First Third of Jets Season vs The Last Two Thirds

[Buffdog]

I've given this idea way more thought than it deserves. Basically, purposefully excluding any outlier of a 16 game stretch in the search for how good (or bad) a team may or may not be is a disgrace to data analysis. That goes for both good AND bad stretches. Also, if someone is going to use one sample size, they should compare it to another of the same size (notn16 games to 32 games)

If someone really wanted to use stretches of games to look at a team's performance, this is how it should be done:

First, decide on an interval of games. Garrett chose 16 games to omit, so let's use that. You need to look at ALL stretches of 16 games. For this season, there are 35 such stretches (1-16, 2-17, 3-18... 35-50)

Look at the number of points (or points percentage) earned during each if those stretches and plot them on a graph. From there, it would be easy to see the outliers (both good and bad) and you'd be able to draw a curve that shows the average

Big outliers (in both the positive and negative directions) like games 1-16 (30 points) and games 19-34 (15 points) should be discussed, but never omitted

 

[BobTheSolarsystem]

Yet, in fact, he had not given it enough thought... or read the article.

 

[macmaroon]

Why are some trying so hard not to enjoy the best team in the Western conference and 2nd best in the NHL??? I don't understand and never will. If we were 0-10 in our last 10 games i could see people being upset but jesus try and enjoy our team just a little instead of always saying ya but.

This is what HFJets is all about @Jets 31, we are never satisfied...

 

[Heldig]

Panic in advance of the playoffs

Jets are a top team in the league atm but the only record that matters is how they do in the playoffs.

 

[Jets 31]

Considering that middle 3-7 stretch was 10 games in 19 nights with a B2B, that had the team on go from the East Coast to the West Coast on the same road trip. You take out the hardest stretch of games we are looking real good. I wonder how each team's toughest stretch of games looks like compared to other stretches of the season? I betcha they are all pretty similar.

That's a very big part of the "bad" part of our GREAT season so far, that road trip from hell with a couple of single home games sprinkled in would be tough for every single team in the league. A 13 game stretch with 11 road games and 2 single home games sprinkled in, basically those 2 home games were like road games. That's a awful schedule and yet here we sit 33-14-3 1st in the conference and 2nd best in the NHL, i couldn't be happier with our season right now.

 

[Board Bard]

What a series of bizarre reactions to Garret's article... and too many expected 'so therefore I didn't read it' - because they had confirmed their own biases early haha...

'This isn’t to say that Winnipeg is bad or that 16-13-3 reflects the “real” Jets.

The 15-1-0 run was just as much a part of the team’s story. Both stretches contribute to the full picture of what defines this team. The question is where in the middle is this team?'

I am missing the part where he is 'anti-Jets'

'Interestingly, despite being the simplest metric—focused solely on shot quantity—Corsi consistently predicts team and player success better than expected goals or goals.'

'Be skeptical of anyone dismissing the insights from Corsi simply because goals or expected goals metrics suggest a different narrative that matches their own opinion. Shot volume, as measured by Corsi, often provides the most reliable foundation for analyzing team and player performance.'

'They have solid goaltending. They have an elite power play. Yet, there are still gaps—particularly in their 5v5 play—that need addressing if they want to maximize their playoff chances.'

Inflammatory stuff.

I honestly don't see the negative in his article... EVERYONE on this board knows we are pretty good at 5v5 but that our powerplay is the straw that stirs the drink and the Helle wins extra games for us.

The charts do meet the eye test and they do show that lines/pairings with more ice time are closer to the mean.

Maybe the anger stems from the data showing that Stan is not good.... I don't know anymore.

Some people think that anyone who can add, subtract, multiply and divide is a pencil-necked nerd who couldn't possibly know a hockey puck from a decimal point. Similarly, entrenched oldtimers won't credit any stat that Don Cherry and Danny Gallivan didn't use. Then you've got your blind homers who will countenance no opposing opinion and whose favourite sport is not hockey but appealing to authority. The extra pompous only have to know that something exists to know that it's crap; this can be a useful trait as it saves on reading time. Sometimes some or all of those traits exist in one person, like a Marvel superhero.

 

[Cotton Eye Joe]

Some people think that anyone who can add, subtract, multiply and divide is a pencil-necked nerd who couldn't possibly know a hockey puck from a decimal point. Similarly, entrenched oldtimers won't credit any stat that Don Cherry and Danny Gallivan didn't use. Then you've got your blind homers who will countenance no opposing opinion and whose favourite sport is not hockey but appealing to authority. The extra pompous only have to know that something exists to know that it's crap; this can be a useful trait as it saves on reading time. Sometimes some or all of those traits exist in one person, like a Marvel superhero.

Bingo!
Those posters who say they will not read an article on here are the exact ones who most should.
Be open to alternative positions other than their own. Refrain from juvenile attacking those who don't agree with them. Closed-minded, a multitude of incorrect predictions and statements.
Yet believing they have all the answers. Just ask them.
Usually a lost cause though.

 

[gojetsgo]

I read the article before this thread was even made, my posts have nothing to do with this article but are towards the op and others who have been calling us a .500 team...

 

[BobTheSolarsystem]

That's a very big part of the "bad" part of our GREAT season so far, that road trip from hell with a couple of single home games sprinkled in would be tough for every single team in the league. A 13 game stretch with 11 road games and 2 single home games sprinkled in, basically those 2 home games were like road games. That's a awful schedule and yet here we sit 33-14-3 1st in the conference and 2nd best in the NHL, i couldn't be happier with our season right now.

Except no one has framed anything as bad...

The corsi and xgf are interesting and should be acknowledged... no idea why fans want to dismiss it

Vgk is one spot lower than us so playoffs should be a breeze haha

 

[Buffdog]

Bingo!
Those posters who say they will not read an article on here are the exact ones who most should.
Be open to alternative positions other than their own. Refrain from juvenile attacking those who don't agree with them. Closed-minded, a multitude of incorrect predictions and statements.
Yet believing they have all the answers. Just ask them.
Usually a lost cause though.

Name one of these incorrect predictions and statements

You're the one who had the most recent incorrect statement that I can recall (claiming that CSV got owned in the playoffs last year). When corrected with the data, you refused to admit you were wrong

Reading an article or looking at a spreadsheet doesn't mean you understand anything. Just curious... what's the highest level of hockey have you played or coached?

 

[raideralex99]

Here's what I have noticed over the years following hockey and why schedules are a big part of a teams success.
It days back to the Jets 1.0 Gretzky era. I can show lots of examples but I will keep it short.
When the Jets went on the road and did the Alberta trip ... they would first go to Edmonton play a great game before losing in the 3rd period to Gretzky and the next night they would get blown out by the Flames in Calgary. Come playoff time Jets first round opponents were the Flames but this time the Flames did not have the advantage of playing a tired Jets team and guess what ... the Jets knocked off the 3rd best record team in the league in the first round. I made a lot of money on that series.
The NHL finally straighten out their schedule by not giving some teams a great advantage. I remember one season it seemed like the Canucks always played a team B2B so I checked the Canucks schedule and they played 22 of their 41 home games where the visiting team played the night before or it was their 3rd game in 4 nights and of course come playoff time the first place Canucks lost in the first round.

 

[Skidooboy]

Except no one has framed anything as bad...

The corsi and xgf are interesting and should be acknowledged... no idea why fans want to dismiss it

Vgk is one spot lower than us so playoffs should be a breeze haha

the sampling of the article was absurd.

general rule in stats . anything under 30 data points is probably too small a sample for analysis.

make samples EVEN and consistent. 15/15/15? might give you a result

drop the first 15 for “reasons”? then look at the next 8 then the 22 after that?

Its a joke and i can’t believe an educated mathematician would claim it’s revealing of anything….

especially when looking at advanced stats…. because Xgoals etc are all about probabilities not results… so the data from the first 15 should 100% count.

if i went out and was sampling say minnow egg viability….

if i find 15 perfectly viable eggs at the front of my sample.. do i say “It’s a fluke” and cancel the data? hells no. so why does anyone pretend hockey stats should have that kind of exclusion?

it’s like last year.
I was told the Jets were bad last year in the second half because there was a six game losing streak….. and if you broke up the jests season in a way that ignored the record before that streak, and ignored the 8-0 the jets finished with down the stretch….. there was a period between Late Feb and April, where the jets were 3-6-1 or whatever ….it showed the jets were “bad at hockey” in the second half the season…..

well duh…if you arbitrarily cut data up willy nilly, & only look at the data that proves your narrative? Then yeah, you can find data to show whatever you want…

 

[Buffdog]

the sampling of the article was absurd.

general rule in stats . anything under 30 data points is probably too small a sample for analysis.

make samples EVEN and consistent. 15/15/15? might give you a result

drop the first 15 for “reasons”? then look at the next 8 then the 22 after that?

Its a joke and i can’t believe an educated mathematician would claim it’s revealing of anything….

especially when looking at advanced stats…. because Xgoals etc are all about probabilities not results… so the data from the first 15 should 100% count.

if i went out and was sampling say minnow egg viability….

if i find 15 perfectly viable eggs at the front of my sample.. do i say “It’s a fluke” and cancel the data? hells no. so why does anyone pretend hockey stats should have that kind of exclusion?

it’s like last year.
I was told the Jets were bad last year in the second half because there was a six game losing streak….. and if you broke up the jests season in a way that ignored the record before that streak, and ignored the 8-0 the jets finished with down the stretch….. there was a period between Late Feb and April, where the jets were 3-6-1 or whatever ….it showed the jets were “bad at hockey” in the second half the season…..

well duh…if you arbitrarily cut data up willy nilly, & only look at the data that proves your narrative? Then yeah, you can find data to show whatever you want…

I can understand re-examining an outlier data point for accuracy but once it's been confirmed, you can't just eliminate it

If our 15-1 record to start the season was an error, then yeah it should be discarded

 

[scelaton]

Reading an article or looking at a spreadsheet doesn't mean you understand anything.... Just curious... what's the highest level of hockey have you played or coached?

Just thought I'd point out that this question, which you often ask at the end of an argument with other posters, is the equivalent of "You can't teach height" when arguing the merits of Stanley. Both are factors, but not major or determining factors in contesting an otherwise legitimate, opposing POV.

I'd suggest that reading articles, understanding analytics and having played the game at a high level is the ideal combination, but you if you are astute and open-minded, you can get away with any combination of two of them.

 

[Skidooboy]

I can understand re-examining an outlier data point for accuracy but once it's been confirmed, you can't just eliminate it

If our 15-1 record to start the season was an error, then yeah it should be discarded

except it wasn’t.

there were three separate events in succession.

break it up by recent “streaks” and throw the playoffs in or out…. split it into 8 game increments of reg season.

8-0, (0-4), 8-0 ,7-2…

or 8-0 8-0 7-2.



so how “unrepeatable is it?

 

[Buffdog]

Just thought I'd point out that this question, which you often ask at the end of an argument with other posters, is the equivalent of "You can't teach height" when arguing the merits of Stanley. Both are important, but not major or determining factors in contesting an otherwise legitimate, opposing POV.

I'd suggest that reading articles, understanding analytics and having played the game at a high level is the ideal combination, but you if you are astute and open-minded, you can get away with any combination of two of them.

Personally, I do read articles and understand analytics. I've also played, coached and officiated at relatively high levels. Which is why I ask people when I'm discussing topics with them

Go through my post history and you'll see I reference stats and analytics often, but always as a tool of providing context rather than as an end all be all

I'd like to end by reiterating what Skidooboy said... anyone who doesn't laugh at the premise of this article has zero understanding of how stats actually work... and I don't mean in hockey, I mean "stats, period". Cherrypicking datasets is a career-ender in that field

 

[Skidooboy]

let’s look at on ice day in day out ice and skates and blood.

85 wins in the last season and a half…

85 out of 132 64.3% win rate .643 if you like.

.264 possible points jets got 169… 64.01 of available points.

tell me again how this is flukey or unsustainable….. based on what? a four game sample from the playoffs?

 

[BobTheSolarsystem]

Here's what I have noticed over the years following hockey and why schedules are a big part of a teams success.
It days back to the Jets 1.0 Gretzky era. I can show lots of examples but I will keep it short.
When the Jets went on the road and did the Alberta trip ... they would first go to Edmonton play a great game before losing in the 3rd period to Gretzky and the next night they would get blown out by the Flames in Calgary. Come playoff time Jets first round opponents were the Flames but this time the Flames did not have the advantage of playing a tired Jets team and guess what ... the Jets knocked off the 3rd best record team in the league in the first round. I made a lot of money on that series.
The NHL finally straighten out their schedule by not giving some teams a great advantage. I remember one season it seemed like the Canucks always played a team B2B so I checked the Canucks schedule and they played 22 of their 41 home games where the visiting team played the night before or it was their 3rd game in 4 nights and of course come playoff time the first place Canucks lost in the first round.

I also remember you could literally injure your opponents best player ... except gretzky

let’s look at on ice day in day out ice and skates and blood.

85 wins in the last season and a half…

85 out of 132 64.3% win rate .643 if you like.

.264 possible points jets got 169… 64.01 of available points.

tell me again how this is flukey or unsustainable….. based on what? a four game sample from the playoffs?

But what are you arguing against? Because i cant find a post.. including garret where anyone has said this...

Id just like some indication that @Buffdog or @Skidooboy had actually read the article... because garret clearly points out that the difference between the two parts of the season is 14th or 15th place.. so he doesnt seem to be making the point you are arguing against.

 

[Jets 31]

Some people think that anyone who can add, subtract, multiply and divide is a pencil-necked nerd who couldn't possibly know a hockey puck from a decimal point. Similarly, entrenched oldtimers won't credit any stat that Don Cherry and Danny Gallivan didn't use. Then you've got your blind homers who will countenance no opposing opinion and whose favourite sport is not hockey but appealing to authority. The extra pompous only have to know that something exists to know that it's crap; this can be a useful trait as it saves on reading time. Sometimes some or all of those traits exist in one person, like a Marvel superhero.

Or you can just be a bitter person that isn't going to be happy no matter how good the Jets play. I choose to not be and i am thoroughly enjoying this incredible season so far. I am Iron man i guess according to you, that's cool.

 

[BobTheSolarsystem]

Or you can just be a bitter person that isn't going to be happy no matter how good the Jets play. I choose to not be and i am thoroughly enjoying this incredible season so far. I am Iron man i guess according to you, that's cool.

There is a very real symptom of a very good or a very bad team when fans start comparing whos the better fan... luckily this year it means we are good

 

[Jets 31]

There is a very real symptom of a very good or a very bad team when fans start comparing whos the better fan... luckily this year it means we are good

If we were getting outplayed every game and Helly was stealing us games i could totally understand poking holes in the season. That's not happening, we are playing some good hockey this season so we should enjoy it. It's so hard to be a really good team in this league, really easy to be just fighting for a playoff spot every year especially with the disadvantages we have just being in Winnipeg. We have to be on the most no trade lists so we almost have 1 hand tied behind our backs.

 

[Hobble]

Panic in advance of the playoffs

Exactly, just trying to get ahead of rationalizing an early playoff exit due to past traumas.

 

[BobTheSolarsystem]

If we were getting outplayed every game and Helly was stealing us games i could totally understand poking holes in the season. That's not happening, we are playing some good hockey this season so we should enjoy it. It's so hard to be a really good team in this league, really easy to be just fighting for a playoff spot every year especially with the disadvantages we have just being in Winnipeg. We have to be on the most no trade lists so we almost have 1 hand tied behind our backs.

True. But what the analytics say is that we are holding our own at 5v5 and have a massively dominant powerplay and massively dominant goalie. And we all know this... everyone knows you cant win the cup without a great goalie.

The huge positives i see this year are:
- no debates about csv even though there are now cracks showing
- ehlers critics have melted into the walls.... until the playoffs - and many of these guys are stans main defenders
- pionk has proven analytics wrong mostly - i still see cracks in his game but his offence overwhelms the negative and samberg is dillon plus
- helle is mega focused
- demelo is back to form
- arniel had mixed lines with success

But we are stuck arguing advanced stats because they say a great team regresses to the mean... and yes the jets could use another solid player

And yes every single form of analytics says stan sucks haha

 

[Adam da bomb]

True. But what the analytics say is that we are holding our own at 5v5 and have a massively dominant powerplay and massively dominant goalie. And we all know this... everyone knows you cant win the cup without a great goalie.

The huge positives i see this year are:
- no debates about csv even though there are now cracks showing
- ehlers critics have melted into the walls.... until the playoffs - and many of these guys are stans main defenders
- pionk has proven analytics wrong mostly - i still see cracks in his game but his offence overwhelms the negative and samberg is dillon plus
- helle is mega focused
- demelo is back to form
- arniel had mixed lines with success

But we are stuck arguing advanced stats because they say a great team regresses to the mean... and yes the jets could use another solid player

And yes every single form of analytics says stan sucks haha

Ehlers gets polar reactions around here. He has some really tough critics.


Between Ehlers is an important player and is probably our 4th most important player behind our number 1 scored, our number 1 centre and our big shutdown centre who should be in the running for a selke.

and

no, if he was used right he would have 90-100 points and should be number 1. “

yes, he has a lot more potential you can’t call him a 70-80 point player, until he does it, even though he probably has the potential to produce more.


Also, between Ehlers is a really important part of our pp and is equally important to everyone else,

and

no just admit it, he completely drives the power play and is the reason why it’s at the top of the league.

Give him the same as Schief vs

he is having a great season, but, let’s see if he can stay healthy and can produce in the playoffs.

I also don’t hold with the pair him with mcd or Mack suggestion. I think he does his best work with a Tkachuk, ratananen, Drai.

 

[BobTheSolarsystem]

Ehlers gets polar reactions around here. He has some really tough critics.


Between Ehlers is an important player and is probably our 4th most important player behind our number 1 scored, our number 1 centre and our big shutdown centre who should be in the running for a selke.

and

no, if he was used right he would have 90-100 points and should be number 1. “

yes, he has a lot more potential you can’t call him a 70-80 point player, until he does it, even though he probably has the potential to produce more.


Also, between Ehlers is a really important part of our pp and is equally important to everyone else,

and

no just admit it, he completely drives the power play and is the reason why it’s at the top of the league.

Give him the same as Schief vs

he is having a great season, but, let’s see if he can stay healthy and can produce in the playoffs.

I also don’t hold with the pair him with mcd or Mack suggestion. I think he does his best work with a Tkachuk, ratananen, Drai.

The health part is the gamble for sure... but there are so few real play drivers in the league - i think we have really under valued that