Whenever I try to understand what all the advance stats mean...
I guess I'll always be a "+/-" guy...
I guess I'll always be a "+/-" guy...
Jets Advanced Stats thread
- Thread starter JetsFan815
- Start date
-
Xenforo Cloud will be upgrading us to version 2.3.5 on March 3rd at 12 AM GMT. This version has increased stability and fixes several bugs. We expect downtime for the duration of the update. The admin team will continue to work on existing issues, templates and upgrade all necessary available addons to minimize impact of this new version. Click Here for Updates
just to piggy back off this a bit i know this was posted in a diff thread (off topic) but keep it in stats related.... i know dom gets labeled as a Jets hater for some reason. wonder if ppl labeled him as a fan-boy previously? as he picked the Jets to win the cup in 17-18 prior to the PO's, and as a top team in previous years .
just looking at his projections over the years
[TABLE=collapse]
[TR]
[TD][/TD]
[TD]Proj Pts[/TD]
[TD]Proj Finish[/TD]
[TD]Actual Pts[/TD]
[TD]Actual Finish[/TD]
[/TR]
[TR]
[TD]2016-2017[/TD]
[TD]
90
[/TD][TD]
18
[/TD][TD]
87
[/TD][TD]
t-20
[/TD][/TR]
[TR]
[TD]2017-2018[/TD]
[TD]
100
[/TD][TD]
6
[/TD][TD]
114
[/TD][TD]
2
[/TD][/TR]
[TR]
[TD]2018-2019[/TD]
[TD]
100
[/TD][TD]
3
[/TD][TD]
99
[/TD][TD]
t-10
[/TD][/TR]
[TR]
[TD]2019-2020[/TD]
[TD]
92
[/TD][TD]
17
[/TD][TD]
92
[/TD][TD]
20
[/TD][/TR]
[TR]
[TD]North Division (in-division finish)[/TD]
[TD]
61
[/TD][TD]
t-5
[/TD][TD]
63
[/TD][TD]
3
[/TD][/TR]
[TR]
[TD]2021-2022[/TD]
[TD]
94
[/TD][TD]
15
[/TD][TD]
89
[/TD][TD]
19
[/TD][/TR]
[TR]
[TD]Average (Non North-Div years)[/TD]
[TD]
95
[/TD][TD]
12
[/TD][TD]
96
[/TD][TD]
14
[/TD][/TR]
[/TABLE]
North Division finishes based on just standings in the division
19-20 proj 82 gp points
really, it's not far off. he underestimated their 17-18 season quite a bit. Maybe the change from Mason to Helle as a starter had a ton to do with it, and then KC called up full time ofc as neither were starters to begin the year. but the average of projected vs actual is nearly spot-on.
Last edited:
tbcwpg
Moderator
- Jan 25, 2011
- 16,901
- 20,868
I don't believe Dom is a Jets hater - I believe he doesn't watch enough of the Jets to put any kind of meaningful commentary behind his stats about the Jets, and also has been vocal about Scheifele and Wheeler not being elite level players anymore, which some fans took offence to and caused him to dig in even more.
We can see with New Jersey that a new season and some more experience for younger players can push a team from middling to elite level, like the Jets did from 2016-17 to 2017-18.
JetsFan815
Replacement Level Poster
- Jan 16, 2012
- 19,889
- 26,378
No one could have imagined that moving on a from a "PK strategy" that relies on passively waiting around giving all kinds of time and space to the opposing players to make plays would improve the team's PK metrics 
I usually am more charitable towards Maurice than most here but on the PK (whether it was his scheme or Huddy's) his failings are the most stark.
I usually am more charitable towards Maurice than most here but on the PK (whether it was his scheme or Huddy's) his failings are the most stark.
Romang67
BitterSwede
- Jun 10, 2014
- 59,915
- 33,471
Whenever I try to understand what all the advance stats mean...
View attachment 608649
I guess I'll always be a "+/-" guy...
I have similar trouble with most advanced stats. Deep down, I suspect that half of them don't mean what they are supposed to mean. But don't fall back on +/-. That takes you from 0 or near 0 on the stats understanding scale into negative territory. It is disinformation.
tbcwpg
Moderator
- Jan 25, 2011
- 16,901
- 20,868
Yeah. Dom is an HFJets poster, people just don't like people from outside of Winnipeg saying it
Which is where the problem lays. Your job is to watch hockey and to come up with meaningful stats to reflect in numbers what you watch, and to be able to make predictions about hockey. But you dont actually watch enough if it to confirm if what you are writing is correct?
Like reading an article about the Jets and projecting Perfetti to be star because of his blazing speed?
it seems that way. like some of the things people say about wheeler, scheifele over the years, the made up rumors.... you'd think those posters would be the haters, or TMZ.Yeah. Dom is an HFJets poster, people just don't like people from outside of Winnipeg saying it
anyway as far as "hating the Jets" it seems like he's closer to predicting where they'll finish than some on here. perhaps if the Jets iced better teams or played better, he or other prediction sites would give them more respect
. when they had great teams in 17-18 and a similar roster in 18-19 he had them 6th and 3rd overall, and had them as his cup team in '18 POs. What have the Jets done since to really garner huge amount of respect?but over the years, looks like he's pretty close to their performance. how they got there perhaps is a different story. i don't think these National or overall NHL reporters really have time to go through watching each of the 32 teams equally, dissecting games/shifts etc. id look at Murat for that for Jets specifically.
tbcwpg
Moderator
- Jan 25, 2011
- 16,901
- 20,868
He's not going to be able to watch every NHL game. He's also been right far more often than not. I think this season is a bit different - the sample size is pretty small and a lot of the Hellebuyck stuff he attributes it to happened in only a couple of games.
I don't particularly care that he uses his model to predict and make generalizations. I care that he doesn't wait until there is a sufficiently large enough sample of games before making his sweeping generalizations using his numbers.
His model numbers right now don't say much of anything as the aggregates are still heavily influenced by single event outcomes.
every sport posts power-rankings based on what has happened to-date. this is nothing new. no one is forcing you to read or comment on anything. people. including probably you and me, make knee-jerk, or small-sample conclusions all the time on this forum.
I hold people who are paid to write about the sport to a higher standard especially if they use numbers to back up their opinions.
If you are going to use numbers when you know the sample isn't statically significant then provide some nuance. I'm sorry but I expect more then Helle is the only reason for the Jets success from a paid analyst.
Romang67
BitterSwede
It's sloppy work that gives people with a vested interest against numbers ammunition.
It often is with Dom.
Mud Turtle
Registered User
- Jul 26, 2013
- 8,740
- 20,681
Never mind not understanding advanced stats, I don’t even know who “Dom” is.
the first paragraph from his first 16-stats posting which @tbcwpg cited
"Early in the season, it’s easy to get swept away by small sample madness. After only seven or so games, it’s difficult to separate what’s real and what isn’t — fact from fiction. This early, it’s best to still lean heavily toward priors with some exceptions made from extreme cases. It’s those extremes where you can usually find some signal amid the noise."
it seems to me he does. perhaps he should repeat the same thing in every article just for you
there's also this piece that was missing from that post from tbcwpg
"There’s your answer, same as it always has been for the Jets. They go as far as Hellebuyck takes them, but in Hellebuyck, they have one of the safest bets in the league. Never forget that his “down season” last year had him in the league’s top 10 for goals saved above expected. He’s still one of the absolute best goalies in the world."
i don't necessarily disagree with this. Helle has kept this team afloat given they allow lots of chances against over the years. the Jets have trended well as of late this November, and better vs their 1st 10 or so games. but similarly, concluding that they are a good or great defensive team after a few good games would be making conclusions over small samples.
edit: his article was also posted nov 10th maybe written a day or 2 prior. so the Jets only had played 3 Nov games on there since they've played better in their own end. i think if he re-wrote it today it'd look different tbh.
Last edited:
the first paragraph from his first 16-stats posting which @tbcwpg cited
"Early in the season, it’s easy to get swept away by small sample madness. After only seven or so games, it’s difficult to separate what’s real and what isn’t — fact from fiction. This early, it’s best to still lean heavily toward priors with some exceptions made from extreme cases. It’s those extremes where you can usually find some signal amid the noise."
it seems to me he does. perhaps he should repeat the same thing in every article just for you
there's also this piece that was missing from that post from tbcwpg
"There’s your answer, same as it always has been for the Jets. They go as far as Hellebuyck takes them, but in Hellebuyck, they have one of the safest bets in the league. Never forget that his “down season” last year had him in the league’s top 10 for goals saved above expected. He’s still one of the absolute best goalies in the world."
i don't necessarily disagree with this. Helle has kept this team afloat given they allow lots of chances against over the years. the Jets have trended well as of late this November, and better vs their 1st 10 or so games. but similarly, concluding that they are a good or great defensive team after a few good games would be making conclusions over small samples.
edit: his article was also posted nov 10th maybe written a day or 2 prior. so the Jets only had played 3 Nov games on there since they've played better in their own end. i think if he re-wrote it today it'd look different tbh.
That isn't at all what I meant about nuance and he goes and pretty much contradicts that line about sample sizes when he states we go as far as Helle takes us anyhow.
We just disagree on this and that is fine.
Evolving Wild just put this up:
Jets are doing well across the board except we're scoring below expectations (as we did last year, too) but that should turn around if Connor gets rolling.
Jets are doing well across the board except we're scoring below expectations (as we did last year, too) but that should turn around if Connor gets rolling.
tbcwpg
Moderator
- Jan 25, 2011
- 16,901
- 20,868
Goaltending is the lion's share for sure, but shot rate and quality differentials are both positive - which is pretty good considering the Jets schedule and how they got buried in Vegas.
JetsFan815
Replacement Level Poster
- Jan 16, 2012
- 19,889
- 26,378
tbcwpg
Moderator
- Jan 25, 2011
- 16,901
- 20,868
Yeah, the lack of a long bar on the negative side is very nice, just seems like many models put goaltending as the main factor.
It was just a matter or time... He wasn't shooting sub 5% overall and 0% at 5v5 forever.
