Runner's Knee: Why Your Knees Hurt When You Run

You started running a few weeks ago. It was going well. And now there is an ache around the front of your kneecap that shows up a mile in, gets worse going down stairs, and grumbles when you stand up after sitting at a desk all afternoon. The obvious conclusion, the one almost everyone reaches, is that your knees are not built for running.

Almost certainly, they are. Knee pain is the single most common thing that happens to new runners, and the reason it is so common has very little to do with your knees being defective. This guide covers what runner’s knee actually is, why the fix usually lives at your hip rather than your knee, and the order to do things in. If you want the broader picture of what a stride can tell you, the cornerstone guide to running gait analysis is the place to go next.

One thing up front: this is general information, not a diagnosis. There is a short list of symptoms further down that means “stop reading and go see someone,” and it is worth skipping ahead to that first if anything about your pain feels dramatic.

What runner’s knee actually is

“Runner’s knee” is a nickname, and the thing it usually refers to is patellofemoral pain, meaning pain at the joint between your kneecap (the patella) and your thigh bone (the femur).

Your kneecap is not a fixed bolt. It sits in a shallow groove at the end of the femur and glides through that groove every time your knee bends and straightens. Running sends a lot of force through that contact, and the pain tends to come from how that load is distributed rather than from anything being torn or broken.

The pattern is usually recognisable:

• A dull, diffuse ache around or behind the kneecap, rather than one sharp point you can put a fingertip on.
• Worse going down stairs or hills than going up.
• Worse when you sit with the knee bent for a long time, which is common enough that clinicians nickname it the “moviegoer’s sign.”
• It often warms up a little during a run, then returns worse afterwards.

That last one is the trap. A knee that feels better ten minutes into a run convinces a lot of new runners that they can run through it. Sometimes you can. Often the run is quietly making a withdrawal you have to pay back later in the week.

You are not unusual

It helps to know the scale of this. A systematic review pooling the research on patellofemoral pain found the incidence among amateur runners in the general population to be roughly 1,080 cases per 1,000 person-years. Read that number carefully, because it is easy to misread: it does not mean everyone gets it. It is a rate that counts repeat episodes, so it reflects a condition that both affects a large share of runners and tends to come back.

The useful takeaway is not the exact figure. It is that this is the most well-trodden path in recreational running, it has been studied heavily, and there is a reasonably clear picture of what helps.

Does running ruin your knees?

No, and the evidence points modestly the other way for recreational running. A systematic review and meta-analysis of hip and knee osteoarthritis found arthritis in 3.5% of recreational runners, compared with 10.2% of sedentary people and 13.3% of competitive elite runners. Recreational running was associated with a lower occurrence of arthritis, not a higher one.

Two honest caveats, because this finding gets flattened in both directions.

First, this is observational, so some of the gap is self-selection: people whose knees hurt tend to stop running and end up counted as sedentary. Nobody should claim running is proven to protect cartilage.

Second, look at the competitive column. The elite group, meaning many years of very high volume, came out worse than the recreational group. The relationship is not “more running is always better.” It looks more like a dose curve where recreational amounts sit in a good place.

Your knee hurting after six weeks of running is not early arthritis and it is not your joints wearing out. It is far more likely to be a tissue that has not yet been given time to catch up with your enthusiasm.

Why the fix is usually your hip

Here is the part that surprises people. The most consistently effective conservative treatment for patellofemoral pain is not knee exercises. It is hip exercises.

The evidence for this is genuinely strong. A systematic review with meta-analysis found that adding proximal work, meaning hip and trunk strengthening, to quadriceps training produced better pain reduction and function than quadriceps training alone, with the advantage still visible at one-year follow-up. The authors concluded that proximal rehabilitation should be part of standard conservative management.

The mechanical logic is straightforward once you see it. Your kneecap tracks in a groove on your femur. If the hip muscles are not controlling the femur well, the femur rotates and drops inward underneath the kneecap during single-leg loading. The kneecap has not gone anywhere; the bone underneath it has moved. That inward collapse is the sign covered in the guide to dynamic knee valgus, and it usually traces back to the lateral hip, the same muscle group behind contralateral pelvic drop.

So the knee is often the place that hurts, not the place that is failing. Which is good news, because hip strength is very trainable.

The other half: you probably did too much, too soon

Strength is one half. Load is the other, and for a brand new runner it is often the bigger half.

Your cardiovascular system improves fast. Within a few weeks of starting, your breathing settles, your easy pace quickens, and running simply feels easier. Your tendons, bone, and cartilage adapt on a considerably slower clock, over months rather than weeks.

That mismatch is the beginner injury engine. Your fitness gives you permission to run more before your structure is ready to absorb it, and because the extra running feels comfortable, nothing warns you. Almost every new runner who ends up with sore knees got there by feeling good and doing more.

You will see the “10% rule” quoted as protection here, meaning never increase weekly mileage by more than 10%. Treat it as a rough habit rather than a law, because trials testing it directly have not shown it prevents injury on its own. The durable principle underneath it is the useful bit: increase one thing at a time, gradually, and hold each step long enough to see how your body answers.

Should I keep running if my knee hurts?

Usually yes, but less of it, and only at a level that stays comfortable both during the run and the following day. Complete rest is rarely the right tool, because tissue that has been rested does not come back stronger, it comes back less prepared. The aim is to find the amount of running your knee currently tolerates and work from there.

A rule of thumb widely used in rehab is to treat pain as a traffic light. Mild discomfort that stays low, settles quickly after you stop, and is gone the next morning is generally acceptable to keep training through. Pain that climbs as the run goes on, makes you change how you move, or is still there the next day means that run was too much. It is a guide rather than a law, but it is a far better instrument than willpower.

Two things that are not acceptable regardless: pain that is sharp and pinpoint rather than dull and diffuse, and pain that makes you limp. Both are worth getting looked at rather than negotiated with.

What to do first

Roughly in this order.

1. Reduce, do not stop. Complete rest tends to be the wrong tool, because detrained tissue is not more robust. Drop to a volume and pace that is genuinely pain-free, including the day after. Walk-run intervals are useful here. If nothing is pain-free, that is a signal to get assessed.

2. Nudge your cadence up. This is the highest-value quick change available. Increasing step rate to 110% of preferred reduced peak patellofemoral joint force by about 14% in a biomechanics study, largely by reducing knee flexion during stance. Taking slightly shorter, slightly quicker steps means less force through the exact joint that hurts. Aim for around 5% to 10% above your current number rather than chasing a magic figure, and see the running cadence guide for how to find and shift yours.

3. Train the hip, and the quads. Based on the evidence above, do both, with real attention to the hip:

• Side-lying hip abduction and banded lateral walks, for the glute medius.
• Single-leg glute bridges and hip thrusts, for hip extension strength.
• Step-downs, lowering slowly with the knee tracking over the middle of the foot and the pelvis level. This directly rehearses the pattern that fails.
• Split squats and single-leg RDLs, for single-leg control under load.
• Spanish squats or leg extensions in a pain-free range, for the quadriceps, which the research supports alongside the hip work rather than instead of it.

Two or three sessions a week. Expect to give this six to eight weeks before judging it, because strength adaptations are not a one-week story.

4. Then rebuild volume slowly. Once you are running comfortably, add gradually, and change one variable at a time.

Notice what is not on this list: new shoes, arch supports, and stretching. None of those is forbidden, and if a stretch feels good then have at it, but none of them is the thing that resolves this. The evidence points at strength and load management, so that is where your effort should go.

How long does runner’s knee take to heal?

Longer than most people expect, and that is the single best argument for dealing with it properly now instead of running through it.

Plenty of people improve substantially within six to twelve weeks of consistent strength work and sensible load management. But patellofemoral pain has a reputation as self-limiting that the long-term data does not support. A multicentre study following people up five to eight years after diagnosis found that 57% still reported an unfavourable recovery. More than half were still dealing with it years later.

That is not written to frighten you, and it is emphatically not a prediction about your knee. It matters because of what the same research found about who does badly: having had symptoms for longer than 12 months was itself a predictor of a poor outcome. Duration is not a neutral bystander.

Which flips the usual instinct. The common approach is to wait and see whether it settles on its own, then act if it does not. The evidence suggests the opposite is smarter: act early, while the problem is young and the odds are best. Six weeks of proper hip work starting now is a far better bet than the same six weeks starting next spring.

When to stop self-managing and see someone

Book an appointment with a physiotherapist or doctor if any of these apply:

• The knee swells, visibly or as a feeling of fullness.
• It locks, catches, or gives way underneath you.
• Pain came from a specific traumatic moment, a twist, a fall, a pop.
• There is pain at rest or at night, or pain that is sharp and localised to one point.
• You have made sensible changes and there is no improvement after about six weeks.

None of that means something is seriously wrong. It means the situation deserves someone who can put hands on your knee, which no article and no app can do.

Where a gait check fits

If the hip is usually the culprit, the obvious question is whether your hip is the culprit, and that is exactly what a short video can start to answer. Film yourself running from behind and from the side, then look for the knee drifting inward under load and for one side behaving differently from the other.

This is where RunGait is designed to help: record a short clip, get the left-versus-right asymmetries surfaced automatically, and turn them into strength priorities rather than a vague instruction to work on your hips. Retest in a few weeks and see whether the gap closed.

Two honest limits on that. A single phone camera carries real measurement error, so trust the asymmetry between your sides rather than any precise number of degrees. And a screening tool gives you a training hypothesis, never a diagnosis. If your knee hurts, the strength work here is a sensible place to start, and a qualified professional is still the right call.

The short version

Your knees are almost certainly fine. Runner’s knee is the most common thing that happens to new runners, recreational running is not associated with wrecking your joints, and the two levers with the best evidence behind them are hip strength and a more patient approach to load. Shoes are not the answer, and neither is toughing it out.

Back off enough to be pain-free, lift your cadence slightly, train your hips properly for a couple of months, and build back gradually. That is a boring plan, and it is the one that works.