....asked how many people have used the trailer in this way. Chris, the manager,of the store asked me. I told him I had no idea, If anyone who has been on here a long time could help out estimating....
It seems to me this is a dead setup for a random sampling estimate. This means you RANDOMLY select (say N=40) builds (as opposed to builders) and determine which have used a HF frame. Say you find K=16 used an HF frame. Then the estimate of the total number using HF frames is ( 16 M / 40) where M is the total number of builds.
Since we probably don't know M, we do a similiar estimate. For example take another random sample of the subscribers to this forum and calculate the average number of builds per member in that sample (including a lot of zeros because a lot of subscribers never built anything). Then multiply that average by the number of TnTTT subscribers. You would then have to apply a correction for builders who are not members of the forum or qualify your result as the number of frames used by TnTTT subscribers.
All this may appear complex, it is not, but it is a lot of work that requires a lot of attention to detail. It also should have an error analysis which is not so simple.
Wiz
PS. My user name (Wizard Of ODDS) was selected years ago when I taught probability & statistics.