Can someone explain why Hit-by-Pitch rates (when divided by the total number of batters faced) jump starting around the late 90s?  Is there a MLB ruling on shin guards or something that might explain it?  For some reason, by 2000 there is one HBP for every 100 batters faced (1% rate) - double the rate of the past 40 years.

HR rates (per hits or ABs) is a significant predictor but this doesn't explain half the jump observed.  Using the data I have I don't have pitches per plate appearance (the more pitches you see the more likely you get hit), but dividing HBP by BB (which should show equally higher rates if pitches per PA is a factor) shows the same huge jump.

CLARIFICATION: One possible explanation, and one I'm trying to evaluate, is that Moneyball philosophy increasingly rewarded players for HBP giving them greater incentives to get hit.  I want to make sure there's nothing else driving this.