NEWARK, NJ - JUNE 02: Jonathan Quick #32 of the Los Angeles Kings allows a goal from Ryan Carter #20 of the New Jersey Devils (not pictured) as Stephen Gionta #11 and Colin Fraser #24 look on during Game Two of the 2012 NHL Stanley Cup Final at the Prudential Center on June 2, 2012 in Newark, New Jersey. (Photo by Paul Bereswill/Getty Images)
I've heard a few people say lately that Jonathan Quick can be beaten high (this is true of all goalies, of course - if you get the puck up, shooting percentages are twice as high as if you don't) but this observation just sounded a little fishy to me. You need to watch a goalie face a lot of shots before you can decide if he's any good, and given that only 20% of shots are up high, you'll need to watch way more than you're used to in order to figure out if a goalie can be beaten anywhere in particular.
I took shot locations by net quadrant over the last two seasons and ran our standard even-odd test to determine how much shooting percentage regresses to the mean. I restricted the sample size to goalies who faced 1000 or more shots. Basically, in that time frame, shooting percentage in a given spot (say high stick) regresses by 60-65%.
What does that mean for Jonathan Quick? Here are his save percentages by quadrant, both raw, and regressed by an appropriate amount:
From what we've seen so far, we can conclude that Quick's performance has been a bit below normal high stick side - he has given up nine goals more than average over 352 shots, but that's just not a huge sample size. (For what it's worth, I haven't cleaned the data at all, so any mistakes by the stringer are still there.) This observed performance is less than one standard deviation below the mean, and we literally need to watch him face 1000 shots in order to determine that he's one goal weaker than average high stick side.
I don't think there's evidence much here that anyone can run with confidently. If I offered you even odds that Quick's shooting percentage high stick side would be .830 or better over his next 1000 shots, would you really take the under?