|
After the 2007-08 season, the NHL changed a rule in order to increase offense on the power play. In the past, when a penalty was called, the first power play faceoff was taken wherever the puck was touched. Starting with the 2008-09 season, faceoffs were now going to be in the penalized team’s defensive zone. This had the desired impact – teams no longer had to skate the length of the ice when they had the man advantage, so they had the opportunity to spend a little bit more time on their actual power play. The overall impact was a nine percent increase in power play scoring.
Fast-forward to the first round of the playoffs this season and you'll notice that power play scoring is down nine percent. How is this possible? With over 500 shots taken on the power play, any reasonable estimate would show that there is no likelihood this happened by chance. Indeed, there has been no reduction in overall offense:
Per 60 Minutes PP 2007-08 2008-09
Saves 47.9 49.2
Goals 6.34 5.74
Blocked Shots 24.3 23.7
Missed Net 17.9 21.2
Total Shots at Net 96.5 99.8
The total number of shots directed on goal on the power play during the first round has gone up more than three percent over the last season, while the number of shots reached the net remains unchanged. Additionally, the average shooting distance, which is directly correlated with shooting percentage over a large number of shots, has dropped from 33 feet to 32 feet, implying that we should be seeing more goals, not less.
However, this remains a rather coarse estimate of the quality of opportunities that teams have had so far. Using historical real-time shot location data, we can create an expected shooting percentage matrix for any spot on the ice on the power play, then determine how many goals would have been scored, on average, on the shots taken so far in the postseason. This method also suggests that we should see more goals, not less:
PP Shots 2007-08 2008-09
Avg Shot Distance 33.2 32.1
Expected Shooting Pct 12.2 12.7
Actual Shooting Pct 11.7 10.4
# of goals less to expectation 4.0 11.0
So what accounts for the much larger gap in shooting percentage this year compared to last year? Simply using x-coordinates and y-coordinates for shot location is an imperfect measure of the quality of an opportunity, not accounting for screens or defended shots and subject to the eyes of official scorers. If we had a larger power play data set, we could also split this data by shot type, which could explain part of the difference.
However, the largest single factor in reducing scoring rate differences between this year and the previous season is indeed tangible: improved goaltending. Using each goalie’s expected 4-on-5 save percentage above-average (EGAA) based on 2007-08 and 2008-09 performance, we can estimate the impact a player had on overall scoring. First, let's take a look at the goalies who played in the first round both seasons:
Goaltender EGAA/60 TOI 07-08 TOI 08-09 EGAA 07-08 EGAA 08-09
PRICE -0.75 102.6 41.0 -1.28 -0.51
OSGOOD -0.13 28.7 50.3 -0.06 -0.11
NABOKOV -0.21 78.2 73.0 -0.28 -0.26
LUNDQVIST -1.84 58.9 89.9 -1.80 -2.75
THEODORE -0.16 84.0 19.2 -0.22 -0.05
FLEURY -0.69 51.0 81.8 -0.59 -0.94
BIRON -0.24 106.1 82.3 -0.42 -0.33
BRODEUR -1.01 58.2 93.7 -0.98 -1.58
KIPRUSOFF 0.19 67.6 57.2 0.21 0.18
THOMAS -1.76 102.5 45.5 -3.00 -1.33
TOTAL -8.44 -7.70
So a little over half of the total 4-on-5 minutes are played by goaltenders who made the playoffs both seasons. The big difference is among the goalies who are in this year playoff's, but weren’t in the postseason last year:
GOALIE EGAA TOI 07-08 EGAA 07-08 GOALIE EGAA TOI 08-09 EGAA 08-09
HUET -0.99 106.1 -1.74 VARLAMOV -2.02 84.2 -2.84
GIGUERE -0.93 87.8 -1.37 HILLER -1.60 73.7 -1.96
HASEK -0.88 43.4 -0.64 KHABIBULIN -1.30 65.0 -1.41
ELLIS -0.77 70.3 -0.91 LUONGO -1.17 61.5 -1.20
GERBER -0.29 51.3 -0.25 HALAK -1.01 4.5 -0.08
BACKSTROM -0.11 79.8 -0.14 S. MASON -0.85 50.3 -0.71
HARDING 0.16 4.6 0.01 VALIQUETTE -0.67 13.4 -0.15
TURCO 0.19 89.4 0.28 WARD -0.56 93.7 -0.87
JOSEPH 1.04 10.6 0.18 C. MASON -0.32 61.5 -0.33
MCELHINNEY 1.46 7.8 0.19
TOTAL -4.57 -9.55
This year’s crop of goaltenders is substantially better than last year’s, with Nikolai Khabibulin, Roberto Luongo, Jonas Hiller and Simeon Varlamov all expected to stop more shots than the best of the pack in 2007-08. The overall difference between the two goaltending crews is 4.25 goals, which is a bit over one-third of the 11 goals that NHL shooters have fallen below expectation on the power play in the first round of the 2008-09 playoffs. | 
0 comments have been left for this article. 
Conn Smythe Watch: Lundqvist Leads