Penis size is, for better or for worse, a universal concern. It is a visible secondary sex characteristic (more visible for some than for others), and it is taken by many as a sign of virility or manhood.

A figure of “five inches” (12.7 cm) is often quoted as an “average” penis size, usually in the context of reassurance. The academic literature on penis size can be cherry-picked to support a wide variety of beliefs regarding the average penis size. The Kinsey study gave an erect penis length of 15.7 cm (\(SD=1.96\), \(n=2770\)) and a girth of 12.3 cm (\(SD=1.80\), \(n=2628\)).1 Other studies have found means as high as 16.4 cm length with 12.6 cm girth in a sample of homosexual men,2 and lengths as low as 12.7 cm in a sample of Turkish men,3 with means reported at a number of intervals in between.4

In what follows, we have examined the data collected by to determine the average penis size (length and girth), to examine the effect of different levels of verification (self-reporting, photo-verification and face-to-face “realtime” comparison). We examine differences in size by language, the relationship between length and girth, and the change in sizes of users on as a function of time.


These data were collected in the context of a competition—a literal internet cock size contest, and so there is an obvious source of bias: Namely, the desire to win.

There are three levels of “verification” reported here: 1) unverified user-submitted self-reported sizes, 2) photo-verification, and 3) among the photo-verified, users could participate in face-to-face “realtime” (RT) comparisons.

For length, users are instructed to measure the shortest path along the top of their erect cock, from the base to the tip. For girth, users are instructed to choose a cross-section that is neither the thickest nor the thinnest, but most representative of their girth as a whole. A user’s language is just the language in which they have chosen to use the site.

To complete a “realtime” comparison, users must meet face-to-face with another user of to compare cock size. This is not a measurement that occurs over webcam. This is an actual in-person meeting, which also must be photo-verified. We will take the sizes of realtime-verified users to be the “gold standard” sample for cock size measurement for three reasons: 1) There is a very high incentive for users to report their own highest measurement. 2) There is a very high incentive for users to force their opponents to admit to as small a size as they can. 3) Meeting in person to compare provides a very high level of accountability with little room for cheating.

For volume calculations, we assume that every user’s cock is a perfect cylinder, and calculate the volume using the formula \(V = \pi r^2 h\). This is of course, incorrect, but it is a useful simplifying assumption.


Sample demographics

Our sample included 15697 users’ data, 772 of whom were photo-verified, and of those, 139 of whom had completed a realtime comparison.

Highest level of verification \(n\)
None (sample total) 15697
Photo-verification 772
Realtime comparison 139

Comparing length and girth across differing levels of verification

Average length and girth

The mean for cock length among photo-verified users of is 15.34 cm, 95% confidence interval (15.17, 15.51). The mean for cock girth among verified users of is 13.29 cm, 95% confidence interval (13.19, 13.38).

We divided our sample of photo-verified users into users who have done at least one realtime comparison (\(n=139\)) and those who have not (\(n=633\)) and performed a two-sample \(t\)-test at \(\alpha = 0.05\) to see if there is a significant difference in the lengths or girths between them. The differences in lengths (15.78 cm vs 15.24) and girths (13.55 cm vs 13.23) are both statistically significant, with \(p\)-values of 0.015 and 0.011, respectively.


The girth histogram appears at first glance to be bimodal. There is a small ``bump’’ around the 5 cm mark. I believe this can be explained by users who sign up and mistakenly enter their girth in inches.