Post-Game Talk: @boxcar65

[MessierII]

Its illogical that goals would not count among scoring chances in a tabulation of the same. Perhaps the stat should come with an * and large grain of salt in stating exactly what the metric is.

WHY minus the scoring chances that actually succeeded. To create statistical noise?

Would love to hear the explanation on how goals dont count. They are of course the results barometer of what occurs in a game.

It strikes me that not counting goals is asinine.

I agree that’s so absurd. Analytics are a mess.

 

[McTrashBoat]

The real problem is there is a stat to prove anything you really want to.

*cough* Loose puck retrieval *cough*

No idea why a chance wouldn't be counted if a goal is scored, you still count the shot

 

[Drivesaitl]

HDSC are an objective measurement of shot distance. You have to be within a certain distance of the goal to record one. That's it.

They're worth tracking as statistically you are far far more likely to score from this range than anywhere else, but the name really needs a change because confusion like yours is incredibly common.

Yep. But whether the measure be objective or not, the measure is crude, and not taking into account any of the complexity of the quality of a shot, who is taking it, how the shot was set up, was it screened, was it one times, was it across ice, was it Draisaitl. The measure is so crude as to not really be worth the time and effort at all. Unless somebody just wants to collect fairly meaningless data.

Definitely the metric is misnamed. By excluding goals it positions itself as a metric tabulating non success..

Maybe in the Olympics in shooting events they should calculate misses instead of hits..

That's joking, but it hilites the illogic involved with excluding goals from the sample.

 

[Tobias Kahun]

The real problem is there is a stat to prove anything you really want to.

Oilerbear could bend enough stats to show Marincin is a top 5 D

 

[Louis Cypher]

Where's the Eastern media? McDavid 5 points = silence. But if Matthews aka '1B' had 5 points, they'll croak about him being better than McDavid.

TSN must have read your post. There are 4 articles on the front page about the Oilers. This is a first.

 

[Drivesaitl]

Seems like a stat used by teams that are bad that can't score goals.

Oh don't worry we have a ton of HDSC, who cares that we can't score.

Eakins TM hockey. Blaze away.

 

[oljimmy]

4?!

Yeah but considering we scored 6 goals one has to figure the total of 4 HDSC for the OIlers is entirely whacked. How can the goals scored EXCEED the amount of scoring chances. Some of these metrics are just complete garbage in garbage out. Any bean counter that is tabulating less scoring chances than goals for a team is a flat out biased troll. Probably somebody from Toronto doing the math..

"High Danger" scoring chances. Per Natural Stat Trick. Philadelphia Flyers @ Edmonton Oilers, 2019-10-16

I don't doubt that the 16-4 could be a little exaggerated but it's not far off. Oilers got absolutely caved in HD scoring chances, a great predictor of goals.

 

[oXo Cube]

There is something to be said about the Oilers taking their foot off the gas yesterday as they were poor in the first two periods but only really got caved in in the third.

That of course is a problem all on it's own. The team wasn't playing that well in the first place and then decided to just quit on the game 20 minutes early.

 

[elmeroil]

I don't even care how they are getting these early wins, the early points are so important!

 

[Drivesaitl]

"High Danger" scoring chances. Per Natural Stat Trick. Philadelphia Flyers @ Edmonton Oilers, 2019-10-16

I don't doubt that the 16-4 could be a little exaggerated but it's not far off. Oilers got absolutely caved in HD scoring chances, the #1 predictor of goals.

Read through the last 20 posts. Everybody thinks its ridiculous that goals scored are not included in HDSC. Every Oilers goal scored last night was a High Danger chance. we also had Nuge nearly scoring before Drai did.

In anycase the number is a meaningless description of the game and analytics has even posited such stuff as score effects to explain away such data.

For instance the team that is leading nearly the entire game, and that had a multiple goal lead for a significant part of the game didn't need as many scoring chances..

By my count if goals are added the HDSC is more like 17-10 and less pronounced differential. I'm not even counting the Flyers last two goals which were clear garbage time as the Oilers stopped playing entirely.

ps goals are the #1 predictor of goals.

 

[CantHaveTkachev]

Oilers were not good in the first

and the 2nd period was their worst period of the season...until McDavid's amazing goal
Koskinen was the only reason they were leading 2-1

the 3rd is tougher to read cause they clearly took their foot off the gas and the Flyers were looking to salvage something

 

[oljimmy]

Yep. But whether the measure be objective or not, the measure is crude, and not taking into account any of the complexity of the quality of a shot, who is taking it, how the shot was set up, was it screened, was it one times, was it across ice, was it Draisaitl. The measure is so crude as to not really be worth the time and effort at all. Unless somebody just wants to collect fairly meaningless data.

Ha ha ha you're actually holding the line on this. HDSC is a great (not perfect) predictor of goals scored in a game, not quite as good as Corsi but good. It doesn't correlate perfectly with goals because... wait for it.. there is such a thing as a lucky goal. But a team that relies on lucky goals is going to score less and lose games. This isn't hard.

 

[McDNicks17]

HDSC are defined as a shot or attempt that comes from a certain area.

Goals do not count.

Its simple counting.

Is this actually true?

How do you score without a chance to score?

 

[McDNicks17]

The Oilers Facebook with the "MANN the torpedoes!" title on Mandog's goal is on point.

 

[oXo Cube]

Is this actually true?

How do you score without a chance to score?

Hence why the name "High Danger Scoring Chance" needs to be changed.

The stat is fine the terminology is the problem.

 

[oljimmy]

ps goals are the #1 predictor of goals.

. OH man, this is fun.

Let me break this down for you: we are talking about a game where the Oilers appeared to come away with a lucky win. How would we go about verifying that? Well, if we use your preferred metric (goals scored) lucky wins are literally impossible, they never happen, because the team that scores more goals was by definition more "dangerous". But this is absurd, obviously there is such a thing as a lucky win, right? So we need a stat that doesn't correlate perfectly with goals but is merely a predictor of goals. On all such stats last night, HDSC, Corsi and others, the Oilers got absolutely caved and were getting caved even during the second period when they were only up by a goal (score effects, not relevant there). That is why these stats are useful, they help us to verify (if not 100% prove) that the score was generally lucky for the Oilers. Which it clearly was. They are not "useless" stats made up by "Toronto Trolls" as per your first reply...

 

[Drivesaitl]

Ha ha ha you're actually holding the line on this. HDSC is a great (not perfect) predictor of goals scored in a game, not quite as good as Corsi but good. It doesn't correlate perfectly with goals because... wait for it.. there is such a thing as a lucky goal. But a team that relies on lucky goals is going to score less and lose games. This isn't hard.

Not really following your reply. The Oilers are going to win a lot of games where they are even close in the metric of scoring chances because the coefficient of who is getting those scoring chances is off the charts. For instance consider a coefficient multiplier of McD, Dra, Neal, Nuge variable effect.

The Oilers have scored as many goals as they have because we have two generational producers. The two top producers in the game arguably, on the same team.

Nor will those players be lucky to score. McD, Drai may again be among, or the top two producers in the entire league.

HDSC is essentially meaningless without factoring in several other effects, variables, coefficients.

 

[OilerTitanFan]

Funny how it was around this neck of the woods when everyone was cursing Chia for the signing of koskinen. So far, Koskinen's living up to that 4million dollar contract. It would be funny if Koskinen leads the oilers to a playoff run and wins the conn smythe.

 

[Drivesaitl]

. OH man, this is fun.

Let me break this down for you: we are talking about a game where the Oilers appeared to come away with a lucky win. How would we go about verifying that? Well, if we use your preferred metric (goals scored) lucky wins are literally impossible, they never happen, because the team that scores more goals was by definition more "dangerous". But this is absurd, obviously there is such a thing as a lucky win, right? So we need a stat that doesn't correlate perfectly with goals but is merely a predictor of goals. On all such stats last night, HDSC, Corsi and others, the Oilers got absolutely caved and were getting caved even during the second period when they were only up by a goal (score effects, not relevant there). That is why these stats are useful, they help us to verify (if not 100% prove) that the score was generally lucky for the Oilers. Which it clearly was. They are not "useless" stats made up by "Toronto Trolls" as per your first reply...

The Oilers are only lucky to win this game in that their lineup includes McD, Drai, Neal, Nuge, klef. All of whom had their game going more than any Flyer except Voracek.

I don't even have to look but based on what I'm seeing the Oilers probably have half a dozen players with as many points as any Flyer this season.

The bean counters look at none of that.

The Oilers knew they were winning this game the whole time. The players knew it, were confident, and could score when they needed to.

Breaking it down the trouble with your argument is that NONE of the Oilers goals scored were actually lucky. They were great finishes. Something Philly failed throughout to do.

But the team with the very best players in the game, that were leading almost the entire time, that were leading at one point 6-1, were "lucky to win" according to you.

They probably win this game, against this Flyers team, 8/10 times.

Oilers were winning 6-1 but the slant forehead analytics beancounters have it "that the Oilers were being caved"


lets just agree to disagree

 

[Del Preston]

classic McJadedog...always seeing the worst in the team

That post just needed another misdiagnosed injury to complete my bingo.

 

[Darkwinter]

Funny how it was around this neck of the woods when everyone was cursing Chia for the signing of koskinen. So far, Koskinen's living up to that 4million dollar contract. It would be funny if Koskinen leads the oilers to a playoff run and wins the conn smythe.

If that happen I do not think one Edmonton fan would be upset for eating crow pie

 

[OilerTitanFan]

If that happen I do not think one Edmonton fan would be upset for eating crow pie

There might also be a bring back chia thread especially if players like Bear and a bunch of Chia players step up.

 

[Arpeggio]

What the stat shows is that if McDavid and Draisaitl regress to 1.0 ppg pace, or even a 1.5 ppg pace, the Oilers are going to start losing their fair share of games. While I would be thrilled if McDavid scored 150 points this season, I think most of us can agree that it is unlikely to happen. What the stats tell us is that the Oilers are getting killed at even strength right now, so when the PP normalizes to a more sustainable percentage, the team may be in trouble. Same goes for both goalies currently stopping pucks at a higher percentage than usual.

HDSC is useful in my opinion because most goals are scored from close to the net. So if you give up a ton of those, then over the course of the season you're probably going to give up more goals. Winning 4-5 games while getting out-chanced significantly means you better fix that aspect of your game up before McDavid and Draisaitl slow down.

That said, if McDavid scores 150+ points, it probably won't matter what the advanced stats say.

 

[Arpeggio]

Not really following your reply. The Oilers are going to win a lot of games where they are even close in the metric of scoring chances because the coefficient of who is getting those scoring chances is off the charts. For instance consider a coefficient multiplier of McD, Dra, Neal, Nuge variable effect.

The Oilers have scored as many goals as they have because we have two generational producers. The two top producers in the game arguably, on the same team.

Nor will those players be lucky to score. McD, Drai may again be among, or the top two producers in the entire league.

HDSC is essentially meaningless without factoring in several other effects, variables, coefficients.

The problem is that for the last few games, the Oilers haven't been close in scoring chances. They need to have the puck more, and the bottom six needs to start producing in order to sustain their early success.

 

[McDNicks17]

Hence why the name "High Danger Scoring Chance" needs to be changed.

The stat is fine the terminology is the problem.

Man, that's silly.

Why have a stat that's pointless unless you add HDGF/HDGA to it?