Team wins don't really matter much to me when I'm evaluating trends in a hitter's results. If we want to extrapolate that to team success that's another matter. That five month gap can mean a lot or mean nothing and for many (most?) players it generally means nothing. I could just as easily argue that a logical cut-off of evaluation could be between adjustments by Dozier. We make similar cut offs when we judge young players (like we are with Buxton). The truth is, that five month gap is arbitrary as a cut-off for player production. We use it for convenience sake, not because it has any special meaning. (Unless there is a specific reason to believe it does, like injury rehab) We also make many cases for player evaluation based on cutting off statistical analysis at various points. We talk all the time about things like "Since June 1st Brian Dozier has slashed...." or "Since being moved to the top of the order" or "Pre-allstar". All of these are arbitrary. It doesn't invalidate them. Short of taking the whole players career into focus, we're always relying on arbitrary time frames. That isn't a valid argument against a statistical study, especially if it takes a large, time consistent sample size.