Hockey is a sport where one player can ruin another team’s night just by being on fire. That one player is a goaltender. As a New Jersey Devils fan for years, I have seen more than my fair share of my favorite team doing a whole lot right but scoring. And the issue is not so much what the Devils did wrong, but the other team’s goaltender just playing out of their mind. This experience has caused me to see and hear phrases like:
“A no-name goalie? Watch the Devils make him look like a Vezina candidate.”
“A backup goaltender? They know we’re not good but watch him be great anyway.”
“Why can’t we finish? Oh, because the other goalie is a wall”
And so forth. Recently, Jared aptly described the 1-2 loss to Winnipeg as the Devils getting “goalie’d” by Eric Comrie. In their most recent game, a 2-1 shootout (!) win over Anaheim, the Devils were arguably having the same experience against backup Ducks goaltender, Anthony Stolarz. Since Nico Daws was about as great and the Devils won, there is less teeth-gnashing over it. Still, the topic is worth exploring. We know the 2021-22 Devils are Bad. Does this mean they are seeing a lot of “lesser” goaltenders? Are they getting robbed in games by goalies and, if so, what does that look like and when? Conversely, have the Devils absolutely pounded a goaltender and, if so, how often and when? Let us take a closer look.
For the purposes of this post, I only care about who is the starting goaltender against the Devils. By my count, the Devils have chased only three starting goaltenders in their 59 games played so far. In terms of whether the opposition goaltender is a starter or not, I needed to make some decisions on what that looks like. This is not the late 1990s or early 2000s where most teams clearly have a starter who plays a ton and a backup who does not. The gap in games played between #1 and #2 can be so small as to be more of a 1A/1B situation. Therefore, I decided on these roles with these definitions:
- Starter - The goaltender who leads their team in games played and has 10 or more appearances than the second goaltender on the team.
- 1A - The goaltender who leads their team in games played and has 9 or fewer appearances more than the second goaltender on the team.
- 1B - The goaltender who is second on the team in games played and has 9 or fewer appearances than the first goaltender on the team.
- Backup - The goaltender who is second on their team in games played and 10 or more fewer appearances than the first goaltender on the team.
- Depth - The goaltender who is third, fourth, fifth, or sixth in games played on the team.
One can quibble as to whether a gap of 9 or fewer makes it a 1A/1B situation. Likewise, some teams are finding out that their backup or 1B goaltender is quite good and is on pace to get more games than their starter or 1A goaltender. Boston appears to be doing this with Jeremy Swayman (1B by a game) over Linus Ullmark (1A). St. Louis might with Ville Husso (1B) and Jordan Binnington (1A). The gap definition also means these roles are a little fluid; just last night Karel Vejmelka moved into “Starter” territory with now 10 more games played than Scott Wedgewood, who has been downgraded to “Backup.” Of course, availability has played a role in this construct - how else to explain that Dustin Tokarski leads Buffalo goalies with 20 games played? However imperfect these definitions may be, they allow the larger questions to be asked. How well have the Devils done against these various kinds of goalies?
I counted up each game the Devils played against each goaltender and filtered out the five different roles defined above. As the questions relate to how well the Devils have or have not done against them, I decided to collect all situations data from Natural Stat Trick from each of the 59 games played so far this season. I want to know how the Devils have done against whoever the opposition starts against them. That includes power plays, shorthanded chances, and overtime situations. Everything but shootouts and empty-net situations by the opposition.
The 2021-22 Devils Against Starting Goaltenders
Believe it or not, the Devils have faced more Starters and 1A goalies than 1B goalies, Backups, and Depth goalies. Not by much, it’s 31 to 28 games, but the plurality of opposition starting goaltenders are Starters by the above definitions.
The above stats are fairly straight forward. CF is Corsi For, or shooting attempts by the Devils. SF is shots on net by the Devils; SCF is scoring chances by the Devils; HDCF is high-danger scoring chances by the Devils; xGF is expected goals by the Devils; and GF is how many goals the Devils scored. QS is a bit different. It stands for Quality Starts, a concept Rob Vollman popularized through his Hockey Abstract books. His 2014 book defined it as a goaltender who had a performance better than the league average save percentage or posted a save percentage better than 88.5% if there were fewer than 20 shots against. This season, the league average save percentage is 90.8% and the Devils have been held under 20 shots against in one game this season (Pittsburgh, 12/19, 2-3 loss).
Of course, the above chart is a gross total. Here is a per-game average for the Devils’ team starts for the same groups of goaltenders.
The Devils have definitely attacked their opposition goaltenders. The Starters have done their best to collectively keep the Devils from coming close to xGF - both by gross and per-game rate. About 65% of their games have had them post a Quality Start. Given the number of games, that their per-game rates are lower than some of the other groups may be a bit more reflective of how the team does. It is not so low that it is particularly concerning, though.
While the population size is low, it’s the 1B group that has given the Devils the most issues. The average save percentage this season is 90.8% and they have collectively stonewalled the Devils at 94.15%. All of seven of those games were Quality Starts; even in the one win the Devils have had against them - which was last Sunday’s 3-2 overtime win over St. Louis. As you can see from the per-game rate chart, it is not for a lack of trying. Their rate of scoring chances are fairly high, their xGF rate is the highest outside of the Depth group, and they have seen the most attempts per game. Again, population sizes should cause one to take this with a grain of salt; but this is what they have accomplished sofar.
Oddly, they have been the most successful against the 1A goalies and the Depth goalies. The 1A group is small; but the Devils have done well to either attack or finish in those three wins. Those three wins were against Dustin Tokarski (Buffalo, 2-1 overtime win on 10/23), Ilya Samsonov (Washington, 4-3 overtime win on 1/2), and Jordan Binnington (St. Louis, 7-4 win on 2/10). What about those no-name goalies buried deeper on a depth chart? Those four Depth goalies were Joey Daccord (Seattle, 4-2 win on 10/19), Ukko-Pekka Luukkonen (Buffalo, 4-3 win on 12/29), Jack LaFontaine (Carolina, 7-4 win on 1/22), and Cayden Primeau (Montreal, 7-1 win on 2/8). In those four games, the Devils offense just pounded the opposition. Given that Seattle, Buffalo, and Montreal are not good teams, perhaps that is not a surprise. Doing it to Carolina, well, that was impressive. Still, one of the People Who Matter should not lament seeing a third-or-below stringer start a game against the Devils this season. The 2021-22 Devils have made them work and suffer.
While an unknown name has not frustratingly denied the Devils goals that they have worked for, the Devils have been goalie’d multiple times this season. This is normal. Everyone runs into a hot goaltender at one point or another. It even happened in their last two games. How often has it happened? Assuming getting goalie’d means the Devils have scored at least one fewer goal than their expected goal amount, it happened in 15 games.
Out of all 15 games, the Devils went 2-12-1. The two wins went beyond regulation, too. By role, 4 were by Backups, 3 were by 1B goalies, 1 was by an 1A, and the remaining 7 were Starters. Again, it is not for a lack of trying as the Devils put up over 30 shots on net in all but two of these games. Yet, these goalies were just that much better on these given nights. I do not think many of the People Who Matter would argue that the goalies in those games were made to look good. No, they did their job quite well. Sure, seeing Eric Comrie and Anthony Stolarz on the list is a bit of a surprise, but they did achieve it. The rest have all been established in the league in the past save for Karel Vejmelka. Not all of these kinds of frustrating nights were the same. The first seven on the list were especially bad nights of getting goalie’d as the Devils were two (or more goals) below the expected goals model at Natural Stat Trick.
Curiously, even if the Devils met expectations in these respective games, the result would not always change. For example, scoring three or four against Dallas on January 25 would have just meant a closer loss assuming all things would be equal. Ditto for the 1-4 loss to Spencer Knight on November 18 or the one and only shutout loss to John Gibson on November 2. Incidentally, Braden Holtby can claim being the one to most impressively deny the Devils - even more than Gibson, who shut them out. I would like to think that a few more goals would have changed the tide of the game, depending on when and how it was scored. It could have flipped the recent 1-2 loss to Winnipeg, made the loss to Our Hated Rivals much more dramatic, and possibly yield points out of the bottom three on this list. Alas, this is speculation and nothing more. It was what it was.
As one last tidbit: No single goalie dominated the Devils more than once this season. It could still happen as the Devils have two more games against Our Hated Rivals and Carolina (I doubt Jack LaFontaine will play in both?) and one more game against Washington, Buffalo, Florida, Dallas, Arizona, and the Islanders. That will depend on who is in net in those games and how they will do.
Crushing the Opposition Goaltenders
This post is not all about the bad times. No, there were some good nights for the offense. Incidentally, the Devils have also had 15 games where they absolutely surpassed the expected goals model against whomever the opposition started. This group includes the three times they chased the initial goaltender.
The Devils went 13-2-0 in those games. They risked it in three of those games as they went to overtime, but they prevailed. The two losses are among the worst of the season. The January 31 loss to Toronto was a result of a meltdown in the third period and arguably Damon Severson’s worst play of the season. The Devils were tied 4-4 in the final ten minutes in Chicago on February 25 and blew it with goals against in back-to-back shifts. Still, the Devils did get wins in a majority of games where the offense lit up the goalie even beyond their xGF in all situations.
For this post, I counted only who started the game; hence the lower numbers for Jaroslav Halak on February 28, Tristan Jarry on February 24, and Jack Campbell on January 31. Two of the three they chased were their team’s respective Starters, the one Backup who was chased - Halak - was the biggest blow-up yet this season with the Devils scoring six goals on attempts yielding 1.5 expected goals. Needless to say, their respective coaches were right in yanking them. If Holtby had the best performance against New Jersey by difference of xGF and actual goals, then Halak definitely had the worst.
Going by role for the other twelve games, you can see three of the four Depth goalies the Devils have faced this season in this group. There were two 1A goalies in Jordan Binnington and Ilya Samsonov where the Devils just made them suffer. They picked on 3 Backups, which includes Spencer Knight, who was on the Goalie’d list. It is three out of four listings as Kevin Lankinen showed up twice. Chicago annoyingly got three points out of those two games, almost in spite of their goaltender. Three Starters also appear on this list, including another double appearance of Ilya Sorokin and a surprising appearance from Andrei Vasilevskiy from that third-period comeback 5-3 win on November 20.
Games Where the Devils Were Expected to Score At Least Three Goals
Now for a bit of twist. What about games where the Devils were expected to score at least three goals? How often did that happen? 22 times, which includes some of the games where the Devils were goalie’d and some where they ripped apart the opposition goaltender.
I highlighted the seven times where they scored four or more and six other times where they scored three goals. The other nine times, well, they were done in by a goalie on some kind of fire that night. A few observations to ponder:
- The Devils went 9-11-2 in these games. Of those 11 losses, seven were getting goalie’d. The other four had other issues. Namely, giving up much more than they were able to score. This was most apparent in that wacky 4-8 loss in Winnipeg back on December 3 and the more recent five-unanswered-against 3-6 loss to Tampa Bay on February 15.
- While the Devils have faced rather good goaltending in this past week, their second best xGF performance of the season was the March 8 comeback win over Colorado. It was not like Darcy Kuemper was bad, but the Devils found ways to beat him - fortunate as two of those goals were against him. You can see that the quantity was there with 42 shots on net, 38 scoring chances, and 18 high-danger chances. There is an argument to be made that particular comeback win may be among the Devils’ best wins of the season.
- Even though the Devils have been goalie’d by a few of the Backups, they did crush Martin Jones per expectations back on November 28. While they did not win either game, they did as well as the model suggested against Joonas Korpisalo on October 31 and January 8.
- Half of these games did yield Quality Starts for the opposition goaltender. Not all of them were in games where they kept the Devils over a goal less than their expected goal amount. It includes Korpisalo’s game in October and Husso’s game last Sunday.
- The best of the Depth goalie performances was by Joey Daccord from October 19. He was just a save away from a Quality Start, too.
- Robin Lehner managed to win a game with not so good performance on December 16. He conceded 3 and was expected to give up 3, but also posted a sub-90% save percentage on 26 shots. He was also technically replaced by Laurent Brossiot for 15 seconds in the game, but I’m not counting that given the end of that game. Odd.
- What about the inverse? Outside of the games where they chased a goalie, the Devils have put up an xGF less than two in six games. They went 3-3-0 in those games, which includes the blow up of Binnington and Vasilevskiy (11/20), the first shootout win of the season (James Reimer, 11/6), and three sadder games (2-4 loss to Juuse Saros on 11/26, 2-5 loss to Thomas Greiss on 12/18, and the stupid-Ruff-played-a-sick-Blackwood 1-6 loss to Carter Hart on 12/14.)
Conclusions & Your Take
So what does this all mean? The Devils’ offense may have some systemic flaws to make more inefficient than it needs to be (read: too many shots from distance and by defensemen), but it has been able to put oppositions under plenty of pressure. In 15 games out of 59, the Devils have been able to push ‘P’ and rack up G’s on their way to some W’s. And there have been 22 total games where the Devils are putting up a lot of attempts, shots, and at good enough locations to be expected to score at least 3 goals. In other words, the Devils’ offense is really not that big of an issue as a whole. Could it be better? Of course. A different set of power play approaches (maybe a different way of using a 1-3-1 entirely), getting some better wingers, and giving the defensemen a “yellow light” for shots instead of a “green one” can yield even more. But it is not as if the Devils’ offense lighting up goalies is few and far between.
As for the outstanding complaints about lesser goalies always looking fantastic against New Jersey, the data really does not bear that out. None of the third-and-below stringers have had good games against the Devils. Sure, a handful of Backups have goalie’d the Devils but there have been plenty of Backups who have not. The 1B group is not necessarily that much worse than their respective 1A’s and the stark difference in results between the two could be more due to variation among a small population than anything meaningful. Lastly, while opponents may be more willing to use backups/1Bs against the Devils in back-to-backs or rest disadvantage situations; it is more likely they have seen a Starter or a 1A goalie than not so far this season. That could change as the season goes on. The Devils do have four games with rest disadvantages plus two double-disadvantages left; and opponents may see a Bad Devils team as a good time as any to give someone a shot. But it is not the norm 59 games into the 2021-22 campaign. And the data so far suggests it could blow up in their face every so often. Will the Devils get goalie’d again? Possibly. This is hockey, these things just happen. Will it happen repeatedly? Eh, not necessarily.
In other words, the old complaints of the past are not exactly applicable to the 2021-22 Devils. There is much to complain about, but I do not think the opposition’s goaltending as a whole is one of them. The Devils’ own goaltending, well, you know that is a different story and absolutely worth complaining about. That is a subject for many other posts. In terms of opposition goaltenders, it is not such a massive issue provided the Devils keep attacking them. Which they have been for the most part.
What do you make of these findings? Are you surprised that the Devils have been done in by hot goaltenders in only 15 games? Or that there have been 15 games where the Devils just crushed whoever was put in the crease against them? Or that over a third of their season so far had the Devils’ offense generate at least 3 xGF? Do you expect opposing teams to keep playing their better goalies against the Devils, or will that shift as the season winds down to its end? Please leave your answers, reactions, and other thoughts about the offense in the comments. Thank you for reading.