Officials
NHL referees: penalties and power plays by referee
Which referees call more, and which let them play? Every NHL referee since 2017-18, measured against what the two teams on the ice would normally produce, so a referee who keeps drawing penalty-prone teams does not look whistle-happy.
Tonight's crews are on Power Play Tonight once the NHL posts them.
How much do referees matter?
The full test (one season against the next) runs once every season's crews are loaded. On the 1,747 games loaded so far, true referee differences look small (about ±0.22 power plays per game for both teams combined) and only weakly repeatable (odd vs even games: r = 0.18 across 39 referees). Until it passes, the effects are descriptive only and are not used in the power-play forecast.
Active referees
Effect = power-play opportunities per game (both teams) above or below what the teams were expected to get, shrunk for sample size. Coloured when the 90% range excludes zero. Regular season, 30+ games in the last four seasons.
| Referee | Games | This season | PP opps/gm | Expected | Effect (90%) | Penalties/gm | Home PP% |
|---|---|---|---|---|---|---|---|
| Michael Markovic#31 | 81 | 2 | 6.54 | 5.99 | +0.350.07 to 0.63 | 7.90 | 53% |
| Jake Brenk#26 | 102 | 3 | 6.46 | 6.23 | +0.300.02 to 0.58 | 7.99 | 54% |
| Cody Beach#12 | 82 | 1 | 6.44 | 6.01 | +0.300.02 to 0.58 | 7.94 | 54% |
| Chris Rooney#5 | 66 | 3 | 6.62 | 6.34 | +0.20-0.11 to 0.52 | 8.06 | 52% |
| Francis Charron#6 | 102 | 2 | 6.64 | 6.22 | +0.19-0.09 to 0.47 | 8.20 | 50% |
| Gord Dwyer#19 | 95 | 1 | 6.61 | 6.30 | +0.17-0.11 to 0.46 | 7.86 | 53% |
| Stephen Hiff#33 | 48 | 2 | 6.25 | 5.95 | +0.17-0.14 to 0.48 | 7.54 | 52% |
| Jordan Samuels Thomas#37 | 76 | 1 | 6.21 | 6.05 | +0.16-0.12 to 0.45 | 8.37 | 51% |
| Kyle Rehman#10 | 102 | 2 | 6.36 | 6.21 | +0.16-0.12 to 0.43 | 8.03 | 53% |
| Chris Schlenker#3 | 87 | 1 | 6.36 | 6.17 | +0.15-0.13 to 0.43 | 7.75 | 54% |
| Garrett Rank#7 | 101 | 1 | 6.24 | 6.31 | +0.13-0.15 to 0.41 | 8.18 | 50% |
| Mitch Dunning#20 | 61 | 0 | 6.03 | 6.01 | +0.09-0.21 to 0.39 | 7.64 | 51% |
| Brian Pochmara#16 | 68 | 0 | 6.18 | 6.33 | +0.07-0.24 to 0.38 | 8.12 | 53% |
| Francois St-Laurent#8 | 89 | 3 | 6.03 | 6.26 | +0.02-0.27 to 0.31 | 7.71 | 53% |
| Brandon Blandina#39 | 82 | 2 | 5.80 | 5.98 | +0.01-0.27 to 0.29 | 7.51 | 55% |
| Mike Sullivan#47 | 40 | 1 | 5.75 | 5.91 | +0.01-0.30 to 0.34 | 7.00 | 53% |
| Kelly Sutherland#11 | 102 | 2 | 6.09 | 6.23 | −0.00-0.28 to 0.28 | 8.33 | 49% |
| Eric Furlatt#27 | 93 | 3 | 5.92 | 6.29 | −0.01-0.29 to 0.28 | 7.63 | 54% |
| Tom Chmielewski#18 | 99 | 1 | 5.97 | 6.24 | −0.03-0.31 to 0.25 | 7.53 | 52% |
| Jean Hebert#15 | 88 | 1 | 6.22 | 6.34 | −0.03-0.32 to 0.26 | 8.08 | 52% |
| Peter MacDougall#38 | 74 | 3 | 5.72 | 6.00 | −0.04-0.33 to 0.25 | 7.35 | 50% |
| Graham Skilliter#24 | 102 | 2 | 5.91 | 6.22 | −0.05-0.33 to 0.23 | 7.28 | 53% |
| Justin Kea#36 | 84 | 2 | 5.74 | 6.06 | −0.05-0.33 to 0.22 | 7.43 | 52% |
| Frederick L'Ecuyer#17 | 104 | 3 | 5.67 | 6.23 | −0.06-0.34 to 0.21 | 7.23 | 52% |
| Thomas John Luxmore#21 | 103 | 2 | 6.10 | 6.25 | −0.07-0.34 to 0.21 | 7.85 | 50% |
| Morgan MacPhee#43 | 39 | 2 | 5.59 | 6.03 | −0.07-0.39 to 0.25 | 6.72 | 55% |
| Corey Syvret#23 | 80 | 1 | 5.69 | 6.06 | −0.07-0.35 to 0.21 | 6.83 | 51% |
| Pierre Lambert#25 | 88 | 2 | 5.65 | 6.09 | −0.07-0.35 to 0.20 | 7.04 | 51% |
| Graedy Hamilton#41 | 32 | 1 | 5.44 | 6.00 | −0.09-0.42 to 0.24 | 6.47 | 49% |
| Brandon Schrader#34 | 80 | 1 | 5.63 | 6.09 | −0.12-0.40 to 0.17 | 6.92 | 54% |
| Kendrick Nicholson#30 | 99 | 1 | 5.82 | 6.19 | −0.12-0.40 to 0.16 | 7.15 | 52% |
| Carter Sandlak#29 | 81 | 3 | 5.52 | 6.01 | −0.13-0.41 to 0.15 | 6.48 | 53% |
| Wes McCauley#4 | 105 | 3 | 5.50 | 6.24 | −0.15-0.43 to 0.12 | 6.96 | 52% |
| Beau Halkidis#22 | 44 | 1 | 5.39 | 6.08 | −0.16-0.47 to 0.16 | 7.29 | 55% |
| Trevor Hanson#14 | 103 | 3 | 5.93 | 6.26 | −0.18-0.46 to 0.10 | 7.93 | 55% |
| Chris Lee#28 | 91 | 0 | 5.62 | 6.19 | −0.18-0.47 to 0.10 | 7.12 | 55% |
| Furman South#13 | 86 | 3 | 5.44 | 6.05 | −0.19-0.47 to 0.08 | 6.76 | 50% |
| Liam Maaskant#45 | 30 | 1 | 5.03 | 6.06 | −0.20-0.53 to 0.14 | 6.90 | 53% |
| Jon McIsaac#2 | 100 | 2 | 5.80 | 6.21 | −0.20-0.48 to 0.08 | 6.84 | 54% |
Other referees since 2017-18
| Referee | Games | This season | PP opps/gm | Expected | Effect (90%) | Penalties/gm | Home PP% |
|---|---|---|---|---|---|---|---|
| Dan O'Rourke#9 | 47 | 0 | 5.72 | 6.59 | −0.22-0.56 to 0.11 | 6.96 | 48% |
| Ghislain Hebert#22 | 25 | 0 | 7.04 | 7.13 | −0.01-0.38 to 0.36 | 8.76 | 53% |
| Alex Lepkowski#44 | 23 | 0 | 4.83 | 6.01 | −0.19-0.53 to 0.15 | 6.96 | 45% |
| Dean Morton#36 | 22 | 0 | 7.54 | 7.15 | not estimated | 9.18 | 53% |
| Steve Kozari#40 | 22 | 0 | 6.77 | 7.09 | −0.03-0.41 to 0.34 | 8.73 | 54% |
| Dan O'Halloran#13 | 20 | 0 | 6.85 | 7.27 | not estimated | 8.25 | 56% |
| Justin St. Pierre#12 | 20 | 0 | 6.00 | 7.19 | not estimated | 6.65 | 55% |
| Tim Peel#20 | 20 | 0 | 7.20 | 7.16 | not estimated | 8.20 | 51% |
| Brad Meier#34 | 19 | 0 | 5.89 | 7.19 | not estimated | 6.95 | 55% |
| Ian Walsh#29 | 19 | 0 | 6.95 | 7.25 | not estimated | 8.37 | 52% |
| Marc Joannette#25 | 19 | 0 | 7.05 | 7.18 | not estimated | 8.32 | 52% |
| Brad Watson#23 | 18 | 0 | 5.56 | 7.32 | not estimated | 6.83 | 50% |
| Kevin Pollock#33 | 18 | 0 | 6.72 | 7.05 | not estimated | 8.94 | 48% |
| Dave Jackson#8 | 17 | 0 | 5.94 | 7.10 | not estimated | 8.59 | 47% |
| Tom Kowal#32 | 17 | 0 | 5.59 | 7.26 | not estimated | 6.06 | 52% |
| Riley Brace#49 | 16 | 0 | 6.38 | 6.13 | +0.07-0.29 to 0.42 | 8.44 | 48% |
How we measure referees
- Every NHL game since 2017-18 lists its two referees and two linesmen. We store the crew for each game and the power-play opportunities and penalties each team got.
- For each game we compute how many power plays the two teams were expected to get before the game, from their season-to-date penalty-drawing and penalty-taking rates (the same model as the PP forecast). The referee is credited with the difference: more or fewer power plays than those two teams would normally produce.
- A referee's effect is that average difference, shrunk toward zero by an amount the data choose: with dozens of partners, partner effects average out, and the shrinkage stops a 40-game sample from crowning anyone. The interval is a 90% range.
- Before any of this goes into a forecast we test whether it repeats: odd games vs even games, one season vs the next, and whether last season's effects improve predictions for this season's games.
Data: NHL game-day officials and NHL team power-play and penalty reports, regular season since 2017-18.
FAQ
Which NHL referees call the most penalties?
The table ranks every referee by how many more or fewer power plays their games produced than the two teams would normally get, shrunk for sample size, alongside raw power plays and penalties per game. A referee at the top has called more than expected over a large sample; the interval shows how sure that is.
Do referees really change the number of power plays?
Possibly a little; it is not proven yet. Most of the game-to-game swing in power plays is the teams and plain chance. On the games loaded so far referee differences look small (about ±0.22 power plays a game for both teams) and only weakly repeatable, so read the effect column as a description of past games, not a prediction.
Who is refereeing tonight?
The NHL posts crews close to puck drop. When they are posted, the Power Play Tonight page (/fantasy-hockey/power-play-tonight) lists both referees for every game, with each referee's effect.
Why does a referee have no effect listed?
Referees with few games are shrunk almost all the way to zero, and a referee who only worked as a linesman has no referee games. Effects are estimated on the last four seasons of regular-season games.