Despite Saros having a 3-7-1 record and a GAA of 2.85, his GSAE is 8th in the league.

[x Tame Impala]

Which of these is more likely:

1. Saros randomly became a bad goalie after being great for years
2. Saros is just as good as before but his team is playing terrible defense, which causes his numbers to look bad when you're not taking the context into account

It's not that complicated.

Expected goals assume a baseline goaltending. If you're saving more goals than expected, then you're outperforming the baseline goaltending.


Yeah, but when you also add that he's 8th place in goals saved above expected, you could assume that his low SV% is a product of the team allowing more high-danger scoring chances against than average.

What I don't understand is just looking at a small subset of stats, coming to conclusions based on those, and just ignoring other statistics that would bring further context to those numbers.

Sounds like we’re coming to a similar conclusion regardless. He’s playing in front of a struggling team which isn’t helping him keep pucks out of the net (GAA). His lower SV% suggests he’s letting in more shots than usual. Some of that is him, some of that is the team. (SV%)

What’s the baseline we’re using for above/below expected? His previous level of play or a league wide expected rate based on shot location? I’m genuinely curious. Also, I obviously haven’t watched a ton of him this year so it’s tough to say how many bad goals he’s given up.

Just trying to get a better understanding here

 

[x Tame Impala]

Whatever helps you sleep at night. Trotz went all in and they'd best pull this out of the fire here soon if they have any chance of making playoffs.

Saros is a top 15 goalie. Why he hell they hitched their wagon to an older smaller goalie like that for 8 years is asinine. Then signed two older slower shoot first wingers with a gaping hole in the middle. Trotz got into the screech on Canada Day.

You don’t know why they signed a top 15 goalie? I don’t question that at all. Nashville won’t stay this bad for long. They’re just off to a rough start. They’re only 1/9th of the way through the season.

 

[Albatros]

Saros is a top 15 goalie. Why he hell they hitched their wagon to an older smaller goalie like that for 8 years is asinine.

Top 15... The only goalie with Vezina votes each of the last four years, lowest placement 6th.

But hey, he's already 29 so I guess they should just get rid of him while they still can. Why take a boat when you can also have the mystery box that could even be a boat?

 

[Reality Czech]

Another example of how subjectively valuable/invaluable these advanced stats are. Saros is 22nd in SV% and 30th in GAA. With a small amount of context all that says to me is that he's not playing well and also playing on a struggling team.

For the life of me I don't get the value in these "above/below expected" stats and I think using them is not seeing the forest through the trees. Fundamentally they cannot account for the inherent dynamic activity that goes into every single puck possession, shot attempt, shot on net, pass, etc...

What does his 8th place ranking say to people who do like these stats? That he's actually a top 10 goalie playing on a bad team?

"IF HE'S A GOOD HITTER WHY DOESN'T HE HIT GOOD?"

I agree advanced stats are way overblown in many cases, especially with regard to goalies, but you could say the same about GAA and Sv%, which are as much a team stat as goalie stat.

Not to mention the fact that a small number of bad games (by the goalie or team in front of him) can totally tank his numbers. If a goalie plays well 8/10 games but then his team leaves him out to dry the other 2 his stats won't look so impressive. But I'd say a goalie who plays well 8/10 games is a good goalie regardless of what the numbers say.

Not talking about Saros specifically, but in general. Solely relying on stats to judge goalies is a fool's endeavor. Just ask Gretzky.

 

[kanucks25]

His lower SV% suggests he’s letting in more shots than usual. Some of that is him, some of that is the team. (SV%)

Yeah but you don't know that for sure without actually watching all his games.

It's very possible his numbers are lower despite playing even better than he usually does.

I'm not saying that is the case here, but a lower SV% doesn't automatically mean he's playing worse.

And that's what expected goals tries to show: how well a goalie is playing relative to the shot quality he's facing, not just shot volume with no context.

That said, every site has their own model and I think it's common knowledge that you should take public expected goal numbers with a grain of salt because it's very questionable how sophisticated their systems are in determining shot quality.

For example, a shot from X location may only have a 10% chance of going in but if there's two screeners in front, it's more like 20%. Or shot from Y location only has 20% chance of going in but if the pass came from a certain location, it's more like 40%.

What’s the baseline we’re using for above/below expected? His previous level of play or a league wide expected rate based on shot location?

League wide average.