Behindthenet Blog: Goaltender Rebound Percentage Leaders

In his work on Shot Quality, Alan Ryder noted that shooting percentage is extremely high on rebounds. That is, in the two seconds following another [i.e. initial] shot [the average sv% is .694]. The overall save percentages for 2008-09:

Overall   1-2 s   3 s   4 s

919     694 807      892

After four seconds, shooting percentages are in the low-to-mid 900s. So it is significant to look at what percentage of a goalie’s shots are rebounds allowed less than 2, 3 and 4 seconds after a previous shot. Here are the league “leaders”:

Goalie PCT2s PCT3s PCT4s

EVGENI NABOKOV 5.52 6 7.61

MIIKKA KIPRUSOFF 5.33 6.24 8.25

CAM WARD 5.31 6.38 7.09

CHRIS OSGOOD 5.25 5.59 6.48

MATHIEU GARON 5.19 5.73 6.55

MICHAEL LEIGHTON 5.16 6.25 7.06

MARTIN BRODEUR 5.08 5.72 6.67

TOMAS VOKOUN 4.98 6.63 8.2

MANNY FERNANDEZ 4.68 5.81 6.78

PETER BUDAJ 4.59 5.37 6.06

And at the other end:

Goalie PCT2s PCT3s PCT4s

CHRIS MASON 3.25 4.42 5.96

STEVE MASON 3.24 4.43 6.22

KARRI RAMO 3.21 5.22 7.22

ERIK ERSBERG 3.18 4.67 5.09

JONATHAN QUICK 3.08 3.77 5.71

JASON LABARBERA 3 4.6 5.61

RYAN MILLER 2.84 3.84 5.76

BRENT JOHNSON 2.69 3.92 4.9

MARTIN GERBER 2.6 3.9 4.72

JOSH HARDING 1.79 3.58 5.67

via Behindthenet Blog: Goaltender Rebound Percentage Leaders.

I’ll have to think about what this really means. But the first thing that occurs to me is: Jonathan Bernier. Gives up 20-25% fewer rebounds than Quick. Regardless of your “rebound” save percentage, that’s one way to solve the problem. Ersberg, we know, is pretty good at rebound control, too.

One stat I would like to see — maybe the guy who did the original numbers already did this, I will have to look — is a SV% that treats a shot and its rebounds as one event. Then you would see the effect of a goalie’s rebound control. For example, a goalie could suck on rebounds, but if he doesn’t give up any, the number is meaningless. Conversely, a goalie could be great on the original shot, but if the rebound always goes in, that — um — sort of cancels it out.