Two words, one habit
In the shop, both words get used for the same complaint: the string is longer than it should be. The peep is rotated. Timing has shifted. Something moved, and the archer wants it not to.
But the mechanism that lengthens a well-built 452X string over the first hundred shots is not the mechanism that lengthens the same string after five thousand shots. Nor is either of those the mechanism responsible for the sudden 3/8" length change an unstabilized string will show after its first competition. Three different processes, three different timescales, three different remedies. Calling them all "stretch" is how bad tuning habits get preserved across generations of archers.
Stretch — the geometric mechanism
Stretch is what happens in the first few hundred shots of an unstabilized bundle. It is geometric. The strands are re-arranging themselves inside the bundle, air voids are collapsing, wax is being redistributed by shot cycles rather than by the bench, and the whole bundle is finding a diameter and helix angle it should have found under 200 lb of load at 115°F. It looks like the string got longer. It did get longer. But not because any fiber elongated.
The mechanism is bundle contraction perpendicular to the length. As the strands press against each other more tightly under repeated shot loads, the bundle diameter shrinks by a small amount. Because the total helical path length of each strand is conserved — the wax is not letting them slide — a smaller diameter means a shallower helix angle, and a shallower helix angle means a longer bundle for the same amount of strand. The math is exact. Halve the bundle diameter and, at a fixed twist rate, the length of the axial column that the same strand traces grows by a predictable percentage.
Stretch is fast. It happens in shots, not weeks. It is bounded — once the bundle has reached its equilibrium diameter, further shots do not extend it further. And it is preventable: it is exactly what the 200 lb tensioning-and-burnishing dwell is designed to consume at the bench.
Creep — the polymer mechanism
Creep is a fundamentally different process, and it lives inside the polymer chains themselves. UHMWPE — Dyneema — is a linear polymer. Under sustained tensile load, its long molecular chains slowly slide past each other along their axes and re-orient more perfectly parallel to the applied load. That sliding and re-orientation is irreversible on human timescales. The fiber, at the molecular level, is very slowly re-organizing to be a slightly better fiber in the direction it is being pulled.
Every high-performance polyethylene fiber does this. It is not a defect. It is a defining property of the material class. Dyneema creeps. Spectra creeps. Every UHMWPE fiber that exists creeps. The relevant number is not whether, but how much per unit time under what load at what temperature, and for engineered fibers those numbers are published and consistent.
At room temperature and moderate load, UHMWPE creep rate is small but nonzero. For a bowstring in service, "moderate load" means the string is at brace tension between shots — a resting load of somewhere in the tens of pounds — and briefly at full-draw tension during shots. Both loads produce creep. The rate is highest at the highest sustained load, which for a compound bow means the archer's holding time, not the shot itself.
Vectran — the LCP fiber blended into 452X — creeps at approximately one-tenth the rate of UHMWPE under the same conditions. This is why the blend exists. Pure Dyneema creeps too much for stable service in a target bow. Pure Vectran does not creep enough but fatigues under shock loads. The blend gets a creep rate that lives between them, and closer to Vectran's than to Dyneema's.
The timescales
| Mechanism | Timescale | Bounded? | Where it lives |
|---|---|---|---|
| Stretch (bundle geometry) | Shots, not weeks | Yes — hits equilibrium | Between strands |
| Creep (polymer chains) | Weeks to years | No — never fully stops | Inside the fiber |
| Fatigue (fiber damage) | Tens of thousands of shots | No — leads to failure | At weak filaments |
| Environmental (wax redistribution) | Days to weeks | Reversible | Interstitial voids |
Four different processes, all of which can present to the archer as "the string got longer." Only the first two are what most people mean by "stretch." All four are worth being able to distinguish, because the correction for each is different.
The wrong fix
A string that has stretched (mechanism one) does not need more twist. Adding twist to correct the length introduces additional axial-to-radial force conversion, which locally raises the internal pressure on the strands, which changes the bundle geometry, which is precisely the thing that just finished changing on its own. The archer gets a length that reads correctly for a week and then drifts again. Twist correction is treating a symptom while making the underlying condition worse.
The correct fix for stretch is a fresh 300 lb hold — the same final-stretch operation used at the bench during string building. It re-imposes the equilibrium geometry that the bundle wants and lets the strands settle back into their finished positions. On the bench that hold happens under controlled conditions. On the bow it can also happen, if the archer is willing to unstring the bow, place the string back on a tensioning fixture, and re-apply the 300 lb load — but at that point the string is being finished, not tuned.
Creep does need a different fix
Creep, by contrast, cannot be prevented by any bench operation. It is a property of the material. What the bench can do is minimize the load per filament — by choosing an adequate strand count for the peak load the string will see — and choose a material whose blend ratio keeps the aggregate creep rate low. Both of those choices are made before the string is built.
Once creep has occurred in service, the correction is not a re-stretch. Fibers that have crept have re-organized at the polymer level. Re-stretching does not un-organize them, and repeated attempts to compensate creep by twist adjustments simply pile geometric changes on top of a chemical change until the string's behavior becomes unpredictable.
The correct response to significant creep is a new string. This is why creep rate matters as much as strength in string material selection: a fiber that gains a small percentage of length over its service life will need to be replaced when that gain crosses the tolerance the bow's timing requires. For a well-set-up compound at target, that tolerance is small — a fraction of an inch across the string's total length.
How to tell them apart at the bow
An archer with a stopwatch and a shot count can distinguish stretch from creep in one afternoon:
- Stretch announces itself in the first hundred shots and then stops. A tuned string that goes out of paper tune after twenty shots, then holds tune reliably by shot two hundred, was stretching.
- Creep is invisible in the first hundred shots and visible only in accumulated drift over weeks. A string that held paper tune for a month and slowly begins to require timing adjustments after that is creeping.
- Fatigue looks like creep at first but accelerates. If the drift rate is increasing over time, filament damage is accumulating and the string is heading toward end-of-life.
- Wax redistribution is reversible. If the drift correlates with weather changes and reverses when the string is re-waxed and burnished, the mechanism is thermal migration of wax, not any change in the fibers themselves.
Why this matters
The archery world has, for most of its history, spoken about "stretch" as a single generic process that all strings do. That habit has cost archers thousands of tuning sessions spent applying twist corrections to problems that twist cannot fix. Once the four mechanisms above are separated in the mind, the diagnostic protocol changes: the archer sees a length change, asks which mechanism produced it, and applies the correction that mechanism actually responds to.
A well-built 452X target string, tensioned and burnished properly, stretched at 300 lb before it leaves the bench, should show negligible stretch in service, low creep over its service life, and a fatigue-limited replacement schedule measured in tens of thousands of shots. Any string doing worse than that is showing a bench problem, not a material problem.
Connections
See The helix for why the material blend exists — pure Dyneema creeps, pure Vectran cracks. See Tension and dwell for how the bench spends the stretch budget so the archer does not. See Strand counts for why load per filament, and therefore creep rate, is a design parameter set at build time.
← Tension and dwell · Science & Mechanics
Published 2026-07-04 · Axial Bowstrings