Speculation: Roster Building Thread: New Season Edition

[The Crypto Guy]

How does this denote their uselessness exactly?

It kinda speaks for itself when you look at what I quoted..Bad advance stats we were good, good advance stats we are bad. So that's why I'm not sure why people put much into them..they are pretty useless.

 

[SA16]

How does this denote their uselessness exactly?

Simple. It tells the story he wants it to tell.

 

[The Crypto Guy]

Simple. It tells the story he wants it to tell.

And when it doesn't tell the story you want it to tell you fly in with the silly comments?

 

[SA16]

And when it doesn't tell the story you want it to tell you fly in with the silly comments?

It tells the story of a team generating a lot of chances that is having unfortunate shooting luck. Happens all the time in small sample sizes. Now I’m sure you’ll tell me it’s because they’re taking bad shots or some other nonsense to attempt to explain why that is the case which isn’t true.

 

[DanielBrassard]

It kinda speaks for itself when you look at what I quoted..

You quoted a guy who posted their 5v5 stats through 13 games for the last 2 seasons where the Rangers have been shown to have improved dramatically in each category except for Shooting percentage and save percentage. The point of the post being that sometimes, particularly in a smaller sample size, a team even if they have improved their process can be bitten by misfortune. How does this prove analytics are useless? It's not all that self evident.

 

[The Crypto Guy]

You quoted a guy who posted their 5v5 stats through 13 games for the last 2 seasons where the Rangers have been shown to have improved dramatically in each category except for Shooting percentage and save percentage. The point of the post being that sometimes, particularly in a smaller sample size, a team even if they have improved their process can be bitten by misfortune. How does this prove analytics are useless? It's not all that self evident.

All scoring chances aren't equal. Maybe they are just shooting more garbage shots from the outside? Point is there a ton more factors that go into the game than what the "fancy" stats can tell.

 

[SA16]

All scoring chances aren't equal. Maybe they are just shooting more garbage shots from the outside? Point is there a ton more factors that go into the game than what the "fancy" stats can tell.

There are five teams shooting under 7% this season. There are four teams shooting over 10% this season.

In the last five years there have been an average of 2.4 teams under 7% and 1 team over 10%. It's a small sample. These things happen. Do you expect a 41 goal scorer to score exactly 1 goal in every other game?

If they were shooting more "garbage shots from the outside" then their xG would not go up much since those shots have very low value.

I guess Colorado and Toronto are doing the same thing? Rangers 2.1 goals/60. Colorado 2.2. Toronto 2.1. What do you know. All 3 just happen to be 3 of the 5 teams shooting under 7%.

 

[HatTrick Swayze]

I think calling them “advanced stats” has done more harm than good.

For the most part what is in that tweet is essentially just…counting. It isn’t an algorithm or model or anything. It is counting shot attempts and scoring chances instead of just shots and goals.

No, every scoring chance is not created equal. But over time some of yours will be under graded, some will be over graded, and the same for your opponent. It should wash out over time.

No, being the best team at generating shot attempts does not mean than you will win any one game let alone the Stanley Cup. But the truly elite teams tend to…be more successful at playing hockey aka having the puck in the offensive zone creating chances. So if you can do that consistently, more frequently than your opponent who is attempting to do the same thing, chances are you are doing something right I.e. playing good hockey.

I honestly don’t understand why this is controversial. But we’ve been down this road 10000 times.

 

[NoQuitInNewMexico]

I only go down the royal road.

 

[TominNC]

The answer to the roster problem is easy, I still don't get why they don't try:

Kreider-Ziba-Panarin
Kid Line
Blais-Trochek-Kravstov
Vesey-Goodrow-Carpenter/Reaves

The bottom six are basically interchangeable with who is playing well. I get Trocheck was signed for an offensive role but he'll still get his PP time.

They need more even strength goals. Trochek can essentially form an old school shut down checking line and play against the other team's first line.

If Kravstov is not playing well or injured, you move up Goodrow or Vesey with him. Basically it's Trochek with whoever of Kravstov/Goodrow/Blais/Vesey is going hot and the other guys on the fourth line. This ideally keeps Carpenter out of the lineup as much as possible.

Carpenter is the weak link. Getting regular minutes is a problem. If he's used in the Chris "Birdman" Andersen role, he'll be much more fresh and effective.

Because Panarin plays LW. Panarin thrives on coming in the zone and making that cut. Guys play positions. Put players in the positions where they play best.

 

[ncdoc]

Stats look at the long-term picture. It is early in the season. Those numbers are encouraging (outside the save percentage). You cannot apply stats to an individual game.

 

[GoAwayPanarin]

All scoring chances aren't equal. Maybe they are just shooting more garbage shots from the outside? Point is there a ton more factors that go into the game than what the "fancy" stats can tell.

This year through 13 games at 5v5 - 140 high danger chances for (117 against)

Last year through 13 games at 5v5 - 85 high danger chances for (122 against)

While high danger chances are subjective, they certainly aren't garbage shots from the outside. I don't even think they've played THAT well, but the point is that they've been pretty unlucky on top of that.

 

[Off Sides]

There is so much context missing within the stats. A shot from Zbad, from whatever distance at whatever angle, is a better scoring chance if Blais (or anyone else beyond Panarin) is taking the same shot, yet they are treated the same in terms of high danger scoring chances.

Yes it's nice to see the stats where they are taking more 5on5 attempts, a proxy for possession, yet a proxy is just that.

Returning to the mean, it can take forever, there is no rule that says it needs to happen within any time frame. One can flip 100 coins and get heads 70 times, it could take anywhere from 30 to infinity to return to 50/50.

 

[QJL]

i really really hate Panarin and Kakko together

Yes, it really sucked when Kakko posted 8 points in 8 games playing with Panarin last year.

In truth, I agree they should probably be on separate lines, but I do think they can dominate together.

 

[SA16]

There is so much context missing within the stats. A shot from Zbad, from whatever distance at whatever angle, is a better scoring chance if Blais (or anyone else beyond Panarin) is taking the same shot, yet they are treated the same in terms of high danger scoring chances.

Yes it's nice to see the stats where they are taking more 5on5 attempts, a proxy for possession, yet a proxy is just that.

Returning to the mean, it can take forever, there is no rule that says it needs to happen within any time frame. One can flip 100 coins and get heads 70 times, it could take anywhere from 30 to infinity to return to 50/50.

70% of the Rangers 5v5 high danger shot attempts from forwards are from players in the 6.

For the Leafs it's ~61%.

For the Avs it's ~73%

For the Oilers it's ~70%

For TB it's ~65%

For the Sens it's ~70%

This is not a concern. It's almost the same ratio for everyone.

 

[Off Sides]

70% of the Rangers 5v5 high danger shot attempts from forwards are from players in the 6.

For the Leafs it's ~61%.

For the Avs it's ~73%

For the Oilers it's ~70%

For TB it's ~65%

For the Sens it's ~70%

This is not a concern. It's almost the same ratio for everyone.

Maybe McDavid and company are better scorers when it matters, just as Stamkos and company were, just as the Kings were back in the finals.

Rangers have two scorers, Zbad and Panarin, maybe Kreider when he is on. That is it.

 

[Clark Kellogg]

This reminds me of apChem when I turned to the girl sitting next to me and asked “ What am I looking at?”.

 

[Glen Sathers Cigar]

This reminds me of apChem when I turned to the girl sitting next to me and asked “ What am I looking at?”.

Green good, red bad.


Long story short, numbers like this indicate a team that is controlling play, outplaying their opponents, out chancing their opponents but not getting enough saves and not converting on their chances.

 

[Glen Sathers Cigar]

I don't think they are useless, but yeah they are just a tool that some people pretend are the end all be all.

Yeah, this is obviously truth. Calling them useless is just as wrong as saying they are all that matters IMO.

They're a good tool to use in your analysis to make it more well rounded. They're dumb if you're only using them in place of watching the games (which unfortunately many vocal fans online especially do this bc who has the time to watch all the teams in the league all season) but they're great if you want to dig deeper - especially the scoring chance numbers that Vally's company runs. Majority of the time (not always though) the stats end up lining up with what the eye test tells you anyway.

 

[bleedblue94]

It's funny how there are people who genuinely project that they would rather have the great fancy stats opposed to wins.

Analytics are an information tool but any information is useless without proper context.

 

[NoQuitInNewMexico]

It's funny how there are people who genuinely project that they would rather have the great fancy stats opposed to wins.

It's 13 games in! Like there's a good argument to be had between people who would prefer the team to look good and people who would prefer a good record in 1/6 of the season in the sport where playoff seeding basically isn't real.

I'm sitting this one out because unlike other short spells I can remember where they had good shot differentials/advanced stats and mediocre results, I think they kind of look like shit

 

[SnowblindNYR]

this is what happens when you rely on the stretch pass in the year 2022.

if you want to have a good transition game literally just look at the penguins. two guys going full speed and dump it in when they get to the opposing teams blue line, and get in behind the defense hard and fast, and outnumber them.

instead we do a stretch pass with a guy standing on the blue line and the other guy know not knowing he has to chase. this = lost possession

The Penguins have a worse record than us and have lost 7 straight games and counting...

Yeah, this is obviously truth. Calling them useless is just as wrong as saying they are all that matters IMO.

They're a good tool to use in your analysis to make it more well rounded. They're dumb if you're only using them in place of watching the games (which unfortunately many vocal fans online especially do this bc who has the time to watch all the teams in the league all season) but they're great if you want to dig deeper - especially the scoring chance numbers that Vally's company runs. Majority of the time (not always though) the stats end up lining up with what the eye test tells you anyway.

I guess the question are our eyes deceiving us or are these advances stats deceiving us?

 

[egelband]

Agreed and those numbers probably paint a rosier picture than the eye test has lately but to me they're pretty accurate. The team is good, they're not losing because they're a bad team or not good on paper. They're beating themselves, finding ways to lose and playing sloppy and mentally not fully there. A bit discombobulated, some injuries, some players not firing on all cylinders and a team that obviously came out of the gate playing well and decided after a handful of games "okay, when do the playoffs start?"

I think a game or two where everything goes right for them to get some confidence and good vibes going will snap them back into the groove. The Islanders coming to MSG followed by a pretty road heavy schedule should be the perfect opportunity for the equalizing to happen, let's hope it does soon before they lose too much ground.

Agree. They’re a good team playing crappy. Unless there’s something seriously wrong we aren’t aware of, I’m confident they’ll figure it out.

 

[Barnaby]

Just to switch things up… I’d like to see:

Panarin-Zibanejad-Vesey
Lafreniere-Trocheck-Kakko
Kreider-Chytil-Kravtsov
Blais-Goodrow-Gauthier

 

[NYR2007]

I think the lines needed a change a little while ago, I understand the coach sticking to his players, but also happy he finally made some moves.

The frustrating thing is when he throws these struggling players out for overtime when they have been dog shit all game. But then I guess people would be complaining the coach is gonna bench players if they make a mistake. Can’t make everyone happy