Do 'Expected' goals statistics suck?

[Filthy Dangles]

I feel like everytime I go under the hood and really look at them, they just do not represent reality.

For example, Moneypuck had this Justin Danforth goal as .16 XG, which to my understanding is them saying this goal should only go in 16/100 times on an average NHL goalie (please correct me if I am wrong). IMO, it's closer to 96/100 than it is 16.




Idk who watched the Rangers/Jackets tonight but it's absolutely laughable that these stats 'say' the Rangers should have won that game most of the time. They were absolutely dreadful defensively and in pretty much all phases of the game.

I feel like I laugh my ass off at the 'Win O Meter' more than half the time they get posted in GDTs/PGT's (again as being wrong and unreflective of reality)

Help me. Am I misinterpreting the information? Should these stats and source (Moneypuck) be thrown out the window? Is there a better source (where you can actually pinpoint specific shots/goals and see their 'expectedness')

 

[SteelCityCannon]

I pretty much hate all the"advanced statistics". They're absolutely meaningless. Way too many variables in hockey.

 

[Mobiandi]

Expected goals has always been a silly stat

 

[Fixed to Ruin]

Yes

 

[Coffee]

Of course it does

 

[ijuka]

I believe expected goals as a predictor for goals are like a 0.36 for correlation.

They're based on historical data, so it means that shots with the same parameters as that one have historically ended up becoming a goal around 16% of the time.

It's a statistical model, they don't actually look at the videos, they just go by the event data.

Also, here the goalie didn't play it correctly at all. If he had moved even a little bit, it wouldn't have looked nearly as free as it was.

 

[Nogatco Rd]

I feel like everytime I go under the hood and really look at them, they just do not represent reality.

For example, Moneypuck had this Justin Danforth goal as .16 XG, which to my understanding is them saying this goal should only go in 16/100 times on an average NHL goalie (please correct me if I am wrong). IMO, it's closer to 96/100 than it is 16.




Idk who watched the Rangers/Jackets tonight but it's absolutely laughable that these stats 'say' the Rangers should have won that game most of the time. They were absolutely dreadful defensively and in pretty much all phases of the game.

I feel like I laugh my ass off at the 'Win O Meter' more than half the time they get posted in GDTs/PGT's (again as being wrong and unreflective of reality)

Help me. Am I misinterpreting the information? Should these stats and source (Moneypuck) be thrown out the window? Is there a better source (where you can actually pinpoint specific shots/goals and see their 'expectedness')

That ‘deserve to win o meter’ is worthless and undermines the many good things that site does put out. I suspect they know this but leave it up since it drives so much traffic to their site (there are always fans looking for reasons their team “should’ve” won)

 

[norrisnick]

To quote one Redmond, Mickey... "expected by whom?"

 

[Chips]

Deserv

I feel like everytime I go under the hood and really look at them, they just do not represent reality.

For example, Moneypuck had this Justin Danforth goal as .16 XG, which to my understanding is them saying this goal should only go in 16/100 times on an average NHL goalie (please correct me if I am wrong). IMO, it's closer to 96/100 than it is 16.




Idk who watched the Rangers/Jackets tonight but it's absolutely laughable that these stats 'say' the Rangers should have won that game most of the time. They were absolutely dreadful defensively and in pretty much all phases of the game.

I feel like I laugh my ass off at the 'Win O Meter' more than half the time they get posted in GDTs/PGT's (again as being wrong and unreflective of reality)

Help me. Am I misinterpreting the information? Should these stats and source (Moneypuck) be thrown out the window? Is there a better source (where you can actually pinpoint specific shots/goals and see their 'expectedness')

Deservee to win o meter is shit, I say as someone who likes advance stats.


You’re better off looking at different individual stats and trying to interpret them yourself.

Expected goals largely correlates with on ice results. Players who get real goals at a way higher or
lower rate than expected tend to level back.

 

[NjDevsRR]

IMO. I think using them in big sample sizes is good. For example, if you look at season to season xGF stats. Majority of the good teams are in the top half , with a few outliers. So it is largely doing a good job identifying good teams.

But using it in single game samples will give you some crazy results many times.

 

[Goptor]

Its one of those situations where you make do with what you've got.

The NHL keeps a lot of data private that public models can't access. The teams also hire anyone talented enough to develop better models and data collection methods and then wipe their work from public access.

You've already hit on one of the major flaws of 'expected goals'. The idea that every goalie has average ability and every skater has an average shot.

 

[Three On Zero]

Most advanced stats are heavily flawed and based too much on subjective opinions by whom is creating them, or interpreting them. Hockey it’s far too reliant on a 5 player system to use stats accurately. They don’t account well for outside sources that can impact them.

 

[StlBigFly]

I feel like everytime I go under the hood and really look at them, they just do not represent reality.

For example, Moneypuck had this Justin Danforth goal as .16 XG, which to my understanding is them saying this goal should only go in 16/100 times on an average NHL goalie (please correct me if I am wrong). IMO, it's closer to 96/100 than it is 16.




Idk who watched the Rangers/Jackets tonight but it's absolutely laughable that these stats 'say' the Rangers should have won that game most of the time. They were absolutely dreadful defensively and in pretty much all phases of the game.

I feel like I laugh my ass off at the 'Win O Meter' more than half the time they get posted in GDTs/PGT's (again as being wrong and unreflective of reality)

Help me. Am I misinterpreting the information? Should these stats and source (Moneypuck) be thrown out the window? Is there a better source (where you can actually pinpoint specific shots/goals and see their 'expectedness')

Imo

Hockey prediction is closer to pinball than baseball. Much closer.

Statistical comparison and probabilistic outcomes are not something many of our brains like but the puck is either on edge or flat when the one timer comes through, the goalie perhaps can’t even see that an opponent is shooting, the puck hits a guy in the pants, then a skate and goes in 5 hole.

This is pure chaos.

Compare this to: guy throws his 52nd fastball and guy swings.

One is way easier to model than the other, much more predicable with results that are way easier to measure even in weird ways.

What advanced stats do and are is just a handful of measuring sticks. They attempt to place the same measuring stick on every team, player, and event. They can’t help their job is to define chaos.

The root of your problem is ice hockeys chaotic nature, not the evaluation of it.

Understanding their fundamental source will allow you to recognize them for what they are.

I’d suggest that they’re a great tool that joins others to help describe historic events and predict future ones.

 

[Freudian]

Before xG people had to watch the game to claim their team that lost deserved to win. Now they don't have to do that. Progress.

 

[JaegerDice]

xG models are good and useful PREDICTORS over large samples.

That is why the best xG teams tend to perform best over the course of a season or playoff run.

The deserve-to-win ometer is for one game, and using predictive measures to describe how a game went isnt particularly useful.

 

[Bounces R Way]

They are total junk in their current form. A 4th line plug shoveling three successive short range shots directly into a goalie who has the post sealed off at the side of the net is worth something like 10x the xGs of an open Ovechkin cross ice one timer. There's too much noise in a flow game like hockey to have the gall to claim to know how many goals you can expect to score from math.

That isn't to say they're not worth recording or considering, or that the concept isn't worth refining, but they need to be taken as part of a picture. And that part should be around 20% or so IMO.

 

[Bank Shot]

They are total junk in their current form. A 4th line plug shoveling three successive short range shots directly into a goalie who has the post sealed off at the side of the net is worth something like 10x the xGs of an open Ovechkin cross ice one timer. There's too much noise in a flow game like hockey to have the gall to claim to know how many goals you can expect to score from math.

That isn't to say they're not worth recording or considering, or that the concept isn't worth refining, but they need to be taken as part of a picture. And that part should be around 20% or so IMO.

Rebounds are the most dangerous shots statistically.

You are implying that 4th line plugs can juice their numbers by getting flurries of rebound chances, but they don't realistically get those chances often at all. Most bad 4th liners average less than a shot per game.

Ovechkin generally generates 3-4 shots per game. A lot of Ovechkin's chances aren't grade A chances, that is why is career shooting percentage is relatively low for a first line player. Ovie just gets more chances than almost everyone else though.

 

[HugginThePost]

 

[SoundAndFury]

xG basically got straight-up copy/pasted from football (soccer if you insist) where it works very well due to the particularities of the game. And even there the model will be improved at some point to pay more attention to the positioning of the defensive team.

In hockey, where the goalie is 30 cm out of position makes a world of difference a stat that only considers the position from where the shot was taken and none of the other factors is just, basically, redundant and can only be applied in the broadest of applications.

Having said that, the premise of this thread, that a particular play or match is the basis to discard a statistical category, is also rather silly and reflects a fundamental misunderstanding of how statistics work. However, I agree with the notion that "win o'meter" is just a cringe cope device that pounces on this misunderstanding.

 

[Gaylord Q Tinkledink]

Yes, but the win-o-meter is by far the worst.

 

[Shanahanigans]

Individual game xg or any small sample size is pretty useless. Over a full season though, they will tell you who controls play. Now controlling play is important but it isn’t everything - shooting talent and goaltending matter too. It’s why Carolina basically leads the league in xgf% every year but get bounced when faced with teams that have similar , or even slightly worse numbers but better shooting talent. But , those saying advanced stats are useless don’t know what they’re talking about. For reference, these 4 teams were the best expected goals teams in the NHL last year. It matters quite a bit.

 

[majormajor]

What better predicts how many goals per game a team will score next month - the number of goals per game it scored this month or the xG per game it put up this month?

Or what about year to year?

If it predicts better than the current goal pace then it is already very valuable. I'd like the xG sucks crowd in here to consider that.

The "deserve to win'o meter" is kind of tongue in cheek. You're not supposed to take it too seriously.


----------------

On the Danforth goal (holy Werenski!) there is a cross crease pass right before the shot (royal road pass). At one point there were analysts trying to figure out how to build the pre-shot movement into the model, so that, for example, a shot from that spot with a goalie set is like .08 xG while a shot after a cross crease pass is a .4 xG. I would assume that that effort didn't end up in the public models like moneypuck - perhaps pre-shot movement is now built in to the models that teams are using.

 

[ijuka]

They are total junk in their current form. A 4th line plug shoveling three successive short range shots directly into a goalie who has the post sealed off at the side of the net is worth something like 10x the xGs of an open Ovechkin cross ice one timer. There's too much noise in a flow game like hockey to have the gall to claim to know how many goals you can expect to score from math.

Flurry-adjusted expected goals was specifically made to deal with this.
Without flurry:
0.7 + 0.7 + 0.7 = 2.1 xG
With flurry:
0.7 + 0.21 + 0.063 = 0.973 xG

Also, seems like you just don't understand that expected goals aren't tied to the shooter. That doesn't mean that you cannot adjust for the shooter separately from expected goals, by applying shooter-dependent modifiers on the raw expected goals-value. This, of course, is what you should be doing. Using it as an argument against expected goals is, frankly, ridiculous.

It's akin to saying that a player's shots on goal aren't worth tracking because players score at different %s. However, in reality, you should consider both the shooting frequency and the shooting percentage to get the most complete view. It's not rocket science, merely common sense.

 

[Beezeral]

I feel like everytime I go under the hood and really look at them, they just do not represent reality.

For example, Moneypuck had this Justin Danforth goal as .16 XG, which to my understanding is them saying this goal should only go in 16/100 times on an average NHL goalie (please correct me if I am wrong). IMO, it's closer to 96/100 than it is 16.




Idk who watched the Rangers/Jackets tonight but it's absolutely laughable that these stats 'say' the Rangers should have won that game most of the time. They were absolutely dreadful defensively and in pretty much all phases of the game.

I feel like I laugh my ass off at the 'Win O Meter' more than half the time they get posted in GDTs/PGT's (again as being wrong and unreflective of reality)

Help me. Am I misinterpreting the information? Should these stats and source (Moneypuck) be thrown out the window? Is there a better source (where you can actually pinpoint specific shots/goals and see their 'expectedness')

Change it to score adjusted and the math completely flips. You used one metric and didn’t look at the rest. 50+%. Of the Rangers chances came in the 3rd when the game was no longer in doubt.

 

[Machinehead]

Just gonna post what I told you in the Rangers thread.

I do think, to varying extents, these sites underrate scoring chances. These models were developed during the Dead Puck Era, and by and large, xG haven't entirely kept up with actual league scoring going up the past few years.

That being said, you're massively overcorrecting. 16% is a pretty good chance. It might be closer to 25% given what I just said, which is an amazing chance.

The goaltender was still playing his position on that play. The majority of the time, a shot like that hits the goalie or misses the net.

The worst chance you can face as a goalie is a 2 on 0. Do 2 on 0 chances go in 96% of the time?

And I'll add that out of all the sources I use, Moneypuck is my last favorite.

What's the problem with Moneypuck? They give out these ridiculous scoring chances, which get disproportionately high xG counts and then those shots, lo and behold, don't go in. That's a valid criticism except it's exactly what you're suggesting they should do.

Most importantly, this entire readout can be explained by the fact that the Rangers had three more powerplays and played some of the third period with their net empty. You can criticize analytics but I don't know how it's 2025, and we're still doing that by ignoring 5v5 play. The nerds have been screaming "5v5!" at the top of their lungs for 11 years.

NaturalStatTrick, which I trust the most, gave Columbus an advantage at 5v5. And not a huge one, but that tracks with Quick giving up two really bad goals and Columbus basically taking most of the third period off with a big lead. And they still had the advantage in 5v5 xG!