Mastodon
@Vancouver Canucks

Alphabet soup has the Canucks at 3.8% (10th best) to win the cup.



[Link to the article](https://theathletic.com/5330653/2024/03/12/trade-deadline-nhl-team-playoffs/?source=user_shared_article NHL trade deadline aftermath: Ranking where each team stands in the Stanley Cup race) At this point I think it’s just rage bait by a worthless hack.

by jafahhhhhhhhhhhhh

18 Comments

  1. jafahhhhhhhhhhhhh

    It’s all good, Hughes is in contention for the Norris after being labeled 3B… Calling the Canucks, the team that’s currently 1st in the West, a Dark Horse team is just the reverse jinx and endorsement the team needs to win it all.

  2. Twinbladey

    I really doubt most analysts are doing anything but underrating our chances at making it out of the Conference. Call it media bias, call it a hate-on, call it whatever, I just dont see how a healthy Canucks team come playoff time isnt a candidate to fuck shit up across a collection of 7 game series.

  3. bitter-pickles

    Typically the first few prediction articles after the deadline skew heavily towards the most active deadline teams. Without being able to see the article am I wrong in this assumption?

  4. carry-on_replacement

    I don’t think it’s an entirely unfair assessment to be honest. Our season has gone relatively injury free and when they’ve struck, even a small loss like Joshua meant a whole half a month of bad performances. We don’t have the kind of depth to make up for losing middle-6 type players and that’s not even considering players like Miller, Petey, Hughes, Hronek, and Demko. Most true contending teams have the depth to at least make up for losing their first pairs. or 1C 2C.

    Now if you told me, say, the leafs have a better chance of winning it then yeah, it’s stupid.

  5. CrabBeanie

    I imagine a guy taking off his blindfold to see the letter he got on his spoon.

  6. It makes sense from an odds standpoint. Pacific division is the hardest road to the final which is generally why we have lower odds. You have to beat multiple true contenders in division just to get to the Conference Final where you get the pleasure of playing the Avs or Stars or Jets who I believe are legitimately better than literally any team in the East.

    The Oilers only escape this by being an analytical model wet dream.

  7. poolside123

    The only thing is trust is moneypuck. And my own eyes.

  8. Admirable-Sound5198

    Anyone who puts the hurricanes as #1 massively underrates the importance of goaltending in the playoffs lol…. Virtually every athletic prediction for anything seems to ignore the most important position in the sport

  9. StarkStorm

    Dom L is an absolute idiot and his models are not predictive. Don’t believe? Look backwards.

    I cancelled my sub and they gave me an offer for 20 bucks…I still regret subscribing.

    I unsubscribed

  10. Having hoggy, a 20ES goal scorer at 0 and Joshua at -1 tells me all I need to know about this model tbh. 

  11. ClassicChrisstopher

    Good, I would rather have no expectations then be a favourite.

  12. To be fair, teams with very little experience together in the playoffs end up winning the Cup. It’s kind of a reasonable take. I would personally put us somewhere in the range around 5-10%, 6.25% being the average for any of the 16 teams.

  13. notarealredditor69

    See what happens is at the beginning of the season these guys have their takes on who is going to be the best , or the worst etc. things all based on the end of last season plus their opinion of the moves done by management. So when this doesn’t pan out they then spend the time trying to justify their take and why the reality of what we are seeing is not correct. For the Canucks this year it was the PDO that was the scapegoat for why they were wrong. So even now they still have this bias in their head that the Canucks or whoever aren’t actually as good or as bad as they are, because they need their model to be correct.

    The funniest thing about this is that usually their model from last year wasn’t the same as it was at the beginning of the year. So it’s like they have this moving moving target , it’s all about justifying their own existence as hockey “analysts”.

  14. ktbffhctid

    This is fine. We are building a foundation here. Finally. I would rather we win it all but if we fall short and the core gets more experience, it bodes well for the future.

    I begrudgingly have to credit old Blueberries for bringing in Rutherford to fix this mess.

  15. CulturalLevel3189

    This model is wack lmao, way too much weight on individual players. This sport is not a sport in which 1 player can make all the difference. Sometimes 1 player can have a massive impact, and 9/10 times it’s the goalie. Mackinnon, for example, is a phenomenal player, but if a team shuts him down, his top percentile 25 offensive rating means nothing. There’s a reason McDavid hasn’t seen much playoff success yet, cause being the best player in the world means nothing come playoffs when the rest of the team sucks.

  16. Alpacaduck

    Honestly I think it’s pretty fair.

    Analytics undervalue goalie metrics and overvalue individual skaters (see: Edmonton) and strength of schedule (see: Carolina, although they’re legitimately good).

    Once Carolina makes it to the 2nd round by beating Philly or Detroit (yeah, their path is that weak), I’m pretty sure the odds will adjust.

    Likewise if we beat Vegas or Edmonton or LA, our odds will likewise adjust.

  17. Canucks are one of five teams that could be considered contenders in the West, and that’s not counting Vegas. 3.8% sounds reasonable to me.

    But yes, let’s continue to desperately seek validation from one member of the media.

Write A Comment