Dave,
You are right, 1-3 bass per inch group is woefully inadequate to obtain statistically defensible length at age information. In fact the new recommendation by the Texas Parks and Wildlife to its fisheries biologists is 10 fish per centimeter group, or about 25 fish per inch. That's a lot of fish to sacrifice to get the appropriate confidence interval, and way too much for most private lakes. That doesn't make the data any less useful to the pondowner, particularly if the data show extremes one way or the other. If your 5 year old bass are 12" long and you live in Texas, something is wrong. This is a common scenario in my experience.
However, with relative weights, I do think the private pondowner can get statistically meaningful results from an unbiased sampling gear, such as electrofishing. Even angling would be better than nothing, although you are correct that it is not an unbiased sampling gear. Also, as pointed out, Wr's obtained prespawn are not as useful as postspawn or (even better) fall samples. That said, as long as you keep the time of year in mind when looking at the data, they will all work just fine.
It's important to note, in reference to your idea on control groups, that relative weights are derived by comparing your sample to a standard weight table. The standard weight table wasn't created in a vacuum. It was developed by averaging the weights of thousands and thousands of fish from hundreds of collections from all over the United States for each length increment, thus creating a standard weight for that length increment. So in effect, one does have a control group.
There's nothing wrong with trend data, particularly if it is backed up with an occasional hard data point.
In my opinion, electrofishing twice a year is way overkill. I recommend no more than once per year, and once every two years is just fine. In fact, many people get an initial evaluation, then continue on their own.
