It Hasn't Been Perfect, but RPI is Slowly Turning the Ship Around
The Engineers have won 4 of their past 6 games and now sit 7th in the league.
Here’s how they’ve done it:
No Longer Leaking Goals on the PK
It’s no secret that the penalty kill was a bit of a disaster for RPI early in the season. The team was taking a lot of penalties and then giving up goals on about one-third of the ensuing powerplays. Yes, they were facing the likes of Cutter Gauthier and the Nadeau brothers in some of those games, but they also gave up 9 powerplay goals in the Union series and subsequent North Country trip.
I talked about this last month, but there was no way it could stay this ugly for the whole season. The team was on pace for a historically bad penalty kill, and it just didn’t jive with how they’d performed in previous seasons.
Funny enough, if we look at some penalty kill rate metrics, RPI is actually still giving up a lot of goals shorthanded (9.37 goals against/60 while shorthanded would peg them at 59th in the country at the moment).
It hasn’t seemed as bad because RPI is spending a lot less time on the penalty kill. They’ve pretty much cut their time shorthanded in half since the start of the year.
The bigger takeaway from this chart is that RPI’s xGA/60 are way down. They aren’t giving up nearly as many chances anymore. Granted they’ve faced some easier opposition on the powerplay, but outside of a couple matchups against Cornell and Quinnipiac, it’s representative of what they’ll face in this last section of the season.
So, what’s changed?
The players have. Let’s compare PK usage in the beginning of the year to usage now. The team spends much less time on the penalty kill now than they did at the start of the season, so we’ll focus on the percentage of the team’s total PK time that a given player spends on the ice. There are some big differences:
Sutter Muzzatti went from having a limited role to playing the most shorthanded minutes on the team. Regula and Strom have become the clear lock-down defensemen. Jake Gagnon has stepped up.
In short, the team has fallen back on some of its best players to stabilize the penalty kill.
Has it worked? Yes. Is it ideal? Probably not. Sending Muzzatti out this often on the kill eats away from his own time at 5v5 and on the powerplay. But if it prevents a few more goals down the road it might just be worth it.
Evans and Beaton Stepping Up
Tinling and Lee are two of the most important pieces in RPI’s forward group and with both missing some time over the past month, opportunities opened up for a couple guys further down the depth chart. Evans jumped up to the first line alongside Sutter and Heidemann. Beaton stayed as third line center, between Brushett and McNeil, but his ice time increased significantly.
And both have excelled in their bigger roles.
Let’s start by just looking at raw production – the two have combined for 4 goals and 9 assists over the past 6 games. But that doesn’t tell the whole story. Amongst the guys who played in at least 3 of the past 6 games, they rank first and second in Expected Goal Share. Just a monster couple of weeks for these two:
What about the Defense?
Measuring defense has always been much trickier than offense. Defenders don’t get points for preventing goals. You rarely get a good picture of how well a defenseman is doing from looking solely at their point production.
In an effort to demystify the Engineers’ defense a tad bit and see what impact certain players are having I’m going to give a big, messy stat dump and go through it bit-by-bit:
All the way on the left, we have the classic 5v5 metrics. We talk about these a lot because they are generally accepted as good predictors of future success.
Jimmy Goffredo hasn’t had the pleasure of facing some of the top teams in college hockey this season, so his numbers naturally start out a bit higher than the rest of the blueline. Regardless, his metrics are really promising for someone who only just moved up to college hockey.
For the rest of the defensemen, there still aren’t many of these 5v5 numbers over 50%, which is what we want, but they are all trending up. These numbers looked a lot worse back in October. Max Smolinski and Lauri Sertti, in particular, are two guys whose underlying numbers have pretty much done a 180° turn from the beginning of the season.
Penalties drawn is one of those stats that normally isn’t nearly as visible as penalties taken, but it’s just as important. Nick Strom is the one guy on the blueline who is actually completely offsetting the penalties he takes (low to begin with) by drawing just as many for RPI.
The next two categories, hits and puck battles, really give a good perspective of which players are playing with the most physicality, and if that physicality is helping to win the puck back.
We see that Agnew unsurprisingly leads the team in hits, and Sertti is a puck-battle winning machine.
Takeaways and giveaways are another set of indicators to help answer the question: “who is helping to win back and keep the puck from the other team?” Adding to the context of which players are doing the most defensive work.
Not a whole lot of variance amongst the players here in takeaways, but we do see some in the giveaways category. Goffredo has been jumping up into the play in the offensive zone quite a bit, which has been mostly good (and even got him his first collegiate goal!), but it also naturally leads to more giveaways.
“Errors Leading to a Goal” is an Instat metric that they define as “a player’s error which made the opponent’s goal inevitable.” They defined two types of errors:
Individual errors: a player’s unsuccessful action. It can be a lost puck battle resulting in a goal, a missed pass which led to a goal or bad puck handling
Positional error leading to a goal: an error associated with the team formation. In this case, we highlight who should have covered the goal scorer.
This is definitely one of the less-concrete metrics in the stat-dump since there’s some judgement to be had in the definitions above, but nonetheless, it paints a picture of which defensemen are making the most costly mistakes. Smolinski stands out here in a good way.
And finally, defensive zone breakouts. Who’s doing them the most? And what type? It’s no surprise that the more offensive-minded defenseman stick out here (Goffredo, Smolinski, and Regula in particular).
A Few, Quick Hits
RPI won the xG battle both nights this past weekend against St. Lawrence (3.5-2.9) and Clarkson (2.7-1.66).
Jack Watson has a Goals Saved Above Expected (GSAx) of -0.34 per game in his last 5 starts. This equates to giving up an extra goal every 3 games, which isn’t great, but it’s a huge improvement from the start of year in which RPI’s GSAx was a staggering -1.59 per game.
Max Smolinski leads defensemen in both xG (2.5) and scoring chances (8), but has yet to net his first goal of the season. He’s due for one soon.
All stats courtesy of Instat Hockey.