Around the League Thread | Post Purge

[ManVanFan]

the people on twitter who make legitimately good analytics models tend to actually get hired by nhl teams

We should listen to the people on public forums that say that Ben Hutton is a legit top 4 D though.

 

[vadim sharifijanov]

JFresh cards are basically a slightly more advanced version of using +/- to determine if a player is good or bad. They're absolutely worthless.

so i have no idea what these jfresh cards i sometimes see referred to here are. but in my mind i always imagined them to be like the hello fresh 10 free meals promotional fliers i can constantly get in mailbox, maybe crossed with handbills for the fresh slices you see in every skytrain station

so yeah, junk and a massive waste of paper and ink (but i imagine these aren’t literal cards right?)

 

[sandwichbird2023]

I’ve known for like the entirety of my life posting on hf that if montreal ever was good they’d be insufferable and i always found them to be the most out to lunch on the main boards

Demidov better be the goat because they’ve built up an insane hype train. I see them hyping up practice videos of him shooting on empty nets dude

To be fair to them, they were right about Hutson.

Demidov is also insanely skilled, so there is some justification in his hype. It isn't like they are hyping some random 4th round pick or something.

 

[Bleach Clean]

Are you saying that the people that run multi million dollar sports franchises are dumb?

What position do you hold in a multi million dollar sports franchise?

MS has a history of dismissing or mocking event tracking for its interpretation. The data is just the data, and has value if used correctly. Throwing out the cards and their micro data breakdowns is 'baby out with the bathwater' territory.

That said, no one needs to hold a position in an NHL office to make a cogent argument for/against the use of said data.

 

[MS]

Are you saying that the people that run multi million dollar sports franchises are dumb?

What position do you hold in a multi million dollar sports franchise?

Data is a useful tool, if used *as a tool* by people who understand the context in which those numbers were generated.

I use these numbers, too. Sometimes they tell you something. Sometimes they don't. Mostly what they tell you - just like +/- - is only how effectively they played the role they were given.

But if you try to present this stuff as evidence of how good a player is ... it's rubbish. It's looking at Marek Malik going +35 and saying he's better than Mattias Ohlund at +11.

MS has a history of dismissing or mocking event tracking for its interpretation. The data is just the data, and has value if used correctly. Throwing out the cards and their micro data breakdowns is 'baby out with the bathwater' territory.

That said, no one needs to hold a position in an NHL office to make a cogent argument for/against the use of said data.

Yes.

But then when you assign a 'WAR %' to that raw data to imply that the data shows how good the player is relative to his peers ... it's garbage.

 

[vancityluongo]

Yes.

But then when you assign a 'WAR %' to that raw data to imply that the data shows how good the player is relative to his peers ... it's garbage.

can't say i follow too closely what jfresh is doing these days, but is he really still doing a version of this?

when those cards first started getting produced 4-5 years ago, they had guys like patrick weircioch and marty marincin at 80% WAR because of exactly what you're talking about, and really dumb people on twitter would cite those cards as if they just discovered a future hhof'er in a snow storm.

surely with the huge advance in publicly available data since then, that's not what they're doing anymore?

i pay for the athletic, so i'll use dom's, but this is his methodology:

That value is according to Net Rating, our all-in-one player value stat based on each player’s Offensive and Defensive Rating. The card showcases all the stats that determine how strong each player’s ratings are on and off the puck. It’s a weighted combination of their production (goals, assists, expected goals, blocks, penalty differential) and their play-driving (on-ice expected and actual goal stats), adjusted for difficulty (quality of competition and teammates). The cards also show each player’s market value based on total contributions.

The cards convey how a player is performing this season and, more importantly, how he’s expected to finish the season. Each card shows a player’s current performance in each category (pro-rated to 82 games), as well as an end-of-season forecast based on projections for the remainder of the season. The projections are based on a regressed weighted average of the last three seasons where the most recent season carries the most weight and players with larger sample sizes are regressed less.

https://www.nytimes.com/athletic/6048288/2025/03/09/nhl-player-cards-pacific/

if we look at the canucks, this very much aligns (at least at a high level) with how the season has gone by my eye test. there are obvious nitpicks (hronek and blueger feel "low", ep40 and hoglander feel "high") but there's a lot more here than just a glorified +/-

 

[SeawaterOnIce]

Maple Leafs vs Senators. It will be interesting to see how Travis Green coaches in the actual playoffs. My gut is saying the Leafs makes quick work of the Sens in 5 or 6.

Oilers vs Kings is brutally painful to witness again.



This is why I cringed at that Trenin cheapshot.

 

[Ernie]

I agree with @MS here - most of the microstats stuff is pure garbage. You don't have to get far into into a 1st year stats course to understand that the sample sizes are just too small. Then they try to massage the data in ways that are even more questionable.

It's hilarious because the whole point of the advanced stats movement was to expand sample sizes using shot attempts instead of just shots and goals. Using shot location to estimate xgoals is a useful advancement as well, but if you're making conclusions with under 200 data points, you're doing it wrong.

Maybe one day we will get access to the full suite of tracking data and will be able to have more fine tuned analysis, but for now, on a game by game basis the eye test is more valuable. For larger stretches, advanced stats can definitely help remove biases from the eye test and that is their primary usefulness.

 

[Reverend Mayhem]

Just saw someone on Reddit claim the 2010 Flyers weren’t “real contenders” or trying to win the Cup. I feel like that was based solely on their finish, but couldn’t be further from the truth. It’s like saying the Kings weren’t going for it in 2012 because they had a disappointing regular season.

I just feel old idk

 

[Vector]

 

[Hodgy]

 

[Vector]

 

[Burke's Evil Spirit]

Couture recently removed all references to the San Jose Sharks from his Instagram (updated bio + deleted pics). It seems like the Sharks strongarmed him into this.

Couldn't happen to a nicer guy.

 

[Hodgy]

I will always remember Couture for that time he accidentally posted that porn link on his Twitter, or whatever it was. I think he then claimed someone hacked his account, or took his phone, only to post the porn video. LOL.

 

[StreetHawk]

Chalk up another late 30's player who ends his career on LTIR.

 

[Mr. Canucklehead]

Always thought Logan Couture looked like a Value Village Sidney Crosby.

Good player, but I'll mostly remember him for being a huge flopper and diver. Still a crappy way for his career to end, hope he has good health.

 

[bandwagonesque]

I will always remember Couture for that time he accidentally posted that porn link on his Twitter, or whatever it was. I think he then claimed someone hacked his account, or took his phone, only to post the porn video. LOL.

I think the only direction to go with this is to keep posting porn links around the clock for a week while claiming you're frantically trying to find the hacker

 

[Hodgy]

I think the only direction to go with this is to keep posting porn links around the clock for a week while claiming you're frantically trying to find the hacker

Yes, he didn't lean into it, that's for sure.

 

[rypper]

I wonder if Logan wanted to be the one to share that news...

 

[Kickassguy]

Man, Demidov couldn't have asked for a better start in Montreal. A goal and an assist already, flashing nice skill on both goals.

 

[Peen]

Man, Demidov couldn't have asked for a better start in Montreal. A goal and an assist already, flashing nice skill on both goals.

Rookie Pettersson aura in this first period

 

[Knight53]

Demidov looks like a instant superstar. What a quick and fast rebuild for Montreal. Very similar to ours expect they got their superstar winger where as we passed on ours (M. Tkachuk). Also, have way better long term thinking management. They probably win a cup within the next 7 years.

Hughes - Hutson
Pettersson - Suzuki
Boeser - Caufield
Demko - Fowler/Dobes
Miller - Slafkovsky
Tkachuk - Demidov
D-Petey - Guhle
Willander - Reinbacher
Lekkerimaki - Hage

God I can’t wait until we get a proper Rebuild with a new owner.

 

[LemonSauceD]

damn, Soucy lost the sauce

View attachment 1012379

Soucy was good for like 25 games where he was injured in between twice in his first season. He was awful for 80% of his time here.

 

[StreetHawk]

Let's pump the brakes on the Habs.
Dcore, has some work to do. Guhle, needs to shake the injury bug. 2/3 seasons in the NHL missing 1/3 of the season or more so far. Have to stay healthy for them. Reinbacher, coming off a serious knee injury which cost him half the season. Likely pushes him back. X, Struble, have they cemented themselves as solid 3rd pairing options?

 

[ephmrl]

Demidov looks like a instant superstar. What a quick and fast rebuild for Montreal. Very similar to ours expect they got their superstar winger where as we passed on ours (M. Tkachuk). Also, have way better long term thinking management. They probably win a cup within the next 7 years.

Hughes - Hutson
Pettersson - Suzuki
Boeser - Caufield
Demko - Fowler/Dobes
Miller - Slafkovsky
Tkachuk - Demidov
D-Petey - Guhle
Willander - Reinbacher
Lekkerimaki - Hage

God I can’t wait until we get a proper Rebuild with a new owner.

We have one of the top 3 players in the sport in their prime and people are still craving rebuilds.

Even if we did a full tank and rebuild, there's no guarantees we'll draft another player like Hughes.