The Bell curve of cylists - how fast are the average TR users?

Because the "A"s make me really suffer. An hour at only 300W with the "B"s is enough of a workout for most days. :laughing:

1 Like

It really depends on what population you are considering for your distribution. All humans? All humans who own a bike? All humans who own a bike and ride at least X hours a week? All humans who own a bike and have Strava? All humans who own a bike and have entered a competitive event on their bike? All humans who own a bike and have a racing license?

You can use Coggan’s chart to see where you fall in these curves - but it varies greatly depending on who you are considering.

Within the trainer road chart will naturally contain a higher W/kg than folks who aren’t attempting to follow any training plan - but it really depends who you are comparing yourself too and who you want to populate your N.

FWIW I’m currently around 4.8 w/kg


I think we can safely assume there will be some absolute monsters in the TR survey!


This chart is interesting and is a good example of why we can take the accuracy of self-reports with a grain of salt.

Those big spikes are at the 200, 250, and 300 watts levels where people have picked a nice round number. I have no doubt the general trend is represented but some parts should be viewed with scepticism.


Also add into the grain of salt people like me who use an Elite turbo muin (or similar) that over reads power by ~20%. I love the numbers it gives me, but I know I’m not that fast.

@willball12 @steveking the data of people with power meters is probably the more reflective of how things actually are. Agreed that the round numbers are suspect, the 200W was used by TrainerRoad as a default, I don’t know whether that is the case now? The round numbers are also interesting, and not necessarily arbitary. I would like to have an FTP of 300, but as this is my goal I may subconsciously stop training once I have reached that level. This is discussed in one of @chad 's recommended reads “How Bad Do You Want It” by Matt Fitzgerald, in the context of why the record for one mile (running) stood for so long just above 4 minutes

1 Like

There is also potential for bias in that people with lower numbers may be tempted not to divulge their FTP (or to inflate it) even if it’s anonymous.

The identification of those spikes has also been used to identify voting fraud. A Russian researcher identified a large number of voting areas which reported turnouts of 70%/75%/80%.


Well, I’m in the bell-end at 1.7 W/kg I guess :dizzy_face:


Yes, based on the large spikes around the round numbers and the likelihood that people would over-report their FTP…I’m guessing the true mean and median are probably at least 10 or 15W less than that graph shows.


Shoot I am a weakling compared to most of you guys. I am at 2.89w/kg, ftp of 194 and weigh 148lbs, and I am one of the faster guys out on my local trail. I am on my last week of ssb2lv and am hoping I have raised my ftp to 200 or better. My goal is to be at 3w/kg, once I reach that I will set a new goal.


We’re twins! I have ten more pounds to lose and twenty more watts to gain. I should be looking at 3w/kg soon.

Shall we?

  • 100
  • 120
  • 140
  • 160
  • 180
  • 200
  • 220
  • 240
  • 260
  • 280
  • 300
  • 320
  • 340
  • 360
  • 380
  • 400

0 voters

20w increments paints with a broad brush, but this puts us within the limitations of the poll options :slight_smile:


I don’t test well. My numbers when racing seem okay. Good 1 and 5 min power. I’ve become more patient and somewhat smarter when road racing crits. sometimes i wait too long though. Usually top 5 in cat 4. Hoping to be cat 3 by mid spring. I was definitely one of the stronger cat 4s at Gateway Cup this year. I did a 3/4 race and was fine.


Using W/kg might be a better statistic to “normalize” the data.


I’d say W/kg has its own limitations as well.

1 Like

From something Nate said on todays podcast (185), “[…] almost perfect bell curve with the average at 3W/kg and about 7% of people above 4W/kg”.

Made using this site..


zwift - a lot depends on your weight and the course. i’m high 3’s low 4, I can race b easy. I can race A on a flat course, but not a hilly as i’m 83-85kgs


It has to be single sided in some fashion because everyone is > 0W/kg.


In theory, yes, Erick. [And it is possible to calculate the probabilities with that restriction.] In practice, no: much less than 1 person in 10 000 is predicted to have a negative w/kg according to this distribution. in fact, this distribution predicts that only 1 in 10 000 would have a w/kg less than 0.49.
1 in 100 will have a w/kg of less than 1.41.
1 in 10 will have a w/kg of less than 2.13.
[19.17 per cent have w/kg less than me! Hurray, I thought I was at the tail end.]
At the other end of the scale, about 23 per cent will have a w/kg more than 3.5.


There’s a lot of Zwitchcraft to racing well on Zwift. The guys that are winning races have huge 1 min and 5 min power. They are also really skilled at sitting in the bunch to recover.

1 Like