"He reviewed all 127? Wow. Oh, and you apparently had 4 errors, Mark." "Sigh." (Photo by Paul Bereswill/Getty Images)
The 2010-11 season wasn't kind of the New Jersey Devils. The team fell flat on it's collective face on October and didn't really get up from the ground until mid-January. From then on, the team's performances were vastly more respectable and successful, but it wasn't enough to make the playoffs or even get out of the bottom third of the NHL.
As with many of the players, it wasn't a good season for Martin Brodeur. He suffered three injuries and he saw the bench for a few games in January and again when Johan Hedberg was blazing in February after his third injury, all of which limited his playing time to 56 appearances and 3,116 minutes. That's enough for many starters in the league, but much less than Brodeur's usual 70+ game seasons. On top of that, Brodeur's numbers were some of the worst in a decade. His total save percentage of 90.3% was his lowest since 2001-02 (90.6%); and his goals against average of 2.45 was his highest since 2005-06 (2.57). Breaking down the save percentages further shows that his even strength save percentage of 91.2% was a drop off over 2009-10's 92.4% and his lowest since 2001-02 (91.2%). The backup goaltender, Hedberg, had better numbers than Brodeur with a considerable amount of playing time last season. Even disregarding how the team did, it's a reasonable conclusion that Brodeur had a bad season.
I'm not going to argue that Brodeur was great in 10-11; but I will argue that I don't believe he was bad as these stats show.
Please recall that all stats are based on events. In this case, it's about goals and saves. Goals are clearly one of the most important events in a game; as they decide them. A goal allowed by a goaltender is a mark against them regardless of how the goal went in or whether his teammate made an error that led to the goal. Fortunately, these events are so important that NHL.com keeps video of every single one. This means we can watch them again and note what went on with the goal and clearly determine whether it was a shot the goalie should have stopped, whether the goaltender had no real chance on the shot, or whether the shot was just too good and would have beaten most goalies short of a miraculous stop. Goalies get the blame regardless, but through looking closer at the goals allowed, we can determine whether they deserved it.
It's that last point that drove me to review all of Brodeur's goals allowed in 2009-10; where I found that a vast majority of the goals against weren't his fault. On top of answering the larger question of whether he should be re-signed, it also drove me to look at the goals Johan Hedberg allowed last season - where I came to the same conclusion. And so it drove me to look at the 127 goals allowed by Brodeur last season, where I found non-soft goals to outnumber soft goals every month. This is the summary of all of those months-in-review, please continue after the jump to get a better perspective on how Brodeur did last season beyond the basic statlines.
The Previous Posts in This Series
The following list consists of posts with links to each of the goals allowed by Martin Brodeur in this last season for the New Jersey Devils, hosted and provided by NHL.com. If there's one in particular you would like to see, or if you want to go through them all yourself, then you are able to do so. The explanation of my approach are in these posts (e.g. what is a "soft" goal, what is a "skater error"), as well as details for the goals allowed in that particular month. They are presented in this list in chronological order of the season. Note: goals allowed will be referred to as "GA" as an abbreviation.
- October 2010 - 31 GA in 12 appearances
- November 2010 - 11 GA in 5 appearances
- December 2010 - 29 GA in 9 appearances
- January 2011 - 18 GA in 9 appearances
- February 2011 - 4 GA in 4 appearances
- March 2011 - 24 GA in 14 appearances
- April 2011 - 10 GA in 3 appearances
The Goals Allowed by Johan Hedberg in 2010-11: 127 Goals in 56 Games
In each monthly post, I reviewed each goal allowed by Brodeur and made a judgment on whether it was a "soft" goal or not. I have defined a soft goal as the following: The goalie must have seen the shot coming; the shot was not deflected or change otherwise in motion; the goalie was in position to actually make the stop; and the goaltender made an uncharacteristic mistake that led to the goal. If all were true, then I deemed the goal as "soft." I've summarized each month's count in the following chart to determine how many soft goals did Brodeur allow all season.
As shown in this chart and the graph, Brodeur peaked in gross soft goals in October and hasn't tied that mark in subsequent months. There was a rise going from November to January in terms of gross soft GAs. Still, the percentage of soft goals allowed per month did not exceed October with the lone exception of January. Nevertheless, Brodeur allowed only 21 goals that he should have really stopped all season. The vast majority of the 127 GAs that got through Brodeur were not solely his fault. This shouldn't be news to those of you who have been following the posts for each month.
What is news is that it is an improvement over last season, where Brodeur gave up 35 GAs out of 168 in the regular season. Brodeur's save percentage went down, his GAA went up, his record was wretched, but the percentage of soft goals allowed dropped by 4.29%. While I only have two seasons of data and it's unclear whether this could just be variation from season to season, it's still an improvement. The same holds true when you calculate these goals into a rate like goals against average.
In the 2009-10 regular season, Brodeur's soft goals GAA was 0.47. Again, 2010-11 showed an improvement in this regard. Given that total GAA rose, it is further indicative that Brodeur's GAs - like Hedberg's - were largely not his fault, and could be attributed to others.
Skater Errors Among the 127 Goals Allowed
In order to identify those others that would be at play that led to a GA, I counted when another Devil made an egregious mistake on a play. I called them "skater errors" as to identify the forward or defenseman who did something wrong or not well enough on a GA. For example, an error could be a lack of coverage on an opposing player; a giveaway; not being in the right position; or some other kind of poor decision. Just as I did for Hedberg, this can answer two questions: How often do they happen and who's most at fault?
This chart answers the first question. With the exception of March, a majority of the GAs for Brodeur in every month had at least one skater make an error. Even in March skater errors were involved in just under half of all GAs. Overall, a majority of all 127 GAs had some kind of error. That's a very important finding. It's evidence that if the players in front of Brodeur make mistakes, then Brodeur - and by extension, the Devils - suffer. And he suffered a lot.
As a brief aside, I can appreciate the concept of a goaltender getting lucky and making a brilliant stop or seemingly be so hot in net that he'll make remarkable stops on a given night. I definitely appreciate the saves themselves. However, those are rare occurrences. They cannot be counted on to happen regularly, like Jay Pandolfo scoring a goal in a way. A goaltender's job can be made incredibly difficult - if not impossible in some cases - when the defense isn't picking up guys at the crease on a rebound or when a forward coughs up the puck en route to a breakaway. Mistakes will always happen as no team can play perfectly sound hockey for 82 games; but that's something opposing teams play for - to capitalize on a mistake. The Devils made that their trademark throughout the late 1990s and early 2000s, in a way. Of course, sometimes the other team just sets up a good play and fires a good shot ; which is not necessarily a skater's or a goaltender's fault. That's why not every single goal allowed has an error associated with it. Still, my main point is that if other teams capitalize more on New Jersey's mistakes in front of the goalie, then that needs to be curbed instead of demanding that the goaltender constantly bail out the team when they happen and/or lamenting when they don't always make that miraculous stop.
Getting back to the review, I will say that some of these errors were recorded even on soft goals. Six of them, to be precise (GA# 10, 24, 80, 84, 85, 102). It is possible that even with the skater screwing up that Brodeur still should have made a stop. Accounting for those 6, that leaves 69 non-soft GAs, or roughly 65% of all non-soft GAs and roughly 54% of all GAs, that had at least one Devil make a mistake. That represents a significant portion of Brodeur's GAs last season; and so I see it as a prime area for improvement. If we're looking for Brodeur to have better stats or the team in general to cut down on goals against in 2011-12, then the first place the coaches should be looking at are these kind of errors. Reducing these is going to yield larger improvements than appealing to Brodeur to cut down on soft goals (though he should do so anyway, every little bit helps).
Now, who committed most of these errors? Well, I've counted those as well. For this chart, a single error means the player was solely called out for their mistake, and a multiple error means they were called out along with another player on the play.
It shouldn't be so surprising to see so many defensemen near the top of this list. Henrik Tallinder, Colin White, Mark Fayne, and Andy Greene all played significant minutes with the team last season; and when they made a mistake, that led to big problems for Brodeur and the Devils. Travis Zajac leads the forwards in errors, which surprised me since he didn't always face the tough competition and his defensive work is pretty good. I suppose this suggests it could be better in a few spots. That said, I wouldn't use this to immediately point the finger at Tallinder and say that he was really horrible, or that we should say "Good riddance!" to White. It just means they could stand to improve in this regard, and that would help out matters in 11-12.
The most interesting thing about this list is that there are several rookies on the list. Brodeur had to play with several rookies in front of him at times in 2010-11, especially in the first half of the season. Among the 92 identified players who made an error, 30 of them come from rookies. Fayne led the group, while Matthew Corrente (still a rookie by NHL standards), Olivier Magnan, and Matt Taormina chipped in their fair share. This actually makes me hopeful that next season, in a way. Some of these rookies won't be around to make errors, and those who will remain in New Jersey (Fayne, Palmieri, maybe Taormina) will improve as players and hopefully make fewer mistakes in games. This will definitely help out in the cause of reducing skating errors next season.
While I don't have anything like this from 2009-10 for comparison, this chart shows some intriguing contrasts to the error totals for Hedberg. Neither Ilya Kovalchuk or Jamie Langenbrunner made many major screw-ups in front of Brodeur. Tallinder and White made more in front of Brodeur than he did in front of Hedberg, while Greene remained consistent (4 singles, 2 multiples). Rolston didn't blow it when Moose was in net, but did so five times with MB30 in net. This may seem more coincidence than anything else; after all, Brodeur played more than Hedberg and so the guys in front of him had more chances to make mistakes that would end up costing the team on the scoreboard. Still, I think it's interesting enough to point out in some regard.
In that vein, I want to highlight where Brodeur was beaten in net over the last season.
The Relative Location of All 127 Goals Allowed by Martin Brodeur
In each month of this GA review, I recorded where the puck got past Brodeur relative to his own position in net. I did this to count how often he got beaten in a certain position or by shots of varying height. It's not a precise method, but it does provide some general understanding of where the puck had been going in so much.
The two most common areas where Brodeur was beaten are interestingly opposite of each other. The most common was the high left side, either over glove, just past an out-reached glove, or over the left shoulder. The second most was low to Brodeur's right, either past his right pad and/or stick, under it, or just a low shot on his right flank. The former suggests that Brodeur's glove side is starting to become more vulnerable, especially since he only got beaten there 27 times in a longer 2009-10 season. However, noticing that he was beaten only three fewer times low and to the other side suggests otherwise.
Representing 58 GAs, low shots were more prevalent than either medium or high shots when it comes to height. That doesn't surprise me so much as keeping the height down on a shot is less risky for a shooter to miss it. Especially if the shot is at a point-blank range like an opposing player putting a puck in off a rebound at the crease. It's going to be more difficult to put height on a shot in that sort of situation. In any case, this may mean the larger quantity of low-right GAs as opposed to low-middle and low-left could be just happenstance. The shooter's going to just put it in where he sees an opening, it could be that just happened to be on Brodeur's right side more so than, say, his five-hole in 2010-11.
As far as width goes, Brodeur did get beaten on the left side 56 times, which is more than the 49 times he was beaten on the right side in 2010-11. This could mean opponents were targeting that side more often in this past season, but I'm not so certain. It is only a difference of 7 GAs. I didn't think 4 was enough last season to really think that was a significant enough gap; and so I'm not so certain 7 is either.
My main conclusion is while Brodeur's stats from last season were bad, I don't believe Brodeur himself was as bad they indicate. He didn't give up a lot of soft goals; 83.46% of the goals he did allow weren't soft. In fact, he gave up fewer soft goals both in gross and in rate than last season. A majority of the goals allowed - about 65% of non-soft goals - also featured at least Devils skater making a mistake of some sort that led to the goal.
This is further indication that if the Devils want to cut down on goals against, then the skaters should be looked at first to cut down on errors. Given that just under a third of all skater errors came from rookies last season, there could be improvement just from having more experienced players and/or not having those players play in New Jersey. As for Brodeur himself, I think he should keep on doing what he has been doing. If the majority of goals allowed weren't his fault and he even cut down on soft goals, then I'm not sure what else he can do. If anything, maybe work on his glove hand, given how often he was beaten there.
Combined with last year's summary and the summary of Hedberg's 2010-11 season, this review serves as further evidence that a majority of goals allowed by a goalkeeper aren't his fault. In the big picture, that's rather important since the natural reaction to a goal allowed is to blame the goaltender. The stats are only concerned with whether a goal was allowed or whether a shot was stopped. By looking at the goal itself more closely, we come away again with a better understanding of what went on. Hopefully, this will lead to more thoughtful judgments of a goaltender's performance - be it Brodeur, Hedberg, or anyone else - and drive more people to break down the play that led to the goal as opposed to just looking at the fact that there was a goal. I doubt that it'll go away completely, but we can improve - just as we want the Devils to do from season to season.
I fully realize that there are flaws in this casual qualitative analysis. You may disagree with my definitions for a skater error or a soft goal. You may have seen some of the videos and come to a different conclusion. You may think the scope needs to be larger to come to more worthwhile conclusions, such as whether Brodeur's 21 soft GAs stand relative to the league or whether other, more successful goaltenders had to deal with as many skating errors. Similarly, you may believe a review of prior seasons, those before 2009-10, are warranted. I fully recognize all of those criticisms and then some. Feel free to bring them up in the comments, though do note that I don't plan to go beyond what I uncovered for this season due to reasons of time and interest.
If you want to use the information in this post and in the other posts that reviewed the goals allowed by Martin Brodeur by month as a jumping off point for further analysis, or to approach it in another way, then by all means, go for it. Even discuss it in the comments. I think an open discussion is beneficial and could lead to a "next step" for this kind of qualitative analysis.
Most of all, this is what I want to know: what's your opinion on all of this? What did you learn from this summary of Brodeur's 2010-11 season? What do you think about Brodeur cutting down on his soft goals count from last season despite allowing more goals in general? What did you takeaway from the skating error charts? How about the shot location data, did it sway your thoughts in one way or another? Do you agree or disagree with my own conclusions? What do you expect from Martin Brodeur in 2011-12 based on this summary? Please let me know all of your thoughts and feelings about Brodeur's play last season as well as this summary in the comments. Thank you for reading.