Digital Gee

Got to thinking about cycling and percentiles the other day. I wouldn't even know how to google this one, so I thought I'd ask here. Perhaps someone knows.

What percentile of the adult population would people who bike 500 miles a year be? 1,000 miles? 2,500 miles? 5,000+ miles?

I suspect that as someone who tries to bike about 2,400 miles a year, I'm in the 90th percentile or maybe a bit higher. However, in terms of just people who post on 50+, I'm thinking I'm maybe in the 50th percentile of that group, if that.

All of this doesn't matter a whit, it was just something I was thinking about. You math folks out there -- got any thoughts?

cyclinfool

I know that between myself and the other two guys I ride with only one of us (it isn't me) can get that many miles/yr on the bike. I don't actually keep track as this is not a goal of mine but if I do the math based on about what an average week is for me during different parts of the season I suspect I am lucky to get 2000. Good for you to be able to get that much time in. For me, November through March is just about a bust. All I can do is the trainer and there I count hours, not miles.

stapfam

Glad you saw sense and saw the light. Leave it to those that WILL worry about stupid things like this and stop reminding me that I do not understand the way the Americans think. If that is what you are doing.

jppe

Let me guess......you've recently been talking with your daughter about her ACT or SAT scores.........

DnvrFox

Here is a bicycling graph with some extra percentiles - perhaps it will help in some way, although I haven't a clue how!

speedlever

Hmm. I have to wonder what the other 45% are if 34% are male and 21% are female.

cyclinfool

We'll from this we do know that as of 2002 that only 27.3% of the population would even come close to the miles listed. Now lets make a few assumptions here. From this web site: https://www.abs.gov.au/Ausstats/abs@....9!OpenDocument
We can get a few more stats. This is for Australia mind you but lets assume the relative percentages are the same for the US. For the cycling population (the 27.3%)
8.3% ride Daily
29.1% ride between daily and once/week
7.9% every 14 days
14.5% once/month
40.2% less than once per month.

So now if we assume that the riders who are more frequent riders ride about 35 miles per ride as suggested by this article
https://adirondackresearch.com/projec...e/hsurvey.html (I think this is high)

Combining this data suggests that 1% of the population exceeds 5000 miles/year, 4% between 1000 and 5000 miles/year and 1% betweem 300 and 1000 miles/year with the remaining 21.3% less than 300 miles/year.

Now these are rough estimates based on data from three very different sources which may not relate to each other - in other words, it's total BS, or should I say - statistics at its finest...

Bud Bent

Interesting stuff.

speedlever

So then the next question is: what percentage of cyclists frequent bikeforums.net? And of those who do, wonder what percentage fall into the higher mileage categories?

Dallyup

seems you are mixing up percentiles and percentages, won't go into it here, any dictionary should help

DnvrFox

That's what makes it so fun!

BlazingPedals

I used to administer the recognition program for my local club. In an area with 200,000+ population, there are 1500 active club members, and of those, approximately 80 per year reach the 2,000 mile point. (Specifically, they ride the year in miles, so it would be 2007 miles this year.) Of those, quite a few go 'way past, but the number of 5000+ mile riders are in the single digits.

cyclinfool

Some of the data given was in percent and some in percentile. I merged the data in a somewhat unscientific way and made some estimates as to what the pdf for the percentile data would look like based on the data in the other citations. Once I did that I could really no longer look at percentiles because I did not know what the distribution tails looked like - how far they extended.

Now if you can get hold of the raw data maybe you would like to perform a simple Kolmogorov Smirnov test and see what distribution to use (although I suspect an exponential or log normal would get you close enough). I think I'll head down to the liquor store where I typically find those two gentleman on the shelf and let you know when the hangover wears off...