source for this section: Choo YJ, Park CH, Chang MC. Rearfoot disorders and conservative treatment: a narrative review. Annals of Palliative Medicine 2020;9(5):3546–3552. doi:10.21037/apm-20-446 (PMID 32787369). The four modifications, the retrocalcaneal-bursitis advice and the statement that their effects have not been studied all come from this review.
What we have not measured, and will not pretend to have
The decisive variable on this page is how soft the heel collar is at the height your bump sits, and we hold no measurement of collar softness for any shoe in our catalogue. Nor does anybody else publish one: it is not on a spec sheet, no laboratory reports it, and it varies between colourways of the same model because the materials change.
That is why this page ranks nothing and recommends no single shoe to buy. The table above is sorted by a figure we can source, not by the property that will decide whether a shoe hurts you — and pretending otherwise would be exactly the kind of confident guess this site exists to avoid. Use the drop column to narrow the field, then use your thumb on the collar to settle it. Building a repeatable bench test for heel-collar stiffness is on our list; until it exists, your thumb is better evidence than our opinion.
Frequently asked questions
Should I just cut the back out of my shoes?
People do, and it works in the short term because it removes the pressure completely. The problem is that a shoe with no heel counter no longer holds your heel, so you get instability and often a new complaint somewhere else within a few weeks. A backless clog or slide is a reasonable rescue for the worst fortnight, not a plan. The better version is a shoe whose collar is soft where your bump is and firm lower down, which gives you both.
Does a higher heel-to-toe drop really help?
The reasoning is sound and the trial evidence points the same way: raising the heel reduces how far the ankle has to bend upward, which reduces both the stretch on the tendon and how hard it is pressed against the back of the heel bone. But the LIFTIT trial that tested it directly was a feasibility study, not a powered efficacy trial, so treat a higher drop as well-reasoned and encouraging rather than proven.
How high should a heel lift be?
The trial started people at 12 mm and about half asked to come down, finishing near 10.8 mm on average. Two-thirds reported some new ache along the way, none of it serious, and the researchers’ own suggestion is to build up gradually rather than start at full height. Put the lift in both shoes, not just the painful side, or you introduce a leg-length difference on top of the problem.
Will a softer shoe fix it?
Not by itself, and this is where most advice goes wrong. Softness under the heel does nothing about a rim pressing on the back of it. The two properties are unrelated — a very cushioned shoe can have a firm, high, seamed collar, and a firm shoe can have a soft one. Test the collar separately from the midsole.
Is the bump going to keep growing?
A Haglund’s prominence is bone, and it does not resolve with footwear. What footwear changes is whether it is loaded and rubbed, which is what produces the pain, the swelling and the bursitis behind it. Managing the pressure well can keep a bump quiet indefinitely; it does not make it smaller.
When should I stop experimenting with shoes and be seen?
If the back of the heel is swollen and warm, if the pain wakes you or is there on the first steps every morning for more than a few weeks, if you cannot push off to climb stairs, or if there is a sudden sharp change with weakness — that last one needs same-day assessment rather than a shoe. Persistent insertional pain also has treatment options beyond footwear, and they work better earlier than later.
What we are testing next
- A repeatable bench measurement of heel-collar stiffness and height — the figure this page needs and nobody publishes. It is the same gap as heel-counter stiffness on our other pages, and the same rig would answer both.
- Where the top edge of the heel counter actually sits on each shoe we hold, measured from the footbed, so the rim-versus-bump question becomes a number rather than a thumb test.
- Whether any of the high-drop shoes above stay high-drop once a lab cuts them open — on every shoe we have compared so far, measured drop has come out higher than the brand states, and the missing millimetres were under the forefoot. The figures are on the heel-to-toe drop chart.
Recently reviewed
Individual shoe pages state their evidence tier, published measurements and who should skip them.
HOKA Clifton 11
Spec AnalysisHOKA’s product page carries a podiatric seal claim the APMA database does not support for this version, and a women’s stack height that contradicts HOKA’s own drop. Not tested by us.
Skechers Hands Free Slip-ins
Spec AnalysisUntested. A heel panel engineered to stay open cannot also be a rigid heel counter, and Skechers publishes no stiffness figure for it — nor a drop, a weight or a width.
New Balance Made in USA 990v6
Spec AnalysisNew Balance publishes no weight, drop or stack height for the 990v6, but it does publish B through 6E widths — and that range is why it still matters. Not tested by…
Measured, worn, scored — never sponsored
No brand has ever paid for a review, a score, a ranking or the seal. Where we earn a commission the link says so, and the score does not change.