One of the emerging area of hockey statistical analysis is a focus on shot quality. Some goaltenders seen a lot of long distance shots from the blue line while other goaltenders seen a big share of point black chances in the crease. Traditional goalie stats like Save Percentage (SV%) and Goals Against Average (GAA) do not adjust for the differences in shot quality. They assume that all shots are equal and all defenses are equal--an assumption that is obviously false.
A number of researchers have begun to move away from the "treat every shot equally" perspective and have developed Shot Quality adjusted SV% and GAA. Gabriel Desjardins over at the Behind the Net blog has just posted the numbers for last season.
The basic idea of shot quality adjustments is "location, location, location." Shots that are close to the net and especially in the slot area are more dangerous (on average) and shots for the boards and blueline are much less dangerous (on average). Based on WHERE the shots were taken you can create an Expected SV% and an Expected GAA. It is far from perfect, but it is more realistic than assuming all shots are equal.
Last seson the Thrashers allowed a large quantity of shots and they also gave up a ton of high quality shots. How did Lehtonen and Hedberg handle these shots? Well based on the location of where the shots were taken, if Lehtonen were an average NHL goaltender his Expected SV% was just .902 and his actual number was .917, his Expected GAA was 3.29 and his actual GAA was 2.76 which means that Lehtonen allowed roughly half a goal per game less than expected. In other words if you traded Lehtonen and installed him behind a league average defense he probably would turn in GAA half a goal better than the NHL average--if we trade to the Flyers this summer he's going to put up some great GAA numbers for them.
Hedberg: Given the quality of shots he faced Hedberg was expected to have a .898 SV% (actual .900) and an Expected GAA of 2.89 (actual 2.82). Hedberg preforms almost exactly as you would expect given the quality of shots he faces.
I'm not a big fan of Hedberg's performance but these metrics suggest that he is actually close to being an average goalie and that the biggest reason his numbers are so ugly are the location of the shots the skaters allow against him.
Loading comments...