I have read some writers that encourage you not to breathe on the set!!! That’s crazy, and for years people believed that stuff.

They still do. Look at all the sent elimination products and clothing that gets sold to deer hunters. Its obvious not many deer hunters are hounds men.

Mostly because some of that stuff was either BS or intentional disinformation.

A lot of the old methods about controlling sent, baits, lures, etc. I really think boiled down to an inability to actually statistically test their effectiveness. Just think about the process of testing a new lure. If you are testing it by placing it in a few locations with trail cameras, your first few nights might actually be random selection, but after that you have site fidelity that is going to make the encounter non-random, rendering it a biased sample. Once biased, you no longer have access to your parametric tests, leaving you with non-parametric test that are not nearly as powerful in detection of a type 1 or type 2 error. Plus, this is assuming you have lets say 20 cameras out to get a reasonable sample size.

Back in the '70s and '80s you would just have no way of knowing if something you did slightly different actually caught that animal the next day, or if it would have been there either way. And if there was not an animal in the trap the next day, you assumed something was wrong with the set, without the ability to KNOW (like with a trail camera) that there was actually no activity at the site.

There are just so many different variables that can be taken into account and tested, that really a Linear Regression might be you only good way of testing some of them, and at that point as a trapper, this has gone from a enjoyable hobby to a almost overly complicated and frustrating chore. And you really have no way to account for the possibility of a trap shy animal or other compounding condition complicating the testing further.