Sabres Previews

[GASabresIUFAN]

Well here is the first Preview of the upcoming season.  Sorry it's from the Athletic and it predicts we'll be the 4th worst team in the NHL.  Don't worry the pitchforks are already out. Dom Luszczyzsyn believes that Sabres will finish with 76 pts next season and actually be worse than the Flyers, Sens and Red Wings among 28 others.  Only Chicago, Mon and AZ project lower.  He doesn't even list JJP as a possible Sabres next season.  He is even using player data which includes data points from the RK era to make his projections.  

Obviously is much easier to predict the performance using analytic tools with veteran players coached by veteran coaches, and he is correct in that the Sabres have so many young players that you don't know how they will grow and develop.  Still one needs a little common sense when the data doesn't match the eye test.  This team turned the corner last year.  If this team fails to score 250+ and move into the mid 80's pt wise, I'll be very disappointed and so will management and the players.

This is his one saving grace:

Quote

 

The best case

While everyone was paying attention to the improved Senators and Red Wings thanks to their offseason adds, the Sabres are the true disruptors of the Atlantic thanks to the growth of their young core. Asking for the playoffs may still be a stretch, but making some noise in the East and showing legitimate strides is a win for this squad.

 

HIs best case scenario IMHO has about a 75% chance of occurring.  His worst case 5%.

Quote

 

The worst case

Thompson’s breakout year was a fluke, Dahlin’s development stalls, and that trickles down to negatively impact players like Power and Cozens sending the Sabres back into the chaos they finally seem to have broken out of.

 

 

https://theathletic.com/3606971/2022/09/20/buffalo-sabres-2022-23-season-preview/

Anyway, If you find other previews please post them here so that we can bash them as well.

Edited September 20, 2022 by GASabresIUFAN

[steveoath]

The flyers are absolutely forked this year.  Couturier out until 2nd half of season (maybe) Ellis not looking like he's playing AND they Risto. 

[DarthEbriate]

The Athletic normally does a decent job with their analytical pieces, but this Sabres preview reeks of halfassery and dianoga poopoo. Using Krueger-season data on players like Tage, Mitts, and especially Skinner seems like a horrible, horrible way to prognosticate anything (unless they're all silently putting lots of money on the Sabres). Including Levi (unsigned) as an option? Nope, that's lazy journalism. These lines about VO: "[Olofsson's] stock has dropped off in recent seasons. At five-on-five his game has fallen off and he’s turned into a bit of a power play specialist thanks to his efficient shot." Every word that they just said is wrong. It's pretty much the opposite of his career trajectory. And no mention that for half of last season VO had a busted wrist/whatever and couldn't shoot. In the 3 months where he was able to shoot his stats line read: 36 GP, 18-16-34 +9. Extrapolate it to an entire season and VO is Ottawa's new toy: DeBrincat.

This will all come down to goaltending though. Ottawa (in addition to sexy UFA forward names) jettisoned their crap goalie (Murray), kept a guy who had a 1B-level season last year, and added a consistent 1B-level goalie in Talbot. They theoretically have two 1Bs... and that's healthy (good SW reference, me) goaltending. Buffalo kept a 2 (Anderson) and added a potential 1B in Comrie. That's the difference in the rosters as I see it.

 

[dudacek]

Meh, nothing to get worked up about.

They have a model, their model kinda works, this is what it spits out about the Sabres.

Their model isn’t really built to account for a bunch of young players dominating the makeup of a roster, or accelerating their growth at the same time. And no broad-based model is going to anticipate the difference the eye test has shown between Krueger and Granato. It’s built to predict based on broad strokes, not outliers.

I look forward to the chance I’ll be reading about ^^^ in next year’s preview when they explain how they missed on the Sabres.

As many have said elsewhere, the general hockey world will continue to expect the Sabres to be bad until they aren’t. And with good reason.

Models aren’t always right, that’s why we play the games.

Edited September 21, 2022 by dudacek

[Thorner]

We’ve seen it with many a rebuild. 

There’s always an “ahead of schedule” year. 

The first year we are “good”, the portents won’t have foretold it, this much is known.

[sabremike]

My favorite part is the idea that we simultaneously have the deepest and best prospect pool in the sport by a country mile and must finish in the high lottery again to add to that pool. Dom is really the most crude stereotype of fancystat types who think the sport is just a math problem to be solved.

[JoeSchmoe]

GUARANTEED - If the Sabres do as well as what we're thinking they will, the know-it-alls at the Athletic et al will come up with the numbers to show how they saw this coming all along.

Edited September 21, 2022 by JoeSchmoe

[LGR4GM]

The other flaw with using 3 years of data is 3 years ago was covids start basically. So a chunk of your data is flawed by a known variable you can't account for. 

I get why you would use 3 years of data because it probably helps find a mean level but between Krueger, covid, and unpredictability of the young guns, that's a hard ask for a model. Intriguing read but if your model doesn't match the eye test you gotta question both. We know that in the last 28 games when finally healthy the eye test and stats align. 

Final thought the article implies Anderson is in line to be the starter but we know he's not. Comrie is the starter and depending on how he does, really tells us what our season will be. 

We won't be worse than the flyers, that's laughable.

Edited September 21, 2022 by LGR4GM

[PerreaultForever]

9 hours ago, steveoath said:

The flyers are absolutely forked this year.  Couturier out until 2nd half of season (maybe) Ellis not looking like he's playing AND they Risto. 

Ya, I don't see how the Flyers can be better than us no matter what happens. The rankings here must have some sort of blind Tortorella miracle belief or something. They are also obviously assuming the goaltending here will be bad (which is highly possible, but not a certainty). 

Aside from goaltending, the season imo hinges a lot on how much of a leap forward Cozens and/or Mitts can make and how much immediate impact Power has. If Cozens takes a Thompson style leap forward we will exceed expectations. If he doesn't, well we could easily be in the bottom 5. 

[mjd1001]

15 hours ago, steveoath said:

The flyers are absolutely forked this year.  Couturier out until 2nd half of season (maybe) Ellis not looking like he's playing AND they Risto. 

I don't see how Philly has any chance of getting anywhere other than one of the worst teams in the league and worse than the Sabres.  The only way I see the Sabres with less points than Philly is a combination of big time injuries in Buffalo, Philly having some unknown emerge into a star, AND a lot of luck on their part also.

[Doohicksie]

15 hours ago, steveoath said:

Couturier out until 2nd half of season (maybe) Ellis not looking like he's playing AND they Risto. 

If you're using Risto as a verb, I like it.  😄

[Doohicksie]

6 hours ago, LGR4GM said:

Final thought the article implies Anderson is in line to be the starter but we know he's not. Comrie is the starter and depending on how he does, really tells us what our season will be. 

Does the article even mention UPL as an option in goal?

[Doohicksie]

6 hours ago, PerreaultForever said:

the season imo hinges a lot on how much of a leap forward Cozens and/or Mitts can make and how much immediate impact Power has. If Cozens takes a Thompson style leap forward we will exceed expectations. If he doesn't, well we could easily be in the bottom 5. 

Remember, to improve one of those centers needs to be better than Eakin.  I think Eakin was one of our worst forwards the last several years but I think there were traits that were less apparent that don't necessarily show up on the stat line (i.e., the wily veteran stuff like the neck roll in Jack's direction and other subtle stuff to get the other team thinking about stuff other than hockey).  I think Cozens may be able to pick up that mantel but he's still relatively new at it.

[LGR4GM]

30 minutes ago, Doohickie said:

Does the article even mention UPL as an option in goal?

Yes

Quote

Anderson, Subban and Luukkonen are back in the fold, but they’re all imperfect fits; Anderson is 41, Subban’s career save percentage is .898 and Luukkonen couldn’t stick in the NHL last season despite the organization’s glaring need for something — anything — resembling a viable option...

...Between the pipes, the Sabres don’t have that typical ‘core’ starter. They’ll likely have a tandem, headlined by Craig Anderson who took over the starting job last year. His workload wasn’t the easiest behind a below-average defensive team at five-on-five that offered little goal support.

Even with that workload in mind, however, he still conceded 7.8 more goals than expected in all situations. The more this team grows, the more support he’ll have. An average goaltender can be fine if the team around him can make up for his weaknesses, but that probably won’t be the case here. There are questions if Anderson can even still be average at 41 years old. Goaltending is volatile as is and only gets dicier as players age. So it’s worth watching just how much he plays this year, and whether anyone else can push for more playing time. Comrie is the best bet, but Ukko-Pekka Luukkonen has a shot too.

 

[MattPie]

14 hours ago, sabremike said:

My favorite part is the idea that we simultaneously have the deepest and best prospect pool in the sport by a country mile and must finish in the high lottery again to add to that pool. Dom is really the most crude stereotype of fancystat types who think the sport is just a math problem to be solved.

I think the stats (both fancy and not) do have a tough time with trending data vs. average. "well, last year xxx averaged below the league for the season, etc.". But as the Sabres are showing, that doesn't always make sense. At best, they stats folks will cut the data down to half a year or since Valentines Day or whatever to show that it's different. But I don't see many "over the course of the year this changed x" type stats.

As others said above, this makes sense when you're dealing with relatively static players and doesn't cope well with very dynamic situations (like the Sabres, 2019-2022). I'm going to pick on @LGR4GMfor a second, in the Tage thread there's a lot of talk about his average shooting % over the last three years, but (IIRC) it's only gone up each year. You can't really take a meaningful average in that case; there's no stability in the data. I think the average that comes out (~11%?) does make sense, but the data used to get there would predict something ridiculous like 30% if taken at face value.

This isn't to excoriate stats, I'm an engineer and I do believe math can explain a *lot* of things. People are still learning what predicts and what doesn't. I think as the methods get refined the stats will get better at seeing trends and not just averages.

Edited September 21, 2022 by MattPie

[LGR4GM]

12 minutes ago, MattPie said:

I think the stats (both fancy and not) do have a tough time with trending data vs. average. "well, last year xxx averaged below the league for the season, etc.". But as the Sabres are showing, that doesn't always make sense. At best, they stats folks will cut the data down to half a year or since Valentines Day or whatever to show that it's different. But I don't see many "over the course of the year this changed x" type stats.

As others said above, this makes sense when you're dealing with relatively static players and doesn't cope well with very dynamic situations (like the Sabres, 2019-2022. I'm going to pick on @LGR4GMfor a second, in the Tage thread there's a lot of talk about his average shooting % over the last three years, but (IIRC) it's only gone up each year. You can't really take a meaningful average in that case; there's no stability in the data. I think the average that comes out (~11%?) does make sense, but the data used to get there would predict something ridiculous like 30% if taken at face value.

This isn't to excoriate stats, I'm an engineer and I do believe math can explain a *lot* of things. People are still learning what predicts and what doesn't. I think as the methods get refined the stats will get better at seeing trends and not just averages.

Idk what you are picking on me for exactly. I explained in detail that Tage shot 49% of all the NHL shots he has taken last year so it is more likely that the averages of years past is not correct. 

[Stoner]

15 hours ago, sabremike said:

My favorite part is the idea that we simultaneously have the deepest and best prospect pool in the sport by a country mile and must finish in the high lottery again to add to that pool. Dom is really the most crude stereotype of fancystat types who think the sport is just a math problem to be solved.

That's great. About the best explanation I've seen for why fancystats is terrible. I'll keep it in my back pocket.

[GASabresIUFAN]

35 minutes ago, MattPie said:

I think the stats (both fancy and not) do have a tough time with trending data vs. average. "well, last year xxx averaged below the league for the season, etc.". But as the Sabres are showing, that doesn't always make sense. At best, they stats folks will cut the data down to half a year or since Valentines Day or whatever to show that it's different. But I don't see many "over the course of the year this changed x" type stats.

As others said above, this makes sense when you're dealing with relatively static players and doesn't cope well with very dynamic situations (like the Sabres, 2019-2022). I'm going to pick on @LGR4GMfor a second, in the Tage thread there's a lot of talk about his average shooting % over the last three years, but (IIRC) it's only gone up each year. You can't really take a meaningful average in that case; there's no stability in the data. I think the average that comes out (~11%?) does make sense, but the data used to get there would predict something ridiculous like 30% if taken at face value.

This isn't to excoriate stats, I'm an engineer and I do believe math can explain a *lot* of things. People are still learning what predicts and what doesn't. I think as the methods get refined the stats will get better at seeing trends and not just averages.

Which is a nice way of saying this author is modelling something that can't or shouldn't be efficiently modeled without significant changes.

I don't think I'd publish the result of something that on the face of it doesn't make sense.  The Sabres, despite injuries, a 40 year old goalie, incorporating new players, while still fielding a team that often played guys like Bjork, Hayden and Eakin played the second half of the season with a record of 20-17-4 or an 88 pt pace.  In the last 1/4 of the season they went 12-6-3 or an 105 pt pace and that included a nearly 500 record against playoff teams during that stretch.  Also a 6-2 record with Power in the lineup.  

Somehow this incredibly poor model suggests a team that was progressing throughout last season, who upgraded goaltending and defense, and that is adding 3 prized rookies (a top overall pick in Power, and two forwards that had historically excellent seasons in the A last year), that the team would somehow be just treading water.

As I pointed out before, even if the returning players just play to last years level, full seasons of Mitts, Krebs, Tuch, coupled with the additions of JJP and Quinn, will add somewhere around 30 more goals to this lineup.  A lineup that scores 260 or better will be in playoff contention.  It's really that simple, but this complicated model doesn't seem at account for that effectively.  

Comparing us again to Det and Ott, the higher placement of this list imho is more of a factor of getting easier to predict veteran players in the off-season rather than substantially improving their teams.  That said, they do have the advantage of more proven goaltending.

 

Edited September 21, 2022 by GASabresIUFAN

[steveoath]

4 hours ago, Doohickie said:

If you're using Risto as a verb, I like it.  😄

Unintentionally,  but I like it also so I'm not changing it haha.

[LGR4GM]

I am trying to imagine how you projects the Flyers defense to be better than the Sabres defense and I can't but the Athletic did it. 

[PerreaultForever]

5 hours ago, Doohickie said:

Remember, to improve one of those centers needs to be better than Eakin.  I think Eakin was one of our worst forwards the last several years but I think there were traits that were less apparent that don't necessarily show up on the stat line (i.e., the wily veteran stuff like the neck roll in Jack's direction and other subtle stuff to get the other team thinking about stuff other than hockey).  I think Cozens may be able to pick up that mantel but he's still relatively new at it.

I don't really follow this logic. 

Eakin is out, Sheahan is in. That's your 4th line center. Sheahan's a little faster, Eakin was better at faceoffs. Both of them are/were scrubs imo. Basically a wash. 

What we need is better 2nd and 3rd lines to be competitive. We were pretty much a one line team last year. If Skinner Thompson Tuch is still your top line we need to figure out how to create solid middle lines and that imo is dependent on Cozens and Mitts (with Krebs having an outside longshot). 

[GASabresIUFAN]

25 minutes ago, PerreaultForever said:

I don't really follow this logic. 

Eakin is out, Sheahan is in. That's your 4th line center. Sheahan's a little faster, Eakin was better at faceoffs. Both of them are/were scrubs imo. Basically a wash. 

What we need is better 2nd and 3rd lines to be competitive. We were pretty much a one line team last year. If Skinner Thompson Tuch is still your top line we need to figure out how to create solid middle lines and that imo is dependent on Cozens and Mitts (with Krebs having an outside longshot). 

Sheahan isn’t making this team unless someone gets hurt in camp.

[PerreaultForever]

17 minutes ago, GASabresIUFAN said:

Sheahan isn’t making this team unless someone gets hurt in camp.

Possibly, so what's your 4th line? You have Zemgus at center? 

I know everybody's optimistic, and the marketing is good, but I still think the Sabres are content (this year)_ to tank for a shot at Bedard etc. I don't think the line up will be a whole lot different than last year. Just a little with Power and maybe one or both of Quinn and Peterka but not a lot will change. 

[LGR4GM]

38 minutes ago, PerreaultForever said:

Possibly, so what's your 4th line? You have Zemgus at center? 

I know everybody's optimistic, and the marketing is good, but I still think the Sabres are content (this year)_ to tank for a shot at Bedard etc. I don't think the line up will be a whole lot different than last year. Just a little with Power and maybe one or both of Quinn and Peterka but not a lot will change. 

THEY ARE NOT TANKING. 

Words have meaning. You don't just call everything tanking because they didn't do what you wanted in the offseason. 

39 minutes ago, PerreaultForever said:

Possibly, so what's your 4th line? You have Zemgus at center? 

I know everybody's optimistic, and the marketing is good, but I still think the Sabres are content (this year)_ to tank for a shot at Bedard etc. I don't think the line up will be a whole lot different than last year. Just a little with Power and maybe one or both of Quinn and Peterka but not a lot will change. 

You mean the lineup at the end of the year that was on a 105pt pace... oh no the horror. 

Also you forgot Lybushkin and Comrie. 

[LGR4GM]

43 minutes ago, PerreaultForever said:

but I still think the Sabres are content (this year)_ to tank for a shot at Bedard etc. I don't think the line up will be a whole lot different than last year. Just a little with Power and maybe one or both of Quinn and Peterka but not a lot will change. 

Actually I needed to add something.

So you claim they are tanking and that is 100% false based on literal facts. 

But let's examine the next part. You express the opinion that the lineup is the same as last year (just a little different). Let us take a peak shall we? Eakin, 69 games and gone. Pysyk, 68 games and gone. Hayden 55 games and gone. Hagg 48 games and gone. Miller 38 games and gone. Butcher, 37 games and gone. Jankowski, 19 games and gone. Caggiula, 18 games and gone. Ruotsalainen, 18 games and gone. So that is 9 players that played a majority or at least a chunk of games for Buffalo who aren't on the roster.

Just a little different: with Power replacing Pysyk... I mean yea sure if you think the 1st overall pick from 2021 is about the same as... checks notes... Mark Pysyk. What about Cody Eakin, let's give his games to Jack Quinn... I will be shocked if there is any difference there with Quinn probably only getting... checks notes again... 12 points. How about Hayden replaced by JJP, no way JJP can get much better than haydens 4 pts, I mean who realistically thinks he will get much more than that? Hagg, idk how Lybushkin can replace Hagg and his 8 whopping points and poor defensive metrics. 

We didn't even talk about Tuch or Krebs playing only parts of last year in Buffalo. They basically replace Caggilula and Ruotsalianen. 

So if you are saying that Eakin, Pysyk, Hagg, Hayden, Caggiula, and Ruostsalianen are the same or only slightly lesser than Power, Quinn, Tuch, Krebs, Peterka, Lybushkin, then yes the lineup is only a little different than what we did for a chunk of last year. Just a tiny amount, in fact I would argue it is almost exactly the same lineup. I mean no way Owen Power is going to be better than Pysyk or Hagg, they are basically the same. Eakin and Tuch, most teams view them as equals. Quinn and Caggiula, probably need a sweetner if you wanted to make that trade but they are almost identical. Because as you say " I don't think the line up will be a whole lot different than last year. Just a little" I get the impression your definition of a little is very different than mine. I didn't even get to the difference between Comrie and Tokarski or Dell. 

A lot did change. That's the intriguing part.