I'll go along with Briman on this one, I used to start at the low end of the published data and work up to the max load in 1/2 gr increments, then the load that shot the best would go at 1/10'th gr increments in both directions. Now I start with an accuracy load from Nosler or Sierra and go 1/4 gr increments in both directions till the sweet spot is found. The thing about accuracy loads is that it is not always what produced the best accuracy but the most consistant pressures.