Say you have three forwards with the following stat lines at five-on-five:
Which of the three has been the best so far? Who has been the worst? Who is likely to the most productive going forward?
In terms of point production, Player C has a clear edge – his six goals in 191 minutes are more than Player A and Player B have in their 682 combined minutes. Of course, if we look at expected goals, Payer C has the lowest rate of the three. He’s also had the most favorable zone starts (by a decent amount over Player A and by a huge amount overy Player B).
Player C’s on-ice goals-for percentage is great (63.6%), but definitely aided by an unsustainably high on-ice shooting percentage (16.3) and strong save percentage (92.2), both of which dwarf Player A’s percentages and, combined into PDO, top Player B’s. But Player C’s underlying shot metrics are in the red across the board – shot attempts, shots on goal, scoring chances and high-danger chances (again, despite cushy zone starts). Player A is nearly the exact opposite – strong possession numbers, but poor results. Player B is a bit of a mixed bag, but considering his extreme zone starts, having an expected goals-for percentage above break-even is pretty darn impressive (and his actual goals-for percentage is even better).
So which of the three has been the best so far? Who has been the worst? Who is likely to the most productive going forward?
As you may have guessed, those three are all Capitals, and they are Richard Panik (Player A), Garnet Hathaway (Player B) and Daniel Sprong (Player C). And if and when the Caps ever get fully healthy, Peter Laviolette is going to have to decide which one of them (or someone else) gets scratched when it’s time to play their best lineup. And maybe that varies depending upon opponent, but, for the sake of argument, let’s ask the simple question: who sits? We put the question to Twitter yesterday, and here are the results:
After reviewing the data at the top of the post, is the result surprising? Maybe. But it clearly reflects a couple of realities. First, we often value results over process. To an extent, that’s understandable – no one has ever won a game solely by outshooting its opponent. But unless there’s a good reason that process and results should be wildly divergent, it’s usually safer to bet on process going forward, as sample sizes get larger. Does that mean that Richard Panik has been better than Daniel Sprong this year? Well… yeah, sorta (though Sprong’s finishing talent does serve to explain some of that divergence).
The second point is a related one, one that we might call The Hagelin Paradox. Among the team’s regular forwards, Carl Hagelin is actually second in expected goal rate, thanks to his team-leading high-danger chance rate. But…
Yeah.
So we intuitively know that Hagelin “should” have more goals than he does, we’re disappointed by that, and we all but ignore the fact that he’s getting those chances in the first place. We shouldn’t do that, because those chances are part of the reason that the Caps’ fourth line has been so effective (and vice versa, of course).
In Sprong’s case, we intuitively know that he’s scoring a lot more than we’d reasonably expect, so perhaps we look past the underlying numbers that aren’t as pretty. To be sure, the guys at the top of this list are generally the team’s more skilled players and can generally outpace their expected goals (think of it this way – if expected goals are calculated based on an average player’s abilities, you’d expect guys like Sprong, Vrana, Ovechkin and the like to do a bit better, and guys like the fourth liners to do a bit worse). Of course there’s going to be random variance here and a guy like T.J. Oshie might end up near the bottom, but over time, we’d expect that to regress upwards.
So which of the three has been the best so far? Who has been the worst? Who is likely to the most productive going forward?
Who sits?
It’s not an easy question to answer, and it’s certainly not as simple as looking at goal or penalty totals.