Tight end is a position that has created some challenges in recent years if forecasting via Similarity Scores is what we want to do. There are fewer tight ends than wide receivers or running backs, and the emergence of guys like Rob Gronkowski and Jimmy Graham has meant that we have multiple tight ends per year that we’re trying to match up with historical seasons from Tony Gonzalez and Antonio Gates.
As usual, the Similarity Scores are going to have the toughest time with extreme outliers like Rob Gronkowski. If you’re forecasting the future based on a similar past, what do you do when nothing in the past looks like the thing you’re trying to forecast. Gronkowski is just in another world in terms of the consistent efficiency that he’s able to display. His targets generate a full fantasy point above average when compared to the average tight end! So the Similarity Scores below might not be that great at forecasting Gronk for 2013. I have some other thoughts as to how we can get at that projection, but that will have to be its own post. Until then, feel free to play around with creating a custom Similarity Score (and projection) by including or throwing out individual games with the app below.