NHL Projections 2021

[Melvin]

I am back and playing with hockey data again. I've been working on some stuff for the upcoming season and it doesn't quite fit anywhere in existing threads so here is a thread to bounce ideas off each other about doing some projections of the upcoming season.

Obviously, none of us know what the season is even going to look like, how many games will be played, if there's division re-alignment, and so on. I am ignoring all of that for now.

I am sticking my fingers in my ears and assuming an 82 game season with the normal divisions. LA LA LA LA LAAAA.

What I am doing is pretty rudimentary which makes it easier to discuss. Please view it as a work in progress!

Here is what I am starting with:

1) Take all of the rosters as of this weekend from CapFriendly.com. Obviously rosters are still in flux and I will update these as they go along. I will probably update them weekly unless something major happens, but for example I doubt Jayce Hawerlyck is going to move the needle all that much so he will go in with the weekly update (probably Saturday.) If the Canucks trade for McDavid or something I'll manually update it right away.

2) I had to do my best to project usage. This meant basically assigning for each player whether he will get 1st line ice time, 4th line ice time, 1st unit PP time, etc. And for goalies it's starter or backup (although some teams are 50/50.) This was very subjective although I did my best to try to have it make sense. I already know that I screwed a few of them up though so will be going back to re-assign a few players (mostly guys who kill a lot of penalties who I don't have on a PK unit.)

3) For skaters, I am using the xGF/60 and xGA/60 numbers from evolving-hockey.com. I have taken each of these metrics for each player over the last 3 years to project the same for next season, using a TOI-weighted average at 3-2-1 (so last season weighted 3x 2018.) I might have to bump that up to account for the shortened-season, not sure. I am then applying an age-adjustment based on this research, and then regressing towards the mean. I have split the factor in half to apply a + to the GF and a - to the GA. This is probably dumb.

I know there's an issue with using this metric, which is that it's tightly coupled with team-mates. I decided I am mostly OK with this, because most players have basically the same teammates. Like, Boeser's numbers might be inflated due to Pettersson... but he'll be playing with Petterssson again next year so I don't think it's a big problem? Does anyone have any thoughts on this? And for the player's switching teams, yes Nate Schmidt will probably suffer a bit due to playing for a worse team, but I'm not sure if it's a big enough problem to worry about.

Anyway, once I have this I can project team xGF/60 and xGA/60 using the projected figures for the individual players and estimated usage of said players. Oh yes, I forgot to mention that I am doing this individually for ES/PP/PK (for now just assuming 48 min ES, 6 PP and 6 PK.) See TODO(a)

4) After I have xGF/xGA for each team, I can do the goaltending. I explained this in another post but basically, I am then applying the goaltending by projecting the Goals Saved/60 of the two goalies and estimating the usage. This number is then added/subtracted from the GA column. So each team has a GF that assumes facing average goaltending and then a GA that accounts for the projected goaltending of the team. I hope that makes sense. I explained in more detail here.

5) I'm not factoring injuries at all in the sense of trying to project playing time, but, players who are currently significantly injured are not included. So Boston is missing Pastrnak and Marchand which is pretty amazing because they still project really well...

~~TO DO~~

a) I want to factor in penalties. My idea here is that it should be easy to project how often each player on the team takes or draws penalties and then I can estimate PP opportunities/PK opportunities. I have not done this yet, so each team is basically assuming the average of 3 mins/game on PP and PK.

b) Obviously, what I would like to do next is plug in the NHL schedule and run some simulations, but that won't be possible until we see the schedule and LA LAA LA ALAL.

c) Any other ideas?

OK, so here is what I have right now, with all of the above caveats that rosters are still changing, season still in flux, and I have some errors in my usage patterns like whoops I can't believe I forgot to put Colton Sceviour on the PK for the Penguins what was I thinking!!!

Since I can't run any sims without the schedule the best I can do is rank teams by their goal differential.

TEAM	TOT_xGF	TOT_xGA	GSAA	TOT_GA	Diff
DAL	192	180	22	158	34
VGK	210	184	6	177	33
CAR	220	192	1	191	29
BOS	185	172	8	164	21
N.J	194	186	12	174	20
T.B	196	179	2	177	19
PIT	198	184	3	181	17
CGY	204	191	3	188	16
STL	191	184	8	176	15
ARI	181	187	20	167	14
COL	190	186	10	176	14
TOR	208	193	-4	197	11
NSH	189	186	2	183	6
ANA	183	192	13	179	4
FLA	178	186	10	176	2
EDM	195	193	0	193	2
PHI	193	190	-1	192	1
CBJ	181	175	-5	180	1
WSH	192	193	0	193	-1
MTL	199	191	-10	202	-3
NYR	188	197	5	192	-4
WPG	189	200	6	194	-5
NYI	189	194	0	194	-5
MIN	178	173	-16	190	-12
L.A	180	187	-6	193	-13
OTT	182	184	-12	197	-15
S.J	190	188	-22	210	-20
BUF	179	188	-12	200	-21
CHI	184	198	-8	206	-22
VAN	182	195	-10	205	-23
DET	172	204	3	201	-29

[TBODY] [/TBODY]

The way to read the above is, for example Dallas have an expected goals-for of 192. They have an expected goals-allowed of 180, but then their goaltending is expected to save them 22 goals compared to expected, so the actual projected goals allowed is 158. Thus, their differential is 192-158 = 34.

OK, so the first thing that jumps out is the generally low numbers across the board. I can explain part of this which is just that I am projecting 82 regulation games so no OT or SO. I also wouldn't have any empty-netters really, but I don't think that fully explains it. I need to do look into this but I'm guessing it's something systemic that affects all teams equally as I doubt it's something that would greatly bias for or against particular teams. But not sure.

....and yeah, this doesn't look too good for Vancouver. Their ES scoring numbers are just...very bad, and the goaltending also projects to be bad. Sorry.

Any other surprises? Dallas comes out looking like they might not be the fluke that everyone thought, and Vegas is still very strong, as is Boston. New Jersey is a pretty big surprise I'd say. My numbers still really like Crawford and Blackwood. Any other surprises? I know I was bullish on MTL earlier but after running the numbers, maybe I had it wrong. Price does poorly with my goalie model though, so idk.

I will be tinkering with this over the coming days and weeks to take my mind off of *waves hand around.*
I welcome any thoughts/suggestions/ideas or to just share your own projections. Would love to compare notes if anyone else has tried anything! I can provide any data if you want to see any of the raw details as well.

Please be nice if I made some dumb errors; that is why I am posting this.

I ... really wasn't expecting the Canucks to look this bad when I started, you guys probably want to know what I did with the usage:

PlayerName	Pos	EV_Line	PP_Line	PK_Line
J.T. Miller	F	1	1	4
Bo Horvat	F	1	1	4
Elias Pettersson	F	1	1
Tyler Myers	D	1	2	2
Alex Edler	D	1	4	1
Braden Holtby	G	1
Nate Schmidt	D	2	2	1
Brock Boeser	F	2	1
Tanner Pearson	F	2	2	4
Jake Virtanen	F	2	4
Quinn Hughes	D	2	1
Thatcher Demko	G	2
Jordie Benn	D	3		2
Loui Eriksson	F	3		2
Tyler Motte	F	3		2
Brandon Sutter	F	3		1
Micheal Ferland	F	4	4
Antoine Roussel	F	4	2
Sven Baertschi	F	4	4
Adam Gaudette	F	4	2
Jay Beagle	F	4		1
Zack MacEwen	F	4

[TBODY] [/TBODY]

To read this, basically it's what you think: Under EV a "1" means I'm giving them 1st line icetime so around 16 ES minutes a game. Under PP and PK I have 3 groupings, "1" for 1st unit PP time which is something like twice as much as "2" for 2nd unit time, and then "4" for players who get just a smattering of PP/PK time, like 30 seconds per game or something.

I know we tend to think of Horvat as the 2C but he actually gets more minutes at ES than Pettersson. Anyway, I can definitely swap this around to what you guys think makes sense but I don't think it will have a significant effect.

For the goalies...I have Holtby getting 65% of the minutes but I have no idea if that's correct. I don't think any of us know right now and also it will certainly depend on performance. I might be better off going 50/50 here. Thoughts?

Click to read more...

 

[Melvin]

Updated rosters, changed a few usage errors, and made some minor tweaks to methodology.

RANK	Change	TEAM	TOT_xGF	TOT_xGA	GSAA	TOT_GA	Diff
1	-	DAL	198	188	23	165	33
2	-	VGK	211	193	4	189	22
3	+1	BOS	196	188	13	176	20
4	-1	CAR	212	197	3	194	18
5	+2	PIT	202	192	3	189	13
6	+2	CGY	200	195	8	187	13
7	+3	ARI	190	196	18	177	13
8	+1	STL	196	192	8	184	12
9	-3	T.B	199	189	1	188	11
10	-5	N.J	197	198	11	187	10
11	-	COL	191	194	9	184	7
12	+3	FLA	190	196	11	185	5
13	+3	EDM	200	201	4	197	3
14	+3	PHI	198	196	0	196	2
15	-1	ANA	194	197	4	193	1
16	+4	MTL	201	196	-5	201	0
17	+1	CBJ	190	187	-3	190	0
18	+4	WPG	197	207	10	197	0
19	-7	TOR	207	198	-9	208	-1
20	+3	NYI	195	195	-1	196	-1
21	-2	WSH	196	196	-1	197	-1
22	-9	NSH	199	194	-6	201	-2
23	-2	NYR	196	206	5	201	-5
24	+1	L.A	194	198	-5	203	-9
25	-1	MIN	184	181	-12	194	-10
26	+5	DET	184	204	6	198	-14
27	+3	VAN	191	199	-10	208	-17
28	+1	CHI	194	203	-8	211	-17
29	-3	OTT	196	201	-18	219	-23
30	-2	BUF	184	195	-15	210	-26
31	-4	S.J	199	200	-28	228	-29

[TBODY] [/TBODY]

Canucks looking better but still not great. Will be working on the penalty taking/drawing rates next.

Leivo leaving didn't affect anything as I didn't have him on the roster anyway.

 

[4Twenty]

This is great.

Even with Markstrom the team would still rank about 15th.

What do you think would be easier? To limit xGA or increase xGF.

I wonder why your model loves Dallas but not the Islanders. If Travis Green could get buy in to a more conservative system I think that’s the key way to improving the xGA performance and hopefully increases the goalie performance.

 

[Melvin]

This is great.

Even with Markstrom the team would still rank about 15th.

What do you think would be easier? To limit xGA or increase xGF.

I wonder why your model loves Dallas but not the Islanders. If Travis Green could get buy in to a more conservative system I think that’s the key way to improving the xGA performance and hopefully increases the goalie performance.

Defense... Barzal/Leddy/Lee all project poorly in xGA and are given a lot of minutes. The Dallas top guys all do quite well in limiting chances.

I'm not sure which would be "easier," but I bet improving xGA is cheaper.

 

[JAK]

I would like links to previous seasons' projections and actual results before I give any value to your wall of text.

While I applaud your dedication and time spent, it doesn't mean much more to me than just some fan typing up what they think.

This type of projection fits a lot better with baseball than hockey.

 

[Pastor Of Muppetz]

Melvin Canuck predictions:

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

 

[Peen]

Yeah take that, Melvin! Instead of 3rd last, they finished 9th last! Gotcha!

Benning definitely competing with the league's elite in 2017-2018 and 2018-2019 like he said he would! Gotcha Melvin!

 

[vanuck]

Great work. I was definitely surprised to see NJD rank so highly the first time.

Some things that I'm curious about:

1. What sort of changes did you make to your methodology for the second projection?
2. Do you apply aging curves to goalies too? I'd be surprised if there wasn't a bit of a dropoff for DAL considering how old Bishop and Khudobin are getting.
3. What about skaters who haven't played at least 3 seasons e.g. EP and Hughes?
4. How do you determine team xGF/60 and xGA/60? Are you taking a TOI-weighted average based on all the skaters' individual xG metrics?
5. Likewise, I also wonder if you did a weighted average - for, say, Canucks skaters using their penalty differentials in 2019-20 - whether it come out close to the number of penalties VAN actually took and drew last season.
6. Since you've already projected TOI/usage for what basically looks like 18-20 skaters on all 31 teams, do you have any plans to do this for individual point production?

 

[Fatass]

We are going to be a lottery team this coming season. I’d say we finish bottom 10 for sure, and maybe even bottom five, especially if there’s just a Canadian division. The only Canadian team we will finish higher than will be Ottawa.

 

[4Twenty]

Mods can you please threadban users who have no interest in discussing the content of the op but instead are here to complain about walls of text and how it’s not baseball. We’ve got enough mindless threads here.

Some of us really appreciate being able to read smart peoples analysis. Some want to protect their fearless leader at any turn.

Thanks in advance

 

[4Twenty]

@Melvin

Do you think your goaltending factor is too strong?

When I looked to last seasons expected goals numbers I was surprised the highest was only 216 but the lowest is 160. Your range doesn’t go that far. Is Detroit that much better this year or is your model?

 

[Melvin]

@Melvin

Do you think your goaltending factor is too strong?

When I looked to last seasons expected goals numbers I was surprised the highest was only 216 but the lowest is 160. Your range doesn’t go that far. Is Detroit that much better this year or is your model?

I know the goals numbers are too low across the board and am not sure what the problem is yet. There are a handful of reasons that I can think of but not quite enough to explain the big drop.

1. My system is assuming 82 60-minute games and projecting totals by splitting those 60 minutes into PP/PK/ES, but we know some games will go 65 minutes and also Shootouts often report +1 in the GF column for the winner, so add in all OT/SO goals. In 2018-19 (i.e. the last 82-game season) the Canucks had 8 of these.
2. My system is not accounting for EN goals although I guess technically the 5-on-6 would be part of the "PK" rates which isn't correct, since the scoring rate is much higher than typical PK, so add EN goals. Canucks had 11 of those in 2018-19
3. There are a myriad of other scoring situations which are not being accounted for correctly, like 5-on-3 which would have a very high scoring rate, but also 4-on-4, 4-on-3, etc. All have different scoring rates but are in such tiny sizes that I am not bothering with them now.

Once the schedule is released I'll start running simulations which will give us the OT/SO goals and maybe if I'm ambitious I'll bake something in for other scoring situations but it's probably not significant enough to worry about.

I don't know if that's enough to completely explain the low goal totals but it accounts for some % of it. As for the lack of range, the range of any projection system is never going to be as wide as reality because of regression towards the mean. Every year there will be some teams that are outliers who just have insanely good or bad seasons because of everything going right, but no projection system will ever project those. I think it's very healthy to see a tighter range as it shows that some regression is baked into it. It's like projecting the outcome of the Loui Eriksson contract, you can picture a best-case and a worst case but probably your projection will be something in the middle. The same is true here for each team, while it's possible to see a scenario where DET is as bad as last year, and it's also possible to see a scenario where they compete for the playoffs, your projection is going to target the average of outcomes for each team; hopefully that makes sense.

 

[Melvin]

Great work. I was definitely surprised to see NJD rank so highly the first time.

Some things that I'm curious about:

1. What sort of changes did you make to your methodology for the second projection?
2. Do you apply aging curves to goalies too? I'd be surprised if there wasn't a bit of a dropoff for DAL considering how old Bishop and Khudobin are getting.
3. What about skaters who haven't played at least 3 seasons e.g. EP and Hughes?
4. How do you determine team xGF/60 and xGA/60? Are you taking a TOI-weighted average based on all the skaters' individual xG metrics?
5. Likewise, I also wonder if you did a weighted average - for, say, Canucks skaters using their penalty differentials in 2019-20 - whether it come out close to the number of penalties VAN actually took and drew last season.
6. Since you've already projected TOI/usage for what basically looks like 18-20 skaters on all 31 teams, do you have any plans to do this for individual point production?

1. I tweaked some of the values that I used for the weightings. With my original projection I used the last 3 seasons, weighted at 3-2-1, which is to say that 2019-20 data was weighted three times as much as 2017-18 data, and the 2018-19 season weighted twice as much as 2017-18 . However, these 3-2-1 values were simply pulled from my behind, and I spent a lot of time this weekend researching past seasons to find the appropriate weightings. What I essentially found is that the most recent data should be weighted quite a bit higher than it was previously, closer to a 6-3-1 weighting. So basically teams were helped if they had key players who had very good recent seasons and not so good 2018 seasons, and it hurt teams if they had key players (*cough*Rinne*cough*) who had very bad recent seasons but good 2018 seasons. I also tweaked my regression amounts to, once again, be based on research and not just made up arbitrarily. Players are regressed towards the mean based on their TOI, so if a player has less TOI then he is regressed more towards the mean. This hurts players like Quinn Hughes who is probably regressed more than most fans would probably like because he only has 1 season worth of data, but it helps guys with poor numbers who played very little as their numbers are more regressed upwards. Meanwhile a player like Schmidt with a lot of TOI is hardly regressed at all. Anyway, the amount of regression applied, like I said, is based on research of prior seasons whereas in the first projection I just made something up (which honestly came really close.)

2. Yes I do. I was lazy and used the same aging curve as skaters. I should probably do a separate one for goalies, yeah.

3. I answered this a bit in [1] but basically their data is used but regressed towards the mean by a more significant factor due to the smaller sample. Having said that, it's based on total TOI between the 3 seasons and so for Petey I think the regression would still be fairly minor as even though he only played 2 seasons, he has a fairly significant amount of TOI over those 2 seasons.

4. Yes, basically I am assigning for each player 3 classes for "ES Line [1/2/3/4]," "PP [1/2/extra]" and "PK [1/2/extra]" So if you look at my second table in the OP, I have Schmidt as getting 2nd-pairing ES time, 2nd-unit PP time, and 1st-unit PK time. Then I am taking the average xGF/xGA of each "unit" and assigning them a % of the ice-time. So you can essentially take the average xGF for the players labeled "line 1" and that's the team's expected scoring rate for the 35% of ES time that I am giving them.

A lot of the assignments I made are completely arbitrary and I am more than willing to swap players around if we think they are going to get more or less ice time than I have them entered for. This was by far the most time-consuming and tedious part of this project! I had to go through all 31 teams and figure out how much ice time to give each player at ES, PP and PK. I would actually like to crowd-source this so I might build some kind of front end where people can vote on the correct usages. IDK.

5. Penalty taking/drawing rates are up next! I have done the initial work on this so will be sharing it soon. Currently the projections are just assuming the league average of 6 PP/PK minutes per game, but I would prefer to change these to better reflect the penalty taking/drawing rates of each team. Pettersson is one of the best players in the league at penalty +/- so this should help the Canucks some.

6. I hadn't planned on it, but it's definitely something I could do pretty easily, yeah. Good idea! Might give me something to do while I wait for the schedule to be released.

Thanks for the questions guys! I appreciate the support/feedback.

 

[Dana Murzyn]

Always a treat when you drop one of these. Thanks!

 

[MS]

Thanks for all your hard work on this. We're not always in the same camp on some of this stuff but the work you do is invaluable and I always learn something from it. Fascinated to see how this will play out.

I think the hardest thing in something like this is modelling goaltending because the position is so insanely volatile in today's NHL almost to the point of being like relievers in MLB. SJ is ranked last on your list largely because the recent track record of their goalies is very bad ... but that could flip in an instant or if it doesn't some mid-season Binnington will inevitably appear. And I'd be very surprised if that isn't a mid-pack team at worst when all is said and done.

My gut feeling is that your xGF/xGA models without GSAA will end up being closer to the actual results than your numbers including goaltending, so I'll be very curious to see if that's the case or not.

 

[Melvin]

Thanks for all your hard work on this. We're not always in the same camp on some of this stuff but the work you do is invaluable and I always learn something from it. Fascinated to see how this will play out.

I think the hardest thing in something like this is modelling goaltending because the position is so insanely volatile in today's NHL almost to the point of being like relievers in MLB. SJ is ranked last on your list largely because the recent track record of their goalies is very bad ... but that could flip in an instant or if it doesn't some mid-season Binnington will inevitably appear. And I'd be very surprised if that isn't a mid-pack team at worst when all is said and done.

My gut feeling is that your xGF/xGA models without GSAA will end up being closer to the actual results than your numbers including goaltending, so I'll be very curious to see if that's the case or not.

Thanks; for what it's worth I agree with you. Goaltending is voodoo and on top of that, it is the position that seems (anecdotally) to be the most team-dependent. We've seen so many times a goalie switches teams and their numbers swing dramatically, so Holtby for example could be anywhere from +15 to -15. Who the hell knows.

And yeah, the other thing you alluded to is that if teams get goaltending that's bad enough, they tend to fix it and it has a much bigger effect than "fixing" a horrible 4th liner for obvious reasons. If S.J's goalies stink that badly and they get replaced by some prospect who randomly comes in and puts up a 930 like every rookie goalie seems to these days, it would make a significant difference.

 

[Frankie Blueberries]

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

Really, you’re going to use a season that was suspended due to a global pandemic, at a time where the Canucks’ MVP and top defensive defenceman were slated to go on LTIR, as a good example of proving a projection wrong? I’m sure @Melvin should have factored in a pandemic into his projections, what an ass clown to have not foreseen that.

 

[Frankie Blueberries]

@Melvin how does substituting Schmidt in for Tanev change the projections? Are we expected to have a better or worse GA/GF?

 

[Melvin]

Really, you’re going to use a season that was suspended due to a global pandemic, at a time where the Canucks’ MVP and top defensive defenceman were slated to go on LTIR, as a good example of proving a projection wrong? I’m sure @Melvin should have factored in a pandemic into his projections, what an ass clown to have not foreseen that.

I should probably weigh last season a little bit less than what my researched weights suggests due to the smaller # of games, as all research is done based on "last-season" being 82 games. I don't know the best way to do that though, maybe just by weighing 2018-19 a little bit more to offset.

 

[Melvin]

@Melvin how does substituting Schmidt in for Tanev change the projections? Are we expected to have a better or worse GA/GF?

I don’t have a “with Tanev” projection to compare but it would certainly be better. Schmidt has the best numbers on the team right now, and although that is somewhat inflated likely due to playing on Vegas, it does nothing but help the Canucks.

 

[4Twenty]

The Canucks significantly outscored their xGF last year (224 actual vs 184 expected).

But their xGA and actual were much closer (214 actual and 206 expected).

Have the Canucks figured out a trick to beat xG or does xG need tinkering.

Last years numbers don’t flatter them at all but they looked likely to score and scored a lot (regular season). It’s weird that a top 10 scoring team could get viewed this poorly.

What does your model to actual look like for past seasons? Where is it going wrong?

 

[Melvin]

The Canucks significantly outscored their xGF last year (224 actual vs 184 expected).

But their xGA and actual were much closer (214 actual and 206 expected).

Have the Canucks figured out a trick to beat xG or does xG need tinkering.

Last years numbers don’t flatter them at all but they looked likely to score and scored a lot (regular season). It’s weird that a top 10 scoring team could get viewed this poorly.

What does your model to actual look like for past seasons? Where is it going wrong?

The Canucks at 5v5 scored 144 goals compared to 130 expected. I don't think that's a huge anomaly (10%) and honestly it lines up with the eye test. I saw a team that struggled a lot with 5-on-5 play and were frequently outplayed and out-shot by a significant margin, but had a very good PP and some opportunistic ("clutch") goals.

The power play was more significant, where they ranked 2nd in the NHL with 57 PP goals compared to just 42 expected. So, a 35% over-performance. My model of course does not project their PP to be anything special because it's based on xG, but if you believe that their PP is as good as the results it got last year, which was one of the top teams in the league, then that's worth upgrading them a bit. Sure looked like crap against Vegas though.

Overall I am surprised at how poorly the Canucks fare with this, but I am not surprised by the factors (poor ES play, downgrade in goaltending,) both of which are supported by most of the games I watched last season.

I mean 2 months into the season and the Canucks had scored something like 35% of their goals on the PP. It was completely ridiculous. I think a hot PP carried them for the first half and then mostly goaltending in the second half when the PP calmed down a bit, but that was just my observation and YMMV.

 

[Melvin]

Basically, you can boil this down to:

"The Canucks from last year, if they continue their poor ES play, their elite PP regresses a bit, and they get below-average goaltending instead of elite goaltending." I don't think many would disagree with them being a ~25th rank team if that happens, so a lot of it comes down to how much you think each of those three things will happen.

 

[4Twenty]

The Canucks at 5v5 scored 144 goals compared to 130 expected. I don't think that's a huge anomaly (10%) and honestly it lines up with the eye test. I saw a team that struggled a lot with 5-on-5 play and were frequently outplayed and out-shot by a significant margin, but had a very good PP and some opportunistic ("clutch") goals.

The power play was more significant, where they ranked 2nd in the NHL with 57 PP goals compared to just 42 expected. So, a 35% over-performance. My model of course does not project their PP to be anything special because it's based on xG, but if you believe that their PP is as good as the results it got last year, which was one of the top teams in the league, then that's worth upgrading them a bit. Sure looked like crap against Vegas though.

Overall I am surprised at how poorly the Canucks fare with this, but I am not surprised by the factors (poor ES play, downgrade in goaltending,) both of which are supported by most of the games I watched last season.

I mean 2 months into the season and the Canucks had scored something like 35% of their goals on the PP. It was completely ridiculous. I think a hot PP carried them for the first half and then mostly goaltending in the second half when the PP calmed down a bit, but that was just my observation and YMMV.

I mean my eye told me they got out played but I do think they have a good power play. Vegas neutralized it but Vegas has a strong PK.


I am also a bit surprised the Canucks fare poorly but not totally as results aren’t driving your model xG is and the Canucks tend to get outplayed in terms of possession and shot attempts.

I guess I just want the team to be better than it is and I see it in the middle more than the bottom but I could definitely see ways it finds itself near the bottom.

 

[Melvin]

As for how it performs compared to past seasons, I haven't run it for past seasons at the team level and it would be difficult to do so as I would have to input the rosters and usage for each player as-at the start of each season which would be incredibly labourious and error-prone. Some very obvious things though:

1) Injuries happen
2) Trades happen

Last year I did something sorta like this and IIRC nailed the Pacific Division ranking ..... as of December. After that it all flipped around. Obviously the teams roster is going to change a lot over the course of the year so even if you had a perfect projection system in-terms of the on-ice performance of the season-opening roster you aren't going to be able to account for all the injuries and transactions that change the roster throughout the course of the season.