Melvin's 2021-22 NHL Team Projections

[VanJack]

Count me as somewhat underwhelmed with the Vegas Golden Knights. I know they're coming off a string of seasons where they've gone to the conference and Stanley Cup finals.

But this team has been living in the 'present' ever since their expansion draft. They still don't have a true No. 1, impact center. There's virtually nothing coming up via the draft and some of their mid-range veterans are at the point where their production could start to dip. And this franchise will miss Marc Andre Fleury more than the they realize.

Not a chance in the world they win the President's Trophy as the team with the most points in the entire NHL.

 

[Melvin]

What would your rankings be if you removed goaltending?

We had this discussion last year about the volatility of goaltending and in your last year's rankings it's actually pretty interesting - you had Dallas #1 based on a huge goaltending boost that reversed this year but SJ right at the bottom based on a goaltending mess that did actually materialize.

Also how did you project LA's goaltending to get them at #28 at that position? I would be assuming that the excellent Cal Peterson would be getting 55-60 starts and be carrying the mail there as he did down the stretch last year.

I had a feeling you might ask and do have a separate non-goaltending list. I'll post it a bit later.

I plan on posting a few different rankings FWIW. I know some people might see this as "double-dipping," since, yes, in theory I could post 32! lists and one of them will be spot-on, but the point isn't for me to try to look smart (I think that's impossible -- any pre-season ranking that doesn't know about in-season trades or injuries is not going to be "correct,") but the point is to try a few different things and see, after the season, which approach worked best.

I think I have Petersen getting 65% of the minutes, but that does seem kinda strange.

 

[HockeyNightInAsia]

Well I like that you explained bits and pieces of what the model may not have taken into account (e.g. Tkachuk trade..... wrong first name btw). The Seattle problem is well noted so every reader can adjust to his own accord. And yes team changes, chemistry changes are difficult to factor in, not supposed to anyway. Carolina has interesting downside risk this year.

Things I find interesting -- 1. I bet few would rank Nashville this high (delighted Saros fan here) 2. Philly. Should Hart be ranked this bad? Really? Does the model not care about his rookie season? 3. Not surprised at WPG or WSH being ranked this low, but I would be quite interested into whether you get the timing of WSH right (and why such a big gap between WSH and PIT).

 

[Melvin]

i guess you could test the air of reality of the kraken ranking by seeing how this methodology would have projected the original vgk team.

i am in no way critical of this by the way. i give melvin a hard time often, but this is a lot of work and i think it produces a reasonable apples to apples comparison and a reality check for optimistic fan bases as to how much of a reach that optimism reflects. it is somewhat limited in reliability i think by the important last season of data being abnormal and perhaps less reliable, but it treats every team equally so a lot of that should come out in the wash.

bottom line, if you think the canucks are going to be markedly better than this ranking you are expecting that a lot of things are going to be better in the future than in the past. that's fine and certainly possible.

Doing the VGK is an interesting idea and definitely possible; I would just have to get some data that I dont' have currently as I didn't go back to 2016. But not a big deal. Would be very interesting actually.

And yes, thank you for recognizing the amount of work involved...between going through 800 players and assigning them to the correct teams, estimating their usage in terms of ES/PP/PK play, putting together the data, the formulas, identifying and fixing several bugs, and then writing this massive long post that included a summary analysis of each team, even taking the time to post each team's logo - - like the idea that I did all that just to "troll" is beyond stupid, especially since I have the Canucks at 22 which I don't think is even bad, really.

If I wanted to "troll" I could just spend 10 seconds on a list that has SEA at #1, VAN at #32 and a bunch of teams in random order in the middle. That would have taken 5 minutes instead of ~50 hours.

 

[Melvin]

Oh, and I think I have Petersen getting 65% of the minutes for L.A, but that does seem kinda strange. Maybe I made some sort of mistake though.

 

[Melvin]

Well I like that you explained bits and pieces of what the model may not have taken into account (e.g. Tkachuk trade..... wrong first name btw). The Seattle problem is well noted so every reader can adjust to his own accord. And yes team changes, chemistry changes are difficult to factor in, not supposed to anyway. Carolina has interesting downside risk this year.

Things I find interesting -- 1. I bet few would rank Nashville this high (delighted Saros fan here) 2. Philly. Should Hart be ranked this bad? Really? Does the model not care about his rookie season? 3. Not surprised at WPG or WSH being ranked this low, but I would be quite interested into whether you get the timing of WSH right (and why such a big gap between WSH and PIT).

The model does care about Hart's rookie season, and actually does not project him to be as bad. But also it *hates* Martin Jones, and has him even worse than Hart.

And thanks for catching the dumb name error.

 

[Tables of Stats]

Vegas has more players in that age range than most teams and could experience a San Jose-like fall into the abyss, but you don't tend to predict worst-case scenarios. That and player regression seems exceptionally difficult to model. We know the age range where players are expected to fall off, and could possibly even have some expected level of fall based on warning signs from prior seasons, but unless there are massive flashing red lights predicting an entire team to suffer from sudden age-related regression seems like something a model shouldn't be expected to predict.

 

[VanillaCoke]

VGK shouldn't miss MAF literally at all.

 

[Melvin]

Vegas has more players in that age range than most teams and could experience a San Jose-like fall into the abyss, but you don't tend to predict worst-case scenarios. That and player regression seems exceptionally difficult to model. We know the age range where players are expected to fall off, and could possibly even have some expected level of fall based on warning signs from prior seasons, but unless there are massive flashing red lights predicting an entire team to suffer from sudden age-related regression seems like something a model shouldn't be expected to predict.

I do include an age adjustment, but yeah, you can’t really predict the cliff falls and any model that tries to is going to be more wrong than useful. At the most extreme you expect some really old players to be 5% worse than they were last year, that’s about it.

 

[Peen]

I rate our es offense higher than Melvin’s model which puts them - in my mind - higher up in the bubble team list.

Also, every public analytics model has had Seattle fairly high and had to adjust for it.

I don’t see them as that good.

 

[Melvin]

I rate our es offense higher than Melvin’s model which puts them - in my mind - higher up in the bubble team list.

Also, every public analytics model has had Seattle fairly high and had to adjust for it.

I don’t see them as that good.

what do you mean adjust for it?

 

[Peen]

what do you mean adjust for it?

I’m not sure what they did to adjust. I remember dom lucsycsysysysnzn said after he initially modelled everything that Seattle is inflated and he would have to adjust some things. Let me see if I can dig it up.

 

[Melvin]

I’m not sure what they did to adjust. I remember dom lucsycsysysysnzn said after he initially modelled everything that Seattle is inflated and he would have to adjust some things. Let me see if I can dig it up.

that would be interesting thanks

 

[m9]

People predict Vegas to be bad every year since Year 1 and then eventually they'll be right and we'll see the "told ya!" everywhere.

They've been consistently excellent and while I do think people forget just how close things were in the Pacific in 2019-20, they should still win the division while piling up points vs these awful defenses.

 

[Peen]

I can also already predict that posters will neglect that this isn’t with simulations of the schedule being run

 

[flying v 604]

Using last years weird covid season results as the 'gospel' is folly..IMO..Last season was a 'one off'.

It makes no aense. I only read the writeup on the Nucks but those are some really bad takes.
The biggest positive change so far is the added depth which should really improve the team 5 on 5, the forward group has been upgraded throughout and will have 4-5 new faces all potential upgrades that push the players that might be slotted to high down to a more suitable role.
The goaltending could end up being in the top 3 in the West and the D while not elite is much much deeper and when you factor in Quinn was only minus 1 in 35ish games with Hamnic as opposed to minus 22 when he was with Benn and Myes who the former was terrible and the latter just isn't a good fit.
OEL will also allow Quinn to have easier minutes while a guy like Poolman who Shaw asked for specifically, is a better fit than Nate was but admittedly not as impactful.
Maybe I'm missing something but how can anything that happened last season especially a team like Vncouver who got boned with the schedule and Covid, while having 8+ new regulars and 18 new faces in total be used as any type of prediction for this season?

 

[Melvin]

It makes no aense. I only read the writeup on the Nucks but those are some really bad takes.
The biggest positive change so far is the added depth which should really improve the team 5 on 5, the forward group has been upgraded throughout and will have 4-5 new faces all potential upgrades that push the players that might be slotted to high down to a more suitable role.
The goaltending could end up being in the top 3 in the West and the D while not elite is much much deeper and when you factor in Quinn was only minus 1 in 35ish games with Hamnic as opposed to minus 22 when he was with Benn and Myes who the former was terrible and the latter just isn't a good fit.
OEL will also allow Quinn to have easier minutes while a guy like Poolman who Shaw asked for specifically, is a better fit than Nate was but admittedly not as impactful.
Maybe I'm missing something but how can anything that happened last season especially a team like Vncouver who got boned with the schedule and Covid, while having 8+ new regulars and 18 new faces in total be used as any type of prediction for this season?

because it’s not, pretty much.

what you’re missing is literally the entire way that it works. Feel free to read the post if you want to understand how it works, or just continue to be baffled.

the team results last season are not taken into consideration.

Still, they are projected at 22/32 after finishing 24/31 last year. That is improvement.

 

[Tables of Stats]

Maybe I'm missing something but how can anything that happened last season especially a team like Vncouver who got boned with the schedule and Covid, while having 8+ new regulars and 18 new faces in total be used as any type of prediction for this season?

We used 29 skaters the season before as well and based on the past few seasons expecting us to be significantly better than 20th going into the season is optimistic.

 

[Melvin]

We used 29 skaters the season before as well and based on the past few seasons expecting us to be significantly better than 20th going into the season is optimistic.

I would sincerely like it if these people would tel me based on their own critical analysis where they have the Canucks ranked. Is it 20? Is it 10? Top 5? At least I’m willing to put it out there.

To the people who think I’m missing the mark, I’d like to know by how much you think I’m missing. You think the Canucks will be 18? A marginal difference not worth complaining about? You think they’ll be in the top 10 teams in the league? What?

 

[bossram]

Thanks for doing this again @Melvin!

People are sweating the actual order of teams waaaay too much. A lot of the teams are separated by like 1 or 2 goals in goal differential. In terms of the major groupings of teams, it's pretty close to where I'd slot them. Honestly only St. Louis looks a little bizarre to me, but they were pretty bad at even strength last year. I expect Washington finish higher but I always go into the year not really rating them.

Get over the Seattle thing guys. They will be Actually Good. Very good goaltending and defending. They'll get goals by committee.

And the Canucks are in the mediocre middle tier. That's what they are. Where do you think they should be? 17th?

 

[Tables of Stats]

I would sincerely like it if these people would tel me based on their own critical analysis where they have the Canucks ranked. Is it 20? Is it 10? Top 5? At least I’m willing to put it out there.

To the people who think I’m missing the mark, I’d like to know by how much you think I’m missing. You think the Canucks will be 18? A marginal difference not worth complaining about? You think they’ll be in the top 10 teams in the league? What?

I'm going into your prediction expecting a +/-4 position ranking for every team, with a few outliers falling even further outside of the range. Would it really shock anybody if the Canucks had injury issues in January/February and finished 26th after a decent start or played well and finished at 18th instead? If they really do well they could sneak into a wildcard spot at 18th but this isn't even a surefire bubble team unless a lot of things go well.

 

[flying v 604]

because it’s not, pretty much.

what you’re missing is literally the entire way that it works. Feel free to read the post if you want to understand how it works, or just continue to be baffled.

the team results last season are not taken into consideration.

Still, they are projected at 22/32 after finishing 24/31 last year. That is improvement.

So if nothing that happened last year matters how can someone look at the roster and not see a team that on paper should easily be top 3 in the pacific based on the the amount of moves made that addressed numerous issues?
The D is a question mark I admit but I like the balance and the depth and we don't need Islander type effectiveness IMO because I see a forward group that again on paper with the additions of Garland Hogs, Dickenson, Highmore, hopefully a full season of Motte while unloading the deadweight a massive upgrade. We also have some young kids in Rathbone, Joulevi, Jonah, Lockwood who hopefully not only make the team but are upgrades on those they are replacing.

For basically the last 20 games the team played with as many as nine regulars missing and the players that filled in were waiver wire fodder and a few kids thrown into an impossible situation.
I also think the goaltending has a chance to be a huge advantage for us with the cramped schedule tandems will be huge and compare ours to the teams were fighting and we have a big edge on all of them.

I apologize if I misunderstood your model I just think on paper which right now is the only thing to go off of we improved in every aspect while the Oilers, Vegas and Calgary all got worse or didn't do enough overall.
Vegas lost their vezina goalie and Lehner and whoever is a downgrade IMO, the Oilers have a worse Dcore and are relying on a 40 year-old in an Olympic shortened season.
The Flames did dick to address their needs in net (backup) and not a big fan of Zadorov.

You obviously don't see it like that but I will say I do think people who aren't fans of Benning seem to be going out of their way to cling to the hate instead of looking at the whole picture that includes how we'll he stocked Abby which will obviously be huge when injuries happen and the fact that not only is the bottom six signed to better cap% contracts but it's also looking like a strength not a weakness and the flexibility Green has to work with.

 

[Paulinbc]

It's way early, and I never have the time to put in effort into these things but what the hell, here's my initial thoughts on rankings

1. Colorado
2. Vegas
3. Tampa
4. Islanders
5. Carolina
6. Toronto
7. Edmonton
8. Boston
9. Florida
10. Vancouver
11. Winnipeg
12. Dallas
13. Seattle
14. Jersey
15. Pittsburgh
16. Rangers
17. Calgary
18. St. Louis
19. Nashville
20. LA
21. Montreal
22. Chicago
23. Minnesota
24. Philly
25. Washington
26. San Jose
27. Detroit
28. Ottawa
29. Columbus
30. Anaheim
31. Arizona
32. Buffalo

Any questions, I'll be happy to answer, but like I said, we have 2 months to go.

 

[Tables of Stats]

So if nothing that happened last year matters how can someone look at the roster and not see a team that on paper should easily be top 3 in the pacific based on the the amount of moves made that addressed numerous issues?
The D is a question mark I admit but I like the balance and the depth and we don't need Islander type effectiveness IMO because I see a forward group that again on paper with the additions of Garland Hogs, Dickenson, Highmore, hopefully a full season of Motte while unloading the deadweight a massive upgrade. We also have some young kids in Rathbone, Joulevi, Jonah, Lockwood who hopefully not only make the team but are upgrades on those they are replacing.

So what exactly is your prediction for the team, you've said you see us being 3rd in the division (on paper) but where does that rank in the league? Are you picturing the Canucks as a surefire 16th or better team, 18th and making the playoffs due to a weak division, 10th and severely underrated by everybody?

 

[Pastor Of Muppetz]

It makes no aense. I only read the writeup on the Nucks but those are some really bad takes.
The biggest positive change so far is the added depth which should really improve the team 5 on 5, the forward group has been upgraded throughout and will have 4-5 new faces all potential upgrades that push the players that might be slotted to high down to a more suitable role.
The goaltending could end up being in the top 3 in the West and the D while not elite is much much deeper and when you factor in Quinn was only minus 1 in 35ish games with Hamnic as opposed to minus 22 when he was with Benn and Myes who the former was terrible and the latter just isn't a good fit.
OEL will also allow Quinn to have easier minutes while a guy like Poolman who Shaw asked for specifically, is a better fit than Nate was but admittedly not as impactful.
Maybe I'm missing something but how can anything that happened last season especially a team like Vncouver who got boned with the schedule and Covid, while having 8+ new regulars and 18 new faces in total be used as any type of prediction for this season?

Here's some of Melvins previous projections.

2018-19
"Amazingly, the 218 goals we got last year was our highest total in 3 years, and an improvement of 40 over 16-17. The 21 goals from the D is in line with the season before (22) and the season before that (23.) Expecting anything different wouldn't make much sense.

My guesses:

For = 200
Allowed = 260
Points = 69

Finish: Probably 2nd or 3rd last, win a lottery spot and have everyone congratulate each other on another job well done."..Sep, 2018
........

"I'm honestly thinking this might be the worst season in our history.

I predict we don't win a game in regulation until November.

Benning fired in January." Sep,2018

2019-20
"I'm not really seeing it. I think with luck they could maybe get to 85 points. I could easily see it being closer to 70.This is just a very bad team. I'd guess 77 points with standard luck/injuries and nowhere near playoffs"...2019​