The paper that made stretched length a proxy

Stretched flaccid length is taken in clinics because of one specific finding, not because it is a convenient number to collect. Wessells, Lue and McAninch, published in the Journal of Urology in 1996, measured a group of men in three states in the same session - flaccid, stretched flaccid, and erect - and reported that the stretched reading tracked the erect one closely. That single result is the reason stretched flaccid length appears throughout the later literature at all.

Why three states in one session mattered

Before this kind of study, a clinician wanting an erect measurement had a genuine practical problem: obtaining and measuring a full erection in a clinical setting is not a simple thing to arrange as a matter of routine. Flaccid length was easy to take and available in bulk, but flaccid length is known to vary with temperature, arousal and recent activity, and it correlates only weakly with the erect state clinicians actually cared about.

Measuring the same men in all three states let researchers ask directly which of the easier states, if any, stood in reasonably for the harder one. Stretched flaccid length - pulling the flaccid shaft to a resistance point and measuring along the top - turned out to be the better proxy of the two easier options.

What the study found, directionally

The reported finding was that stretched length sat close to erect length across the sample, closer than flaccid length did. This is a study about a measurement relationship, not a claim about any individual man's numbers matching exactly - a proxy that tracks well on average still leaves room between two given readings on any one person.

The sample was small relative to the pooled meta-analyses that came later, which is normal for a study designed to test a method rather than to establish a population mean. A smaller study answering a specific methodological question is doing different work than a large pooled estimate, and both kinds are needed - the pooled Veale figures could not have chosen stretched length as a proxy on their own; that decision came from studies like this one first.

Why the finding stuck

A result that lets a clinician skip arranging a full erection and instead take a two-minute stretched reading is operationally valuable, and operational value is a large part of why a finding propagates into standard practice. Once stretched flaccid length was established as a workable proxy, later and larger studies - including the pool behind Veale 2015 - could report it alongside or instead of erect length, widening the erect-adjacent data available without requiring every study to arrange full erections.

That single choice, made on the evidence of a study like this, still shapes what a clinical size reading looks like today: pull to the resistance point, read along the top, and know the number approximates erect rather than measures it directly.

A small study doing a specific job

It is worth naming what kind of evidence this is, because it is a different kind from a large pooled estimate and both matter for different reasons. A study built to test whether one measurement predicts another needs enough participants to see the relationship clearly, not the thousands required to pin down a population mean to a decimal place. Judged against that goal, a modestly sized sample measured carefully in one session is strong evidence, and treating it as weak because it is smaller than a later meta-analysis misunderstands what it was trying to show in the first place.

What this post is not

The relationship between stretched and erect length is not perfect, and how well it holds up varies somewhat between later studies - that variation is worth its own read rather than folded in here. Nor is this post about how to take the stretched measurement yourself; the method sits in its own place.

It is worth being clear about what a finding like this is not, too. It is a statement about the relationship between two lengths, nothing about how either number should be judged. A judgement is a different kind of output entirely - the sort a service like Rate Cock, a scoring hub's public results, or a human reviewer produces - and no image analysis of any kind enters into a stretched-length reading, which is taken with a ruler and a hand, not a camera.