Nothing like a depth signing to rile up Oilersnation. If this is foreshadowing how the off-season will unfold in Edmonton, then the next eight weeks will be a wild ride. And I can’t wait. Opinions on players, coaches, management, signings, trades and even anthem singers touch every possible spot across the sporting spectrum. The Devin Shore signing was received quite differently across the hockey analysis spectrum.
I can see the argument for those opposed to the signing and those who liked it. I think I’m in the middle. The signing didn’t excite me or infuriate me. I see both the pros and cons of his deal.
My initial thought was his ability to play centre and wing is a plus. If he is the Oilers’ 13th forward to start the season, then they are better off than they were last year when Patrick Russell and Joakim Nygard were the extras. I don’t believe this signing automatically makes him the fourth line LW, but if he is, the price point is fine. The question will be his effectiveness on the ice.
His goal production this past season was consistent with his previous four seasons. He had five goals in 38 games this season, which prorates to 10 in 76 games and 11 in 82 games. He had five goals in 45 games in 2020, 10 goals in 76 games in 2019 and 11 and 13 goals in 2018 and 2017 in 82 games. He has consistently found ways to produce around 10 goals a season. There is value in that.
The concern is that in his career he has consistently been on the ice for more goals against than goals for. That is a valid concern, and the main reason I was surprised he got a two-year deal. The $850K cap hit ($750K this coming year and $950K in 2022/2023) isn’t a massive issue, because if the Oilers end up with 14 forwards they like better, then Shore can be sent to the minors and not count against the cap.
The salary is fine, but the on-ice results are not. Not every goal scored or goal against is created equal, but over a span of five seasons if you have been on the ice (at 5×5) for more goals for than against, that isn’t ideal. He started more in the defensive zone in Edmonton last season, so that plays a factor. But in previous seasons he started more often in the offensive zone. In Edmonton will be used in the “ugly” areas: more D-zone starts and on the PK. When you start that often in the D-zone it is more difficult to have a positive GF%. You hope it is closer to 46%-48% though, and not the GF34% he had this year.
Shore is a smart player. He is strong positionally. Rarely do you see him out of position, and I sense that is why he was signed. The other factors, which aren’t trackable but still have value, are that he fits well in the room and accepts his role. Those little details do matter, even if they don’t show up in any stat.
I see Shore battling Tyler Benson and likely another veteran signing for the fourth line left wing spot. If Shore, Josh Archibald and Ryan McLeod are the fourth line, I think the fourth line should be close to breaking even in GF-GA.
If Shore isn’t a regular that means the Oilers have signed some better players. That would be a good thing. Edmonton still isn’t bursting with NHL-ready prospects. Dylan Holloway and Benson aren’t locks to be in the NHL this year, so the organization needs NHL forwards. Shore is like many of the other 250 players in the NHL who don’t have good analytics. In a league of 700 players there are always 250 who don’t have good analytics. Of course, you’d prefer to not have any of those players, but that isn’t very reasonable. Even Anthony Cirelli had a 45.4GF% in Tampa Bay. He is a very good player, but even good teams will have a few players who get outscored or out chanced.
Some might not like this, but I still see Edmonton in a stop-gap situation for some places on the roster. I see Shore as a stop gap until there are more young players coming up the system. Sadly, for Oilers fans, the Oilers are still in the building process. There isn’t one free agent fourth line winger who will automatically shift the balance of the bottom six. Shore is a depth player who got a depth player contract.
I think analytics is information. Having more information should help us make better decisions, but it is also important to note that what looks good on paper won’t always equate to success on ice.
I just finished reading The Psychology of Money by Morgan Housel. It was an excellent read, and while it focused on finances, I found some of Housel’s explanations worked well with sports analytics. One chapter focused on rational and reasonable.
This is just one example, and it illustrates how it is possible to have a rational analytic assessment, while also being unreasonable.
Darnell Nurse… look. Elite offence this year, a whole ton of goals, but his defensive metrics have never been any good at all and this season is no different. Maybe Barrie is so bad defensively that he's breaking the model but… #LetsGoOilers pic.twitter.com/1DWAwnY0rP
— JFresh (@JFreshHockey) June 9, 2021
If you look at some of these numbers they don’t look good. However, how is it possible that Nurse only has 1% on the PK, when he played the most minutes on the ninth-best PK in the NHL?
So I asked JFresh about the PK total.
“An expected-goal based analysis of a penalty kill or powerplay will be a lot weaker than one of even strength play because the samples are smaller and the stuff that’s missing from the public data (pre-shot passing especially) is way more consequential,” he said.
I’ve seen others point out the Oilers PK was better without Nurse because in 118 minutes without him he xGA was 6.68 while in his 137 minutes it was 9.9. I asked JFresh about this.
“Everything about xG on the PK is subject to nonsense.”
The other key factor to remember is Nurse usually started the PK against the opposing teams’ top units. Most top units are more dangerous than the second, so a reasonable conclusion would be that Nurse should have a higher xGA, because he faces more dangerous offensive players.
I think it is important to remember what JFresh mentioned about what could be missing from public data, and his blunt honesty about xG on the PK. When we make assessments on players, not all stats are equal or accurate, similar to hits. I don’t believe they are accurate from arena to arena.
The public data is good, but it isn’t always complete, because as JFresh said some data is not available publicly…yet, anyway. But it is a great starting point and it is improving all the time with more information. I’ll be curious to see what comes from player tracking and how much we will see in the public realm.
While the rational equation suggests Nurse is 3% on EV defence, I don’t believe that is reasonable reflection of his on-ice play. I also didn’t have him in the top-three for Norris voting, but I bet he finishes in the top-10.
And lastly, I leave you will this humorous line from JFresh, and keep in mind, he is a strong believer in analytics.
“Frankly, I think the only reasonable thing to do is overreact to a depth player getting a depth player contract.”
Recently by Jason Gregor:
- NHL Random Thoughts…
- Help Me Understand: 7th Edition
- Oilers Off-Season Scenarios
- Oilers Need a Third Line Centre
- Oilers Top Priority: Scoring Wingers