User blog comment:Alemagno12/Quick poll/@comment-35870936-20180823195137

It depends on how accurate I want my results to be.

If I want to be as sure as possible that I'm correct, then I will either do a formal proof or a quite detailed analysis and I will try to avoid all extrapolation unless it is clear that it is ok to do so.

If I want to be reasonably sure of my result, then I will do a moderately detailed analysis (maybe with some extrapolation).

I might do some analyses with major extrapolations (for example, if I noticed that (0,0)(1,1) = psi(W) in UNOCF, and (0,0)(1,1)(2,2) = psi(W_2), I could hypothetically extrapolate all the way to (0,0,0)(1,1,1) = psi(W_w)). However, doing something like that leaves a lot of room for error. If I did that, I wouldn't be very confident in my answer of (0,0,0)(1,1,1) = psi(W_w) unless I did a more detailed analysis past (0,0)(1,1)(2,2). If I found that, for example, (0,0)(1,1)(2,2)(3,3) = psi(W_3), then I would become much more confident in my value for (0,0,0)(1,1,1), because there is a clear pattern that (0,0)(1,1)(2,2)...(n,n) = psi(W_n).