Can a Player Influence his Teammates' Shooting Percentage?

I actually had one player, in particular, in mind - Mario Lemieux. I was talking to Eric T over at Broad Street Hockey today about playmakers driving their teammates shooting percentages and it occurred to me that if anyone could create better chances for his teammates, it would be Lemieux. Gretzky, too, obviously, but we don't have game logs from his prime yet, and Lemieux's injury record makes it easier to see what happened when he wasn't playing.

First, let's look at Pittsburgh's record with and without Lemieux from 1989-90 to 1993-94, a stretch when he played between 22 and 64 games per season:

In case you were wondering, that's absolutely astounding. Lemieux - even when he was constantly injured - was worth roughly eight wins per season. Today, a comparable player (who could stay in the lineup) would be worth Albert Pujols money. But back to the question at hand - did Lemieux influence his teammates' performance?

Here are the top-level regular season numbers:

We see a 55-goal boost per 80 games (in case you were wondering, Pittsburgh allowed the same number of goals whether Lemieux was playing or not) which was worth something like seven wins in this era. And, perhaps more interestingly, we see that Lemieux's teammates were eight percent more likely to score on a given shot when he was playing, regardless of whether they were on the ice with him or not. You might think that Pittsburgh got more power-play opportunities when Lemieux was playing, but roughly 27% of his teammates' goals were on the power-play in either case.

Now we're looking at a broader impact here - Lemieux driving the shooting percentage of players he likely never played with. What did he do for his actual linemates? We don't know exactly who he played with, but we can take a pretty good guess - they're the guys with the event totals (goals for and against, both PP and EV) that are similar to his. Here's our educated guess:

1989-90: John Cullen, Kevin Stevens, Zarley Zalapski, Paul Coffey

1990-91: Mark Recchi, Cullen, Coffey, Larry Murphy

1991-92: Recchi, Stevens, Coffey, Murphy

1992-93: Ron Francis, Rick Tocchet, Stevens, Murphy

1993-94: Francis, Stevens, Jagorim Jarg, Murphy

Seems good to me. And how did they do with and without Lemieux?

The boost isn't substantially different here than it was for the team as a whole, although it's fascinating that their scoring rates (points totals per 80 games) are the same regardless of whether Lemieux was playing with them or not. I'd posit that Lemieux's playmaking contribution is about as large as we're going to consistently find - something on the order of 7-8% - and we can use it to bound the impact that a player can truly have on the quality of his teammates' scoring chances.