Official Coyotes 2022-23 Season Discussion.

[XX]

Any idea how they set the baseline for expectations?

Expected is another way of saying xG. 'Above expected' is xG differential.

xG is shots+shot quality. A point blank chance would result in a large xG number, a flip in from the point would not. This correlates more strongly with outcomes than raw totals like Corsi or Fenwick. It only recently became a thing due to better tracking technology.

Veg and Ingram have a combined 17 goals saved above expected, so they're part of the reason. Offense wise, the Coyotes are barely above expected for the year (1.29 total), so the offense is largely WYSIWYG. There's not an obvious emphasis on attempts like Carolina or Florida. There's not a lot of shooting talent like better teams.

They competed most nights, got decent goaltending, and executed well in their few ozone chances. Tourigny avoided getting caved in most nights like the Sharks, Ducks, and Columbus.

One stat that stands out: very few dzone giveaways. AZ had 280, good for 5th, Montreal had 642. 10th in hit share.

The only thing keeping this team bottom 10 is penalty share (worst) and faceoffs (3rd worst).

 

[Sinurgy]

xG is shots+shot quality. A point blank chance would result in a large xG number, a flip in from the point would not.

What are the possible values for shot quality, is there just a few options like say high quality is 10pts, medium quality 5pt, low quality 1pt or are their several possible values?

 

[MIGs Dog]

I've also wondered, how many people are watching every game and recording every minute stat? A job like that must be exhausting.

 

[Canis Latrans]

What are the possible values for shot quality, is there just a few options like say high quality is 10pts, medium quality 5pt, low quality 1pt or are their several possible values?

You just look at all the shots that have occurred from a general location and chart how often they go in. The higher ones are binned together as high quality and they map out generally to closer to the net.

Then if you're shooting it in more often than what the shot quality says should go in, you have higher expected goals. I think that's the gist of it.

It doesn't factor in important considerations like is there a defenseman actively defending the shooter and if they screened the shot, or was the prior pass cross-ice and forcing the goaltender to move moreso than just skating there would allow. You could try and account for those and I think models do when you break down the shot type that is in the NHL game logs, like if it tracks one-timers that obviously are more dangerous than skating up and tossing a backhander, generally. However, unless you have tons of data and have clear events types defined, it may just be too noisy to get a real effect.

 

[Canis Latrans]

I've also wondered, how many people are watching every game and recording every minute stat? A job like that must be exhausting.

I think that's what Chayka's company was doing, manually defining things beyond what the NHL game logs ever say. It's what Sport Logiq is doing too. It's mostly microstat tracking.

Interesting, with that Big City Greens game they did, apparently they now are visually tracking several points on players down to multiple points on the sticks. They'd need to behaviorally identify what things mean, but you can do that now with stuff like this: DeepLabCut — The Mathis Lab of Adaptive Motor Control

It's just going to get automatized and teams will have data scientists and behavioral scientists teasing apart the game at some point.

 

[XX]

What are the possible values for shot quality, is there just a few options like say high quality is 10pts, medium quality 5pt, low quality 1pt or are their several possible values?

MoneyPuck.com -About and How it Works

About MoneyPuck, Models, Analytics, Corsi, Fenwick

moneypuck.com

Evolving-Hockey

evolving-hockey.com

Expected Goals Model with Pre-Shot Movement, Part 1: The Model

There are few questions in hockey analytics more fundamental than who played well. Consequently, a large portion of hockey analysis has been focused on how to best measure results. This work is som…

hockey-graphs.com

Each model is slightly different with the variables and weighting but it's a sliding value that takes many different factors into account. The models are also based on and tested against historical scoring data, so from a given spot they can pretty reliably determine a league average.

 

[SR]

For what it’s worth, I’ve heard matthews is wanting to come here pending the may vote.

 

[Jamieh]

I don't like to complain too much about Craig as he's the only one really writing Coyotes stuff but I really wish he wouldn't retweet himself a dozen times for every single tweet!!

 

[PuckLife]

What are your thoughts on the latest line juggles? What's the thinking?

 

[Sinurgy]

For what it’s worth, I’ve heard matthews is wanting to come here pending the may vote.

I'd love Matthews but he's going to be way too expensive. If this rebuild goes right, they're going to have several big salaries to contend with already and I don't see Matthews taking a big enough of a home town discount to make up for it. I mean when the time comes, if he's truly interested you have to at least explore the possibility, but as much as I love the idea I just don't see it happening.

 

[FinnishCoyote]

I'd love Matthews but he's going to be way too expensive. If this rebuild goes right, they're going to have several big salaries to contend with already and I don't see Matthews taking a big enough of a home town discount to make up for it. I mean when the time comes, if he's truly interested you have to at least explore the possibility, but as much as I love the idea I just don't see it happening.

If he would sign for 13-14ish per year you pay it instantly.


But i doubt he signs here

 

[Kai Yo T]

If he would sign for 13-14ish per year you pay it instantly.


But i doubt he signs here

I don't think we can afford that until the TED is making money for Meruelo. Aren't we talking about getting rid of Schmaltz over budget concerns? XG using words like "lean" and all?

Edit: Even if Matthews increased ticket sales we're limited to +/- 5000 seats at Mullett.

 

[Jakey53]

For what it’s worth, I’ve heard matthews is wanting to come here pending the may vote.

Were your pants wet when you woke up? Who did you hear that from?

 

[SR]

Were your pants wet when you woke up? Who did you hear that from?

Ha. I’m not necessarily sure I even want us to sign matthews to what he’s likely going to sign for. But, the source is the same as it’s always been and it’s usually factual. Take it for what it is.

 

[Coyotedroppings]

Ha. I’m not necessarily sure I even want us to sign matthews to what he’s likely going to sign for. But, the source is the same as it’s always been and it’s usually factual. Take it for what it is.

I’ll admit, you’ve had some good intel over the years. My sources have all dried up, think I need to change my username.

 

[SR]

I’ll admit, you’ve had some good intel over the years. My sources have all dried up, think I need to change my username.

I lost a couple that were batting .1000, This one isn’t bad, but is in the know as a relatively important employee of the team. Who knows anymore. I’ll share what I hear with my f40 family and if it pans out great if not, we keep moving.

 

[Dirty Old Man]

So, if we decide we wanna win game 82 tomorrow, according to CapFriendly we have...$1.44 Billion with a B in cap space

 

[TheLegend]

So, if we decide we wanna win game 82 tomorrow, according to CapFriendly we have...$1.44 Billion with a B in cap space

 

[Mosby]

I'm back and forth on Matthews, but admit I could sustain on the tears of Leafs fans and their media for years if it happened.

 

[cobra427]

You just look at all the shots that have occurred from a general location and chart how often they go in. The higher ones are binned together as high quality and they map out generally to closer to the net.

Then if you're shooting it in more often than what the shot quality says should go in, you have higher expected goals. I think that's the gist of it.

It doesn't factor in important considerations like is there a defenseman actively defending the shooter and if they screened the shot, or was the prior pass cross-ice and forcing the goaltender to move moreso than just skating there would allow. You could try and account for those and I think models do when you break down the shot type that is in the NHL game logs, like if it tracks one-timers that obviously are more dangerous than skating up and tossing a backhander, generally. However, unless you have tons of data and have clear events types defined, it may just be too noisy to get a real effect.

I think the advanced stats are more for confirming what you think you see or for looking at biases in data on an individual player level. It can be a small part of the equation in todays NHL. Any slight edge is worth it if you can find it somehow.

The Coyotes were 32nd in GF and 30th in GA last year, this year they sit at 27th in GF and 24th in GA, so up 5 to 6 spots. 31st place last year to 27th place this year, 57 points to 69-71 points, so +12 points and up 4 spots.

Given the way BA has sold off assets, I would say this year has seen solid improvement. Next year, if we add another 12-15 points, and move up 5 or 6 more spots, we will be close to the playoffs but just miss. I think that is a reasonable goal.

 

[moosemeister]

I'm back and forth on Matthews, but admit I could sustain on the tears of Leafs fans and their media for years if it happened.

This is pretty much the only reason why I want it to happen. Haha

 

[Jakey53]

I think the advanced stats are more for confirming what you think you see or for looking at biases in data on an individual player level. It can be a small part of the equation in todays NHL. Any slight edge is worth it if you can find it somehow.

The Coyotes were 32nd in GF and 30th in GA last year, this year they sit at 27th in GF and 24th in GA, so up 5 to 6 spots. 31st place last year to 27th place this year, 57 points to 69-71 points, so +12 points and up 4 spots.

Given the way BA has sold off assets, I would say this year has seen solid improvement. Next year, if we add another 12-15 points, and move up 5 or 6 more spots, we will be close to the playoffs but just miss. I think that is a reasonable goal.

I agree, except I think analytics is more than a small part in the equation now a days. You are looking at the results and then trying to figure out where we should be next, where analytics helped you get those results, if that makes sense. Where analytics gets tricky is which one's the team puts the most emphasis on, and how big is analytics in the overall equation for a team. Makes for a more interesting chess game.

 

[MIGs Dog]

We should sign Toews. Could be a great mentor to our young team.

 

[hbk]

We should sign Toews. Could be a great mentor to our young team.

Is he over long covid? And is he willing to play for $1M or less?

 

[MIGs Dog]

Is he over long covid? And is he willing to play for $1M or less?

He'll PROBABLY get over COVID and he should come at a big discount. Plus, we need to fill some cap space.