You know, this is one of those things that just begs for an experiment. In areas where one is allowed multiple flies, tie on two flies of the same size and pattern (one on the point, and one on the dropper) but one has a turned down eye and one has a straight eye, and fish for an hour. Then, switch positions, fish another hour, and so on. Do this for a number of days in a row, each day starting with the opposite set up from the day before (ie. if today I started with straight eye on the point, tomorrow I start with down eye on the point) to ensure any differences are due to time of day effects (you must start fishing at the same time each day).

Keep track of which fly was taken for all fish brought to hand; strikes might be harder to tell and so the data would be more suspect, and in the end, getting more hits but of a poorer quality (fish throw it more often) might not really be considered "better".

Then, calculate what proportion of fish were caught on the down eyed fly and what proportion were calculated on the straight eyed fly. If there is no real difference, then you would expect things to be about 50/50 (and there are ways to determine if the observed scores are likely to be different from that expected ratio too!).

If one style does appear better, then repeat the whole excercise with up eyed hooks.

If you are allowed 3 flies, then you could have 3 flies of the same size and pattern, but one up, one down, and one straight eye. Mixing up the positions gets more complicated since there are 6 different configurations, but that just requires more fishing to sort out the issues, with no difference resulting in an equal number of fish caught at all positions.

Now, just because it does, or does not, make a difference with one type of fly (let's say spiders), doesn't mean eye position wouldn't make a difference for, say, dries, or nymphs, or winged wets, or streamers, or ...

That's a lot of experiments to conduct, which requires a lot of fishing to be done.

Who said science isn't fun?

- Jeff