HP Filter v. Simple Moving Average

#1
Evening everyone. I'm trying to better understand the reason behind the HP filter's relative timeliness in picking up trend shifts relative to a simple moving average. Take for example the DJ-UBS commodity index. The HP filter and 200 day moving average of the time series yields, for the most part, an identically shaped trend. The exception being the HP Filter signals a trend shift roughly 100-150 days before the moving average. Is this due to an inherent weighting mechanism in the HP filter, or is the "window" in the HP filter shorter than 200 days? Driving me mad, so any and all guidance is sincerely appreciated!

Thanks in advance.

John