The second part of the reply is to explain why a race wouldn’t actually help in what I’m trying to achieve and why Gerrie, yourself and others were so far off-beam in proposing a race as the solution.
In a race, you have the starting assumption that everyone is trying as hard as possible to get as high a position in the rankings as their ability allows. The results show you ranked ability. Great if you are racing, but irrelevant for tuning a social ride.
To do this, you need to know what each rider finds comfortable and non-competitive (in a racing with others sense). In my experience, people rarely like to admit when they are uncomfortable with the pace of a ride and will push themselves harder, to try and stay with the group - it’s a well-recognised human trait. Asking them if they are comfortable isn’t going to yield any useful information, but data of the type I have been suggesting gives you those answers. The exam questions I am trying to answer with that data are really straightforward:
If you are hoping to have a social ride at a single power output that would be relaxing for all riders, what would that power output be - the simplest answer would be the power output of the slowest rider.
A more nuanced answer would be the power output that offers a reasonable ride for the most people - perhaps the mean output of the most and least powerful riders, maybe the median or something else.
Alternatively, analysis might suggest that within the ride, you might need 2 or 3 different target power outputs to maximise everyone’s chances of having a reasonably relaxed ride.
You can guess at all of the above, as is done now, but making use of the data being collected is a much better way of doing things. Get it wrong and those who struggle and find it anything but a social ride, may not come back. Those who go off the front and race - who cares; they are deluding only themselves.