Almost every conversation about tennis elbow prevention eventually arrives at the same figure: just under 40 percent. That is the share of recreational players who reported a history of tennis elbow in Gruchow and Pelletier's 1979 survey in the American Journal of Sports Medicine — still the most-quoted prevalence number in the field nearly fifty years later. It shows up in clinic handouts, on brace packaging, and in string marketing. It is a real number from a real study. It is also being asked to carry far more weight than it was ever built to hold.
What the number actually measured
Gruchow and Pelletier surveyed several hundred recreational players and asked whether they had ever had tennis elbow. Roughly two in five said yes. The strongest association in the data was exposure: players who played more hours per week, and older players, reported it more often. The authors also asked what preventive measures players had adopted — equipment changes, grip changes, technique work — and did not find that those measures tracked with lower rates of the condition.
Three limits are worth stating plainly, because nobody quoting the number states them.
First, it is self-report. No clinician examined these elbows, and nothing was imaged. "Tennis elbow" in that dataset means a player who thought they had tennis elbow, which is a broader and blurrier category than lateral epicondylar tendinopathy.
Second, it is 1979. Racquets were wood and early aluminium and graphite: heavier, more flexible, smaller-headed, strung with natural gut or nylon. Commercial polyester monofilament — the stiff, spin-friendly string that now dominates club bags — did not arrive in force until the late 1990s. The 40 percent figure describes a population playing with equipment almost nobody uses today.
Third, it is retrospective and cross-sectional. It can tell you how common a complaint is among people who play a lot. It cannot tell you what caused any individual case.
So the number measures prevalence, and it measures it in a specific era. What it does not measure — and this is the part that matters if you are the one whose forearm aches when you pick up a kettle — is anything at all about your string.
How we evaluated
We are a review desk, not a lab and not a clinic. For this piece we read the epidemiology (Gruchow and Pelletier, 1979), the biomechanics work on vibration transfer (Hennig, Rosenbaum and Milani, 1992, Medicine & Science in Sports & Exercise, which measured how much racquet vibration actually reaches the forearm and found it varies substantially with racquet properties), the damper literature (Stroede, Noble and Walker, 1999, Journal of Sports Sciences), and the treatment trials — Bisset and colleagues in the BMJ in 2006, and Tyler and colleagues in 2010 on isolated eccentric wrist extensor loading. Alongside that we compared manufacturers' published string specs and the stiffness rankings compiled by independent testers, and read owner feedback on arm comfort.
We weighted them in that order: peer-reviewed measurement above independent bench testing, independent bench testing above tester consensus, tester consensus above anything a brand says about its own product. Where sources conflict, we say so rather than picking the tidier one.
Where the gear argument actually lives
The mechanism is not mysterious. At impact, the ball has to be decelerated and reversed. Something has to deform to do that. The stringbed deforms far more than the frame does, so the stringbed is the dominant variable. A softer stringbed spreads the collision over a slightly longer period; a stiffer one compresses it into a shorter, higher-peak impulse. More of that spike ends up in the handle, and from there in the wrist extensor tendons that anchor at the lateral epicondyle.
That is why the string category rank order is consistent across every independent stiffness dataset we could find:
| String type | Relative stiffness | The tradeoff |
|---|---|---|
| Natural gut | Lowest | Expensive, weather-sensitive, less bite |
| Multifilament | Low | Frays and dies faster; less spin than poly |
| Synthetic gut (solid core nylon) | Moderate | Cheap and unremarkable in both directions |
| Polyester monofilament | Highest | Best control and spin; stiffest impact |
We are deliberately not quoting exact stiffness figures in pounds per inch. Published values vary with test rig, tension and reference length, and a precise-looking number lifted out of its test conditions is worse than an honest rank order.
Two qualifiers matter. Polyester is not the villain of this story — it is a material tradeoff that happens to sit at the wrong end of one axis, and plenty of players use it for years without incident. And tension is a real lever within a type: the same string at lower tension gives a softer bed. The nastier detail with poly is that it loses tension quickly and goes dynamically stiffer as it dies, so a set left in for three months is not the string you bought.
Where the evidence is thinner than the marketing: vibration dampers. Stroede and colleagues found dampers reduced string vibration but did not meaningfully change handle vibration or post-impact discomfort. They change the sound. Treat the rest as unproven.
What we would actually do first
Honest ranking, and it is not the one that sells strings. Exposure is the variable with the strongest empirical support — it is the one Gruchow and Pelletier actually found. Progressive loading of the wrist extensors is second; Tyler's trial was small, but eccentric loading has held up across tendinopathy research generally. Gear is third. It is genuinely third — and it is also the cheapest, fastest and most reversible of the three, which is why it is worth doing rather than worth arguing about.
Who this is for, and who it isn't
If you play three or more times a week, string with poly, restring once or twice a year, and have a dull ache that flares on backhands and grips, a full-multifilament or gut-hybrid setup at a few pounds lower tension is the change we would make. It costs a restring.
If your elbow pain arrived without a change in play, wakes you at night, or comes with numbness or grip failure, this is not a string problem and no product in this category will fix it. See a clinician. Equally, if you play twice a month and your arm is fine, there is no evidence you need to change anything.
Clear verdict: your string cannot give you tennis elbow on its own, and it cannot cure it — but it is the only variable on this list you can change tonight for the price of a restring.
Evidence grade — Moderate for the claim that stringbed and frame stiffness modulate the load reaching the forearm. Weak for the claim that any specific string change measurably reduces injury incidence. Nobody has run that trial.
The question nobody has answered
Here is what still bothers us. Stroede's damper finding suggests players' perception of a harsh impact can be decoupled from what is actually transmitted to the arm. If that decoupling extends to strings — if the setups that feel soft are not reliably the ones delivering less load — then a great deal of arm-friendly gear advice, ours included, is being validated by the wrong signal. Is lateral epicondylitis primarily a dose problem, where impulse shape is a rounding error against total hours? Or does peak impulse matter enough that the string genuinely moves the needle? The study that would settle it — matched players, matched volume, randomised to string type, tracked across a season — has not been done, and until it is, anyone who tells you they know is telling you what they believe.