Whenever there’s a way to judge and analyze defensemen, I’m always eager to test it and see how valid it is or how a certain team’s defense corps looks through it. So, when I saw Japers Rink post an article on “defensemen save percentages,” I naturally had to check it out and test it for Carolina’s players. For this project, we’re going to take a look at each defenseman’s on-ice save percentage. Why are we looking at save percentages? Because if a defenseman is constantly put up against an opponent’s first line and imanages to prevent goals, it speaks high of his abilities…that or he’s getting very lucky. Or possibly a little of both. I’ll explain things further after the jump
Going by save percentage is not the best way to judge a defenseman’s ability at all because it is usually the team’s best blue-liners who get matched up against other team’s top forwards and will usually be on ice for more goals than the sheltered defensemen who draw the opponent’s third and fourth lines. A defenseman who is on ice for a less percentage of goals than his teammates is either very good or very lucky. What we’re going to do here is look at each of Carolina’s defensemen from the last four seasons on-ice save percentage, subtract it from the team average and plot it against their corsi relative to quality of competition to see where they stand. Then we can use the other information we know (defensive errors, shots allowed, scoring chances allowed, offensive zone starts, etc.) and find out just how “lucky” some of these defensemen were. I’m also going to do what JP from Japers Rink did and look at how Carolina’s defensemen performed over the last four seasons. It’s a good way to see how Laviolette and Maurice deployed the team’s defensemen over the years and how certain players performances changed over time like Gleason, Corvo and Pitkanen.
First off, I have to address a few embarrassing typos. First, that should be “Gleason 08″ and “Babchuk 10″ in the table and not as they appear there. Also, there’s a few defensemen who aren’t listed here due to small sample size. To clarify the chart a little, the section that is labeled green shows the defensemen who had the highest relative save percentage and faced the toughest competition. The red shows those who saw somewhat weak competition and had low save percentages. The top left yellow quadrant are defensemen who saw weak competition and had high save percentages and the lower right yellow quadrant are those who saw tough competition but had low on-ice save percentages.
Some observations I made here are Tim Gleason facing tough competition the last three seasons and having on-ice save percentage in the last two. Does that mean he is a superb defensive defenseman? His scoring chance, shots allowed and corsi relative data all say that the reason he’s been on-ice for less goals than others is because he’s gotten bailed out by Cam Ward more often than not and he surrenders too many shots. So, we can say here that Gleason was more lucky than good, at least last season. However, he had a pretty respectable “Goal Causing Error Percentage” but that could be due to being on ice for less goals than others. The other heavy-lifter over the last few years has been Joe Corvo and he’s been a lot less lucky than Gleason was over the years. He also led the team in Goal Causing Errors and did not have too good of a scoring chance percentage either, so that shows that most of his mistakes were ending up in the back of the net more times than not.
When thinking of who the best Carolina defenseman was in recent seasons, my guess is that many would say either Joni Pitkanen, Joe Corvo or possibly Gleason. Well, in terms of guys who saw the toughest competition, Bret Hedican and Glen Wesley in 2008 saw the toughest competition and outperformed their expected save percentage by a lot more than others. Shots allowed data suggests that Wesley was the better of the two and he did not have to be sheltered nearly as much. I would not have guessed that either of those two would have ended up on top in this metric.
The guy who I thought would be at the top here, Joni Pitkanen, had to be sheltered in every season until last year and he underperformed his expected save percentage in every year. The underlying numbers we looked in previous posts also show that he’s been pretty below average in terms of suppressing shots and scoring chances too. Looking at his raw data over the last few weeks has definitely shown a lot of interesting things about him as a player. Speaking of sheltered defensemen, Jay Harrison saw weak competition in both of his seasons with Carolina and got really unlucky in both of them. Like Gleason, he had a respectable Error rating but his mistakes were more prone to end up being goals than others.
When looking at some past defensemen, we can see that Dennis Seidenberg had two polar opposite seasons with the Hurricanes. He was put against middling competition in 2008 and he appeared to thrive in that role by surrendering less goals and then struggle in the same role next season. However, when you look at some other details, you’ll see that he surrendered the same amount of shots and the reason why he gave up more goals in 2009 was because he was used a lot more in defensive situations. This illustrates how important zone starts are and how much of a role they can play. What about Niclas Wallin? He was also given tougher zone starts in 2009 compared to 2008, but when you look at the graph above, you’ll see that he faced significantly easier competition and gave up lesser goals.
So what have we learned here? First, we can say that using save percentages isn’t the best way to judge defensemen by, but you can tell that some of the team’s better defensemen ended up on the right side of the graph and were given those tough assignments because of their ability. However, their ability to prevent goals changed a lot over-time so using this may not be the best metric. Also, when you look at other underlying numbers for players like Gleason, you can see the reason why he was on ice for so few goals the last few seasons and that he may have been getting a good amount of luck. Other underlying numbers also explain “phenomenon’s” like Seidenberg’s apparent “drop in play” which was really just him seeing tougher zone starts. The data here could signal some clues as to who will get the tough assignments next year, too.