The New Jersey Devils have started off the season better than most people predicted. Indeed, I was particularly bearish on the team heading into the season, and have been happy to be wrong about that early on, even despite the current two game skid. Jack Hughes has had the start the team desperately needed him to have, as he needs to blossom into that absolute #1. Some players have surprised, such as Miles Wood and especially Scott Wedgewood, who has been better than expected in his replacement role. And he replaced a smoking hot goaltender in Mackenzie Blackwood who played the first couple of games in a Vezina fashion. If he plays like that upon his return, watch out.
However, a little below the surface, there are perhaps some warning signs that this might not be able to last. When some numbers are thrown together, it looks like the Devils have had a bit of luck at the start of the season, and over the course of a full season, should regress to the mean, meaning less success as we move forward. In fact, over the last two games, that might have started to happen already.
To showcase the luck this team has had early on, I will be showcasing various charts from Sean Tierney over at Charting Hockey from both before and after Thursday’s game. First, here is one showcasing luck versus expectation. On the x-axis, we see each team’s expected win percentage, and on the y-axis we see their wins above expectation. This was the chart before Thursday night’s game against Philly:
Heading into that game, no team was luckier than the 3-2-1 Devils. They had almost a full 1.0 wins above expectation, yet only had an expected win percentage around 0.36. The only teams with a lower expected win percentage were St. Louis and Vancouver, but neither of them had anywhere near a full win above expectations.
A lot of that position on the top left of that graph was thanks to the team’s PDO, which was crazy high. Here was the PDO breakdown prior to the games Thursday:
Thanks to an incredible save percentage between Blackwood and Wedgewood, almost no team was luckier. Philadelphia had an unreal shooting percentage, but poor play by Carter Hart dragged down their overall PDO. Tampa hovered somewhere in between, and of course has been winning with that formula.
However, Thursday night’s game helped to drag the Devils away from their incredibly lucky platform. The Devils dominated the first 40 minutes of the game, and outshot Philly 24-8 through two periods, yet the game was only 1-1. That is not a lucky formula, and after a horrible third period and a 3-1 loss, the Devils luck has started to return towards expectation. Here is the PDO chart after Thursday:
Compared to the previous chart, the Devils have moved down and to the left, meaning both their save percentage and shooting percentage dropped after Thursday night’s game. The three goals let up by Wedgewood did not help, but putting all of those shots on Hart and getting only one goal to show for it was deadly. Being in the top left quadrant is not as bad as the bottom left, where the truly unlucky teams are, and the Devils won’t get there any time soon thanks to some excellent goaltending, but they are no longer specifically lucky in terms of PDO.
Finally, with some falling luck, expectations crash a little more. Here is the chart after Thursday night’s games on expected win percentage. Before the Philly game on Thursday, the Devils were in the bottom right, or ‘lucky’ quadrant, but now you can see where they are:
Without luck, analytics and projections once again have the team as “bad”, which is where many expected the team to be, and where a 3-3-1 record will get most teams. It is a rather big difference from the 3-1-1 that they were at the start of the week. And on this specific chart, the reason for the poor positioning is the team’s expected goals, or xG, which is super low. Only Vancouver, St. Louis, and Anaheim are expected to score fewer goals than New Jersey. Without a stronger offense, it will be tough sledding, even with the strong goalie performances the team has been receiving.
Now, despite this, it does not necessarily mean that the Devils will definitely start to lose a lot of games moving forward, and it definitely doesn’t mean that the two game losing streak will continue today when they face Buffalo. Even though the Devils luck numbers are falling back to normal, the main thing keeping them afloat is goaltending, and that can continue, especially if Blackwood comes back and plays like he was. If Devils goaltending remains one of the best in the league, that does not mean luck automatically, it means a strong season from the netminders. And that is not out of the picture for Blackwood. Luck, in terms of PDO, is determined by a team’s shooting percentage + their save percentage. If Blackwood and backups, Wedgewood at this point, maintain a top 5 save percentage, that would add to the team’s luck in terms of these analytics, but it could also just mean great goaltending from a top notch goaltender, which is something Blackwood can be. That could drive the Devils team PDO over 1.00, and could keep them in the luck quadrant on Sean’s charts, but could still mean sustained success. But as you can also see, that also depends on the team improving on their xG. They need to generate more offense, push the puck forward, and score more. That was especially true on Thursday night.
It will certainly be interesting to see how the team plays today, and if the slide continues with more bad luck back towards the mean, or if the Devils can buck this two-game skid and get back on the right track with better offense and more goals. Regardless of goaltending, the Devils will need to score more to win some of these games, and perhaps that can happen today against Buffalo.