I agree with you absolutely La Mer. I have added barrier draw figures to a couple of successful ratings systems and found in both cases that I have had a small increase in winners but a decrease in POT.
If anyone is interested I found that races with less than 8 runners weren't significantly affected by the barrier draw and for over 8 runners the best factor to use was 0.25Kg per barrier position. This didn't differentiate between different courses or distances and that is where any significant gains from barrier assessment must surely come. Barrier draws are complex beasts before you even take into account how a race will be run but overall as La Mer indicated over rated by the punter.

KV