The following is a guest post from Garik, written exclusively for Point Blank on the Isles early season waiver claims. It was originally written prior to Sunday’s game against the Panthers.
In preseason this year, the Islanders picked up three defenders off of the waiver wire. Those guys were Brian Strait, Thomas Hickey and Joe Finley. Strait’s out for the year, but I’d like to talk to you today about the other two guys — Hickey and Finley — who GM Garth Snow took a chance on for this season. Both of these guys have seen a good amount of action this year; Finley has played 16 games and Hickey 23. But is either one a part of the Islander future? Well, neither guy has shown much scoring ability — Hickey has three points, Finley one — so if either of these guys is to be a part of the Islanders going forward, it’s going to have to be because of the things they do as a defender.
It’s more difficult to measure defense than offense for obvious reasons in hockey. There’s no defensive equivalent to goals and assists and the only conventional statistic we have to measure defense is +/-. If you were to simply use +/- to look at Hickey and Finley, you would see Hickey at +4 and Finley at -5 and assume that Hickey was a better defender.
That said, +/- is heavily flawed as a defensive measure. For one, it is way too easily affected by a player’s teammates. Suppose a player has the goalie play amazingly behind him. Then, even if that player was a horrible defender, he’d likely have a good +/- simply because his goalie was standing on his head and not letting in goals (the converse is true as well). In addition, if a player is being a great defender but his forwards simply are failing to score while he’s on the ice, well, his +/- will be negative not because the player is a bad defender, but because his offense can’t score.
We can adjust a little bit for the problem of teammates by using a version of +/- known as “Relative +/-.” Relative +/- is very simple: It compares the +/- of the team while a player is on the ice per 60 minutes to the +/- of the team while the player is on the bench per 60 minutes. Thus, Relative +/- tries to measure the impact a player makes on the team by showing how much better (or worse!) the team is with that player than without.
Turning back to Hickey and Finley:
Hickey’s Relative +/-: +2.18
Finley’s Relative +/-: -1.28
What do these numbers mean? Well, per 60 minutes of ice time, the Islanders’ are 2.18 goals better in goal differential per game when Thomas Hickey is on the ice than when he’s not. That’s an impressive number right there — (two goals per game difference is obviously huge).
By contrast, the Isles are 1.28 goals worse per 60 minutes with Finley on the ice than when he’s taking a breather. That’s horrendous.
Now again, Relative +/- has some of the same problems as ordinary +/-. As mentioned before, Relative +/- can be greatly affected by how the team shoots with a player on the ice and how well the goalie plays; two things which are a good deal out of the player’s control. Perhaps Finley is getting bad luck and Hickey is getting good luck. So we can’t stop here with Relative +/-, we need to look at something else.
So let’s at shots against while each of these player is on the ice and when each is on the bench. We’ll still get some impact of teammates here by looking at shot totals, but we’ll lose the impact of goaltending ability and shooting ability.
So turning to Finley and Hickey:
While Hickey is on the ice, the Isles’ goalies see 20.7 shots on goal per 60.
While Hickey is on the bench, the Isles’ goalies see 26.7 shots on goal per 60.
While Finley is on the ice, the Isles’ goalies see 25.3 shots on goal per 60.
While Finley is on the bench, the Isles’ goalies see 26.8 shots on goal per 60.
So here we see that with Hickey on the ice, the Isles are facing six fewer shots per 60 on goal. That’s a lot! By contrast, while Finley is on the ice, the Isles are basically facing only one fewer shot against. Again, Hickey comes up looking really good, while Finley looks rather mediocre (not horrible though).
There is of course a second part to looking at shots — shots on goal by the Islanders towards their opponent’s net. You may object, “Hey i thought we were talking Defense!”. Well I’m cheating a little here — offense can of course be the best defense at times. How do the Isles do at getting shots on goal while Finley and Hickey are on the ice?
SOG/60 with Hickey on the Ice: 27.9
SOG/60 with Hickey on the bench: 25.5
SOG/60 with Finley on the ice: 17.3
SOG/60 with Finley off the ice: 30.0
Now here we see again a big difference between the players. When Hickey is on the ice, the Isles get about two more shots per 60 on goal than they usually do. That’s not a large effect, but it’s a positive.
When Finley is on the ice, the Isles get nearly 13 shots per 60 less than they would with him on the bench. This is extremely bad. If Finley played 60 minutes, the Islanders would field 13 less shots than normal and would only get 16 shots per game off! Unsurprisingly, this is by far the lowest on the team.
Combining these last two measures into Shot Differential:
With Hickey on the ice, the Isles shot differential per 60 is: +7.2
With Hickey on the bench, the Isles’ shot differential per 60 is: -1.2
With Finley on the ice, the Isles’ shot differential per 60 is: -8
With Finley off the ice, the Isles’ shot differential is: +3.2
In short, putting Hickey on the ice makes the Isles outshoot their opponents by seven. Putting Finley on the ice results in the Isles being outshot by nearly eight shots per 60.
* * *
Now I anticipate two objections to this analyses of Finley and Hickey that I should answer before we conclude:
Objection 1. ‘Well Hickey has had better linemates than Finley!’ Indeed, Hickey has played around half of his ice time with Lubomir Visnovsky, the Isles’ best defenseman. On the other hand, Finley has played near no time with Visnovsky. How do we know that we’re not confusing Hickeys impact for that of his defensive partner?
Well this is a good question but fortunately thanks to stats.hockeyanalysis.com we have the ability to look at a player’s numbers with specific other players. Both guys have played some time with Matt Carkner — 32 minutes for Hickey, 17:33 for Finley — and Mark Streit — 107 minutes for Finley, 42 minutes for Hickey. Looking at these numbers, we can see that both Streit and Carkner have had massively better results with Hickey than without Hickey. By contrast, both Streit and Carkner have worse results with Finley than without.
Thus it’s clear that Hickey clearly has been a much better player than Finley during the time that each player has had with Streit and Carkner, two of their common linemates.
Objection 2. ‘Well, Finley has size and he brings physicality to the team which Hickey does not!’ This is true, but you need to think about this before you make this objection. For one, the +/- and shot numbers are taking into account the player’s impact with his hits. If a player’s hits are truly good defense and are preventing goals or shots, we should see the # of shots and goals while that player is on the ice be lower than otherwise. And yet, Finley’s numbers are far worse than Hickey’s.
For another, it’s true that hits may have an effect on opponents outside of the time Finley is on the ice. I can’t quantify that. However, Finley’s numbers, as I’ve detailed in this post, have been so bad — and especially worse than Hickey’s — that his hits would have to have a monstrous effect to make playing Finley over Hickey worth it. There’s about a four goal per 60 difference between the two. Do you really think his hits weaken opponents by four goals? It’s just not likely.
* * *
In Short, Thomas Hickey has been yet another steal pulled off by Garth on the waiver wire. When he’s on the ice, the ice is tilted heavily in the Islanders’ favor. That’s a great sign for a defenseman doing his job properly, even if they’re not contributing directly to scoring. Hickey should clearly be a part of this team’s future.
The same could not be said of Joe Finley. He may be big and provide toughness, but when he’s on the ice the team suffers dramatically. Barring an injury, there should never be a time when Finley is chosen to play a game for the Islanders over another active defenseman. Not the least of all, Hickey.
(All stats courtesy of Behindthenet.ca and stats.hockeyanalysis.com)