I don't think Ekholm's that much better than Nurse, and tbh I think Nurse's potential as an all around defender is higher. My personal opinion tho

You are over the retention limit

Flames decline unless they can clean up cap

If the Preds decide to move Ekholm, Nurse doesn't really make sense for them as part of the return. The only scenarios I see NSH trading him would be to free themselves from needing to protect four dmen in expansion, or if they have a poor season and decide to blow up the roster and rebuild. Nurse doesn't help them in either situation. NSH could get a better return than the picks and prospect like Maksimov without having to take back someone like Neal, from another team.

I would be ok admitting be loosing value wise but 1-1 nurse for ekholm. Preds get younger and faster oilers get more experienced and a steadier defensive presences. Think it would work out nicely for both clubs. I’m not entirely sure about Nashville but I would think they shouldn’t have any issue resigning nurse where Im not sure oil can ekholm should be more within oil price range given that time frame. Also think ekholm is the perfect absolutely perfect “replacement” for klef while he’s getting hopefully fixed up. I love the idea

Over the past 2 seasons Ekholm has been worth approximately ~9 more Standings Points than Nurse. Which is absurd. In that timeframe there are a total of 50 Dmen who have racked up at least 3000 mins on the ice. Of those 50, Ekholm is 12th...Nurse is 45th. He's been literally below replacement over the past 2 seasons while Ekholm has been one of the league's top defenders. The gap between Nurse and Ekholm in terms of on-ice performance over the last 2 years is equal to the gap between Brady Skjei and Victor Hedman...

As a Nashville fan my problem isn't necessarily a 1 for 1 Nurse/Ekholm Swap. It has risks for Nashville, Nurse is more expensive and may want more on his next deal then Ekholm, he's not as good defensively but the more I've looked at it the more the idea has grown on me. For Nashville, we get younger, faster, and a littler grittier. Nashville might take a step back defensively for a year or two but Nurse is still growing as a dman so i think he'll end up somewhere around where Ekholm is now. Ekholm really started to round his game out when Subban came over 4 seasons ago. That puts him around Nurse's age when his game really started to flourish. Edmonton gets better defensively to help maximize a win now attitude. The cap difference between Ekholm and Nurse might be just enough to sign Bear on a 1 year bridge deal.

My problem is with the rest of the trade. Neal even retained adds too much dead cap space. And the rest isn't near enough for me to be willing to take that dead cap space on.

Ekholm has been more than twice as valuable offensively too. Would you swap Josi for Keith Yandle? That's an equivalent swap in terms of performance over the last 2 seasons...

What does that mean? One is awful at defense and the other is a top 4 dman in the league

It means in terms of performance over the past 2 seasons, Ekholm's been basically equal to Josi. And Yandle's basically equal to Nurse.

Except Josi has spent the past 2 seasons carrying his team, while Ekholm had it pretty easy playing behind... Josi

not sure what you mean by behind in this case. Josi plays with Ellis. Ekholm's never really on the ice with Josi, though they get the same quality of matchups according to the numbers.

In terms of getting help from teammates... the data shows Ekholm's had it pretty difficult while Josi hasn't. especially this season where Ekholm had to drag around Fabbro, who graded out as one of the worst offensive impact dmen in the league.

What do I mean? That Josi has been playing a ton of minutes over the past 2 seasons, quarterbacking their PP while being a key PKer as well.

Replace Josi with a worse d-man than him, say, Werenski, and NSH would’ve been in the bottom of the league. He has been robbed of the Norris this past year

Get good players, keep good players. The Oilers suffered through a decade of darkness precisely by making these kind of trades. You'd think their fans would learn...

Nurse isn’t great at offense and is subpar at defense. He isn’t anything special. If anything I should add more value going to NSH

Ekholm was paired last season with Fabbro who had a very bad year statistically. Year before was Subban. You could maybe make that argument when he was paired with Subban but being paired with a very green dman having a bad sophemore year and averaging 23+ Mins/game should've dragged him down.

12th in what? 45th in what? And in any case, being 45th in the league isn't "literally below replacement" in anything, that's just ludicrous.

Over the past 2 years Nurse was +1.28 GF% rel while Ekholm was -1.90 GF% rel while playing easier minutes. He's 6 years older, he's past his prime while Nurse is just about to enter his prime. Zero interest in a 1-for-1 deal from the Oilers perspective.

Top 5v5 defensemen scorers over the past 3 years (and their age over that time):

1. Carlson: 100 (27-30)
2. Josi: 99 (27-29)
3. Burns: 95 (32-34)
4. Karlsson: 83 (27-29)
4. Rielly: 83 (24-26)
6. Barrie: 78 (26-28)
7. Hedman: 77 (26-29)
7. Nurse: 77 (22-25)

I think it's your expectations that need adjusting. That was an incredibly uninformed comment.

As for defense you can look at comment #17 - Nurse has been providing the Oilers with above average goal differential despite playing the hardest minutes his coaches could find.

oh sorry it's value. The smart guys at Evolving Wild developed a model called xSPAR which calculates a players total value on the ice. Think similar to baseball's calculation for WAR. For Dmen in particular I prefer it. Evolving Wild model is based on the original War-on-ice model from Andrew C Thomas (formerly of the Minnesota Wild) and Sam Ventura (now head of hockey research for the Penguins). You can read about the inspiration how it's calculated here. http://blog.war-on-ice.com/index.html%3Fp=429.html

You cite relative GF% so clearly you have an interest in analytics. Here's a look at aging curves since you seem interested in that as well... https://hockey-graphs.com/2017/03/23/a-new-look-at-aging-curves-for-nhl-skaters-part-1/

Hockey is too nuanced for the current "one number" stats, I don't trust them at all and they severely lack context. There simply isn't reliable enough inputs to create a one-number rating for players without missing a whole lot of critical information.

Your aging article only reinforces my point - trading a 25-year-old for a 31-year-old is an incredibly stupid idea. And just look at the top scorers from the past 3 years (5v5):

1. Carlson: 100 (27-30)
2. Josi: 99 (27-29)
3. Burns: 95 (32-34)
4. Karlsson: 83 (27-29)
4. Rielly: 83 (24-26)
6. Barrie: 78 (26-28)
7. Hedman: 77 (26-29)
7. Nurse: 77 (22-25)

Other than Brent Burns, they're ALL in the 25-30 range, mostly in the 27-29 bracket actually. Rielly is a bit younger, and Nurse is way younger. No other defenseman from his age group is keeping up with the "prime" age guys, that's not a player you trade away for a 31-year-old.

http://naturalstattrick.com/playerteams.php?fromseason=20172018&thruseason=20192020&stype=2&sit=5v5&score=all&stdoi=std&rate=n&team=ALL&pos=D&loc=B&toi=2000&gpfilt=none&fd=&td=&tgp=410&lines=single&draftteam=ALL

"Hockey is too nuanced for the current "one number" stats, I don't trust them at all and they severely lack context. "

*Posts raw points table without context*

The beauty of models like GAR/xGAR is that they actually consider all the contexts and appropriately weights them. Take some time to read up. You might learn something. Here's some more reading to help including some of the components used... https://hockey-graphs.com/2019/01/17/wins-above-replacement-the-process-part-2/

Surely your point is stronger than here's some points, QoC and age right? Surely you realize that 88% of Nurse's points over the last 2 seasons have primary contributions from McDavid or Draisaitl? Of course someone who complains about context would...

You misunderstood, perhaps I wasn't clear. When I was referring to "one number stats" I meant things like WAR, GAR, GSVA and the like. Stats that purport to distill a player's value down to a single number, i.e. 'player A is worth 2.3 while player B is worth 4.1'. They are far from perfect due to the lack of context in the inputs: the NHL either doesn't track or doesn't release a lot of data, unlike, say, baseball. The nature of the game is very different from baseball too.


Do they though? If so, how come there isn't one single generally accepted model that is used universally? There isn't because they all have a slightly different take on things, that's why. Have you done a study to see which advanced stats have the highest correlation with future events, i.e. predictive ability? I'd be interested in reading that. After all, that's the entire point of advanced stats, isn't it?

You also clearly missed my point in posting that table - it was a rebuttal to the aging curves study you posted, which was a dig at my statement that Nurse was just entering his prime at 25. If players peak at 22-23, how come all of the top defensemen scorers were above (most well above) that age?

I think that's actually why I prefer those models. They're trying to find value. The alternative (that you see here) is misapplying a bunch of component statistics to mean "value". Every sports' modeling is always growing and shifting as more data and deciphering of that data comes in. No model is the best...however all of those models do a significantly better job than any RAPM based metric or raw numbers, etc because they all go through the trouble of trying contextualize individual performance through a litany of factors. Separating an individual from the team in terms of performance still has a long ways to go, particularly defensively, but these still do it the best among the tools we have.




I mean there's no one single model in any sport. Everyone has slightly different weighting for different factors. But they're all kinda in the same ballpark. So much of the research is tracing a goal back from its origin. Eric Tulsky (now the mastermind behind the Hurricanes) was the first to link controlled zone entries with goals. And since then we've dissected that further and traced it back.

Try some of this stuff to occupy your time:

https://hockey-graphs.com/2020/05/13/using-sequences-for-analysis-expected-goals-contribution-and-more/

https://hockey-graphs.com/2019/10/24/lateral-puck-movement-in-the-nz/

https://hockey-graphs.com/2019/10/22/passing-clusters-a-framework-to-evaluate-a-teams-breakout/

Do you know what stat has the highest correlation with future scoring? P/GP. Raw.