|
In parts 1 to 3 of this series (which can be found here, here and here, I analyzed how top line forwards, defined as those who get the most even-strength ice time, differ from their peers. We saw in particular that they tend to have better puck control metrics (Corsi and Delta), better finishing ability (on-ice shooting percentage), and much more power-play ice time. Today I will perform the same analysis with defensemen, and we will understand in what ways the selection criteria for good defensemen differ from those for forwards.
As always, data used in this article can be found on Google Docs here.
1. The Data
My philosophy when analyzing data is simple: one player may be a blip, but several players represent a trend. It is often extremely hard to figure out if a player has a particular skill, be it shooting, drawing penalties, or stopping shots at 4v5, because we don’t have sufficient data. Therefore, I decided I would look at Good Players (henceforth GPs) as a group. If I aggregate the data of all GPs, I’ll be able to see what characteristics distinguish them from their peers.
To distinguish GPs, I decided to use even-strength ice time. I chose to use even-strength ice time specifically because I want to be able to look for a correlation between special teams ice time and even-strength, and if I sort by total ice time I won’t be able to do that; by definition, players with high total ice time are more likely to have both high even-strength ice time and special teams ice time.
I counted 366 defensemen who played at least one NHL game in the last two seasons. I then sorted them by their even-strength ice time in 2009-10, from Joni Pitkanen’s 1184 seconds per game (s/GP) to Maxim Noreau’s 203 s/GP (Noreau only played 1 game; in case you’re curious, the “regular” with the lowest ice time was Tim Conboy, who averaged 261 s/GP in 12 games for Carolina). I then summed them into 3 tiers: the 1st tier contained 33% of ice time played by the 62 players with the highest ice time, the 2nd tier contained the next 33% of ice time (75 players), and the 3rd tier contained the last 167 players. I then aggregated the results of all of these players.
2. On-Ice goals and shots for and against
First, let’s go with the basic events that happened while they were on the ice:
2009-10 aggregate results
GP TOI (min) GF GA EGF EGA Corsi +/- Delta Exp +/- (Corsi)
1st tier 4497 77776 3166 3040 3045 2986 842 126 59 67
2nd tier 4695 74550 2896 2929 2861 2879 -397 -33 -18 -32
3rd tier 5900 78188 2895 3058 2902 2991 -964 -163 -88 -77
2009-10 per-60 min results
TOI (s) GF GA EGF EGA Corsi +/- Delta Exp +/- (Corsi)
1st tier 1038 2.44 2.35 2.35 2.30 0.65 0.10 0.05 0.05
2nd tier 953 2.33 2.36 2.30 2.32 -0.32 -0.03 -0.01 -0.03
3rd tier 795 2.22 2.35 2.23 2.30 -0.74 -0.13 -0.07 -0.06
The most striking thing about these results is how closely packed they are. The difference between 1st tier and 3rd tier defensemen was only 4 minutes per game, from 13 to 17 minutes, an increase of 30%; by contrast, 1st tier forwards averaged 60% more even-strength ice time than the depth guys. 1st tier defensemen were collectively +126, which is much better than the -163 put up by the 3rd tier guys; however, by contrast, 1st tier forwards were +578 last year!
What does this mean? Is there no difference in skill between defensemen? There certainly is, but there is less spread than among forwards, at least in terms of influencing results at even-strength. Generally speaking, defensemen don’t drive the results at even-strength; forwards do, especially, as we saw, top forwards. Other online hockey writers have also arrived at this conclusion (see Vic Ferrari here).
We can see that the results were very similar in 2008-09 as well:
2008-09 aggregate results
GP TOI (min) GF GA EGF EGA Corsi +/- DeltaExp +/- (Corsi)
1st tier 4301 70834 2860 2716 2843 2750 1736 144 94 139
2nd tier 4057 62095 2449 2444 2390 2421 -994 5 -31 -80
3rd tier 5071 69764 2561 2691 2617 2683 -713 -130 -63 -57
Didn't
play
in 2009-10 1683 21393 741 836 787 835 -521 -95 -48 -42
2008-09 per-60 min results
TOI (s) GF GA EGF EGA Corsi +/- Delta Exp +/- (Corsi)
1st tier 988 2.42 2.30 2.41 2.33 1.47 0.12 0.08 0.12
2nd tier 918 2.37 2.36 2.31 2.34 -0.96 0.00 -0.03 -0.08
3rd tier 825 2.20 2.31 2.25 2.31 -0.61 -0.11 -0.05 -0.05
Didn't
play
in 2009-10 763 2.08 2.34 2.21 2.34 -1.46 -0.27 -0.13 -0.12
The aggregate +/- and Delta numbers are similar; interestingly, the aggregate Corsi for GPs was much better in 2008-09.
It’s also interesting to see that the players who didn’t make it back to the NHL the following year averaged less than 13 minutes a game and had a +/- of -0.27/60 min, both worse than the average 3rd tier defenseman. If you’re dressing for less than 12 minutes a game, getting smoked, and you’re not 19 years old, chances are your NHL career is in danger.
3. Situation, Opponents and Teammates
There are other ways in which GPs differ from their peers, and that is in the caliber of opponents they face.
2009-10 Delta results
DeltaDS DeltaDO DeltaDT Total
1st tier 7 69 -26 51
2nd tier 8 21 4 33
3rd tier -15 -94 8 -101
We can see in the above table that zone starts don’t vary much by caliber of defenseman, while GPs tend to play against much stronger opponents, and with somewhat stronger teammates. This is what we expect.
It is at this point that I figured out that something is amiss. Even-strength ice time alone was not an optimal filter of defenseman skill. I decided to try other combinations, and eventually settled on even-strength + short-handed ice time. The results were the following:
2009-10 Delta results
DeltaDS DeltaDO DeltaDT Total1st tier 10 85 -23 72
2nd tier 5 16 -16 5
3rd tier -15 -105 26 -94
While this doesn’t seem like a big change, the spread in Difficulty of Opposition has gotten 20% wider. We see less spread, but similar trends, in the previous year:
2008-09 Delta results
DeltaDS DeltaDO DeltaDT Total
1st tier -1 58 -23 34
2nd tier 6 29 -17 18
3rd tier -5 -56 9 -51
That’s the end for now. Next time I will look at Special Teams Ice time, as well as Goals and Assists, and give you my conclusions.
Tom Awad is an author of Hockey Prospectus.
You can contact Tom by clicking here or click here to see Tom's other articles.
| 
Numbers On Ice (11/11)
2 comments have been left for this article. 
Interesting - but it seems to me that the biggest difference is just ice time. So if the difference between top tier d-men and third tier is how much they play (at even strength), and within those differences, their play is fairly similar, could it be that third tier players are feared by their coaches to be exposable if given too much ice time? And if so, why? What proof? I think a good test would be to isolate players whose ice time changed enough to switch tiers season over season. What else changes for them? If their play improves with less time, it may be a sign that the reduced time allowed prevented that exposure.