Here is the average non-coincidental penalty differential (penalties drawn minus penalties taken) per 82 games and 15 minutes of icetime for each position:
| 5v5 | 5v4 | 4v5 | |
| All | -1.106 | -0.770 | -1.642 |
| D | -8.835 | 4.341 | -9.488 |
| C | 3.974 | 0.434 | 2.392 |
| LW | 3.580 | -6.160 | 9.916 |
| RW | 4.855 | -5.656 | 11.525 |
| F | 4.132 | -3.250 | 6.233 |
I think you can construct some pretty compelling arguments for why centers show so little variation in their numbers across all man-advantages. But why do right-wingers have much better numbers than left-wingers? Simple answer: Dustin Brown. He is far and away the best in the NHL at drawing penalties, and he drives the numbers to a different result:
| 5v5 | 5v4 | 4v5 | |
| RW | 4.855 | -5.656 | 11.525 |
| w/o Brown | 4.425 | -6.063 | 10.812 |
| LW | 3.580 | -6.160 | 9.916 |
Almost half of the difference between LWs and RWs is Brown's skill. Lest you think drawing penalties is not a skill for forwards, here's the year-to-year regression to the mean and R^2 for forwards with 500+ minutes at 5-on-5:
| Regress | R^2 | |
| Differential | 39% | 11.70% |
| Drawing | 37% | 39.20% |
Taking penalties has a non-skill component to it – that is, some coaches will send players out to intentionally take penalties. Opportunities to do so will vary from season-to-season, and by whether Daniel Carcillo is seen as an agitating physical leader or an undisciplined goon.
