Post-Game Talk: We suck (Well, the Players, coaches, management, Ice Caretakers)

[tarheelhockey]

But PDO predicts that their shooting percentage will go down and ours will go up soon enough. Someday, right?

The scary thing is there are PDO enthusiasts who believe it actually works this way.

Maybe my least favorite stat after +/-

 

[NotOpie]

I'm one of the few folks who thinks that shooting percentage is more complex of an indicator than most. It is a combo of pure skill and timing. It is also about positional awareness, basically getting into the best position to shoot. It is a partial indicator of how good of a passing team you are. If you're getting juicy feeds in the right spot and at the right time, it's easier to bury them (unless your Jordan Staal or Joakim Nordstrom, even the occasional Teuvo Teravainen). But it also contains a luck multiplier. So arguably, shooting percentage is a pretty good proxy for how good your team is offensively. The top 10 shooting % teams are currently all playoff teams except for the high scoring Islanders and the anomaly, Avalanche.

The Hurricanes, on the other hand, during the Peters' years have sported shooting % of 7.3%, 8.0%, 8.5%, and this year's 7.7% - all in the bottom 1/3 of the league. Considering that the league average shooting % has ranged tightly between 9.1% and 9.2% over the same period, it is pretty clear that the Canes have been bad offensively over the same timeframe. One only has to look at a team like the Islanders to see how far good offensive skill can take you. They sport the 3rd highest shooting % (10.6%) in the league and an abysmal SV% of .901. Still they are hanging around a playoff spot...just like us. Could you imagine if we were just a shade better on the offensive side of the board?

It happens enough years in a row to know it's not bad luck.

I like to look at luck as a multiplier of sorts...in our case it is sort of like multiplying by zero.

 

[Boom Boom Apathy]

@NotOpie what was the Canes shooting % the 4 seasons prior to Peters to compare to the 4 seasons he's been here?

 

[cptjeff]

I'm one of the few folks who thinks that shooting percentage is more complex of an indicator than most. It is a combo of pure skill and timing. It is also about positional awareness, basically getting into the best position to shoot. It is a partial indicator of how good of a passing team you are. If you're getting juicy feeds in the right spot and at the right time, it's easier to bury them (unless your Jordan Staal or Joakim Nordstrom, even the occasional Teuvo Teravainen).

This is why I don't generally put much stock in advanced stats- they're often just more and more complex manipulations of a very limited set of data, and there are way too many interacting variables in hockey for any of that stuff to actually be predictive across all strategies and styles of play. It's extraordinarily clear that shooting percentage is not a random thing, but may of the statheads model it as such because that's the only way they can make their math look right, not because it actually models what you see on the ice. Happens with baseball, too. Framing pitches was a myth from all the old people until the stats dorks found a way to quantify it, then it was a critical skill for catchers again.

 

[Surrounded By Ahos]

@NotOpie what was the Canes shooting % the 4 seasons prior to Peters to compare to the 4 seasons he's been here?

Since 2010-2011 they have gone thusly:

9.2
8.5
8.2
8.0
7.2<—-Pastor Bill takes over
8.0
8.5
7.7

 

[My Special Purpose]

I stole this graph from Tyler Dellow's latest, The NHL's biggest 5-on-5 lineup black holes and the trade deadline, on The Athletic [paywall]. I don't think I'm breaking any rules. I'm not going to repost the whole story, but suffice it to say the Canes get a big writeup with our "first line" and first d-pairing both showing up very high (er, low) in relative goal differential.

I will liberate one paragraph:

Carolina has their first line and first pair close to the top of this list, a remarkable achievement for a playoff contender. Longtime observers of the Hurricanes won't be surprised to learn that this is mostly a shooting and save percentage issue. It's always a shooting and save percentage issue with the Hurricanes.​

If the playoffs are the goal this season, there's no way around the fact that we need to make a move for a goalie. Our .912 save percentage at 5-on-5 is crippling. If we don't, I'll understand. It's hard to carry three well-paid goalies, but attempting to fix anything else while not addressing goaltending, probably won't lead anywhere.

 

[NotOpie]

Since 2010-2011 they have gone thusly:

9.2
8.5
8.2
8.0
7.2p>
8.0
8.5
7.7

Beat me to it...since our last playoff appearance we were #10, #12, #22, #25, #25, #29, #28, #20, and now currently #28 in the league (shooting %). Interestingly enough, in 08-09 our shooting % was only 8.8%, good enough for 25th in the league. But that season we were 8th in goals against, and 14th in shots against....all the rest of our stats were fairly mediocre. However, Cam Ward had one of his best seasons, only giving up a total of 160 goals and coming in 3rd in the league in wins with 39.

 

[NotOpie]

....and for all of you PDO fans, the Carolina "Exercises in Futility" have had only a single season above the magical 100 - 2010-11, where we hit 100.4. As bad as our shooting % has been, our aggregate SV% pales in comparison.

 

[What the Faulk]

This is why I don't generally put much stock in advanced stats- they're often just more and more complex manipulations of a very limited set of data, and there are way too many interacting variables in hockey for any of that stuff to actually be predictive across all strategies and styles of play. It's extraordinarily clear that shooting percentage is not a random thing, but may of the statheads model it as such because that's the only way they can make their math look right, not because it actually models what you see on the ice. Happens with baseball, too. Framing pitches was a myth from all the old people until the stats dorks found a way to quantify it, then it was a critical skill for catchers again.

See, when you say things like this and use terms "stathead", it makes you seem like you don't know what you're talking about and that you have a blinding bias against these things. I don't recall anyone saying shooting percentage is "luck", or trying to force it into a model like you're saying, just that there are generally upper and lower bounds to NHL skill and these sorts of numbers (based on historical data), and if you stray too far from them, you're probably getting some bad/good bounces. I don't even think many people use PDO as a proxy for luck anymore. Sustainable pace is probably a better way of putting it. Remember when Alex Steen shot 15%+ a few years ago and scored 33? It's the only time in his career he's done that. In fact, he's only eclipsed 20 goals in 2 of his other 12 seasons. He turned it into a contract worth a $5.8M cap hit, which was 8.4% of the cap (about $6.3M in 17-18 numbers). Is he worth that now? Probably not (his hit on his new contract is 50k less, but it's a smaller percentage of the cap). This is useful information to have.

Many of these statistics have a correlation to winning percentage, but it is not 1, so obviously seasons like Steen's and teams like the Hurricanes exist.

 

[Finlandia WOAT]

I stole this graph from Tyler Dellow's latest, The NHL's biggest 5-on-5 lineup black holes and the trade deadline, on The Athletic [paywall]. I don't think I'm breaking any rules. I'm not going to repost the whole story, but suffice it to say the Canes get a big writeup with our "first line" and first d-pairing both showing up very high (er, low) in relative goal differential.

I will liberate one paragraph:

Carolina has their first line and first pair close to the top of this list, a remarkable achievement for a playoff contender. Longtime observers of the Hurricanes won't be surprised to learn that this is mostly a shooting and save percentage issue. It's always a shooting and save percentage issue with the Hurricanes.​

If the playoffs are the goal this season, there's no way around the fact that we need to make a move for a goalie. Our .912 save percentage at 5-on-5 is crippling. If we don't, I'll understand. It's hard to carry three well-paid goalies, but attempting to fix anything else while not addressing goaltending, probably won't lead anywhere.

What conclusion are we supposed to draw from this? Staal-Slavin and friends have awful goal differential, and it's because a combination of crap goaltending and low shooting percentage?

Also, shooting percentage is "luck" in the sense that there are factors involved in first generating a shot on goal, and then that shot on goal turning into a goal, that are outside of the player's individual control. The best example being who is present on the ice at the time of the attempted shot on goal.

 

[Lempo]

The scary thing is there are PDO enthusiasts who believe it actually works this way.

Maybe my least favorite stat after +/-

This is the point where I as a rule make a point of pointing out that being in team Top 3 in +/- in forwards/defenders (whichever applicable to the player) is an ELC Schedule A Performance Bonus category, that will fetch you a nice sum of $212.500 if attained.

(minimum 42 Regular Season games played by the Player and the comparison group)

 

[The Stranger]

See, when you say things like this and use terms "stathead", it makes you seem like you don't know what you're talking about and that you have a blinding bias against these things. I don't recall anyone saying shooting percentage is "luck", or trying to force it into a model like you're saying, just that there are generally upper and lower bounds to NHL skill and these sorts of numbers (based on historical data), and if you stray too far from them, you're probably getting some bad/good bounces. I don't even think many people use PDO as a proxy for luck anymore. Sustainable pace is probably a better way of putting it. Remember when Alex Steen shot 15%+ a few years ago and scored 33? It's the only time in his career he's done that. In fact, he's only eclipsed 20 goals in 2 of his other 12 seasons. He turned it into a contract worth a $5.8M cap hit, which was 8.4% of the cap (about $6.3M in 17-18 numbers). Is he worth that now? Probably not (his hit on his new contract is 50k less, but it's a smaller percentage of the cap). This is useful information to have.

Many of these statistics have a correlation to winning percentage, but it is not 1, so obviously seasons like Steen's and teams like the Hurricanes exist.

I don't see it much anymore, but in fairness it was a very popular conclusion/narrative within the past few years.

Maybe saying folks were trying to shoehorn data into a model isn't the best way to phrase it...but I would say people overstated their understanding of the game based on statistics and reached some poor conclusions.

 

[NotOpie]

I do believe it is pretty fair to say that shooting % reflects pretty broadly on how good of an offensive team you are (for the reasons previously stated). Sure luck plays into it, but not nearly as much as skill, positioning, and timing. Save %, to a certain extent, can also be a bit of a proxy for how good you are defensively. Those tenders with high save percentages are likely playing in front of a defensive corps that keeps guys to the outside, blocks a lot of shots, or has a good take-away ratio. Essentially preventing high danger shots...of course there are outliers, like goalies who just plain suck.

 

[What the Faulk]

I don't see it much anymore, but in fairness it was a very popular conclusion/narrative within the past few years.

Maybe saying folks were trying to shoehorn data into a model isn't the best way to phrase it...but I would say people overstated their understanding of the game based on statistics and reached some poor conclusions.

I will admit to being guilty of this. Our understanding of the game and the data behind it is constantly evolving.

 

[My Special Purpose]

What conclusion are we supposed to draw from this? Staal-Slavin and friends have awful goal differential, and it's because a combination of crap goaltending and low shooting percentage?

Also, shooting percentage is "luck" in the sense that there are factors involved in first generating a shot on goal, and then that shot on goal turning into a goal, that are outside of the player's individual control. The best example being who is present on the ice at the time of the attempted shot on goal.

He was attempting to pinpoint a single position at which each team in "contention" would benefit most from a deadline deal, by isolating their biggest "black hole." Our black hole is goaltending. Not exactly a debatable point.

 

[Finlandia WOAT]

He was attempting to pinpoint a single position at which each team in "contention" would benefit most from a deadline deal, by isolating their biggest "black hole." Our black hole is goaltending. Not exactly a debatable point.

Cool your buns, kev, for once I wasnt trying to pick an argument. Just a bit unsure what point you *were* making.

 

[tarheelhockey]

I don't even think many people use PDO as a proxy for luck anymore. Sustainable pace is probably a better way of putting it. Remember when Alex Steen shot 15%+ a few years ago and scored 33? It's the only time in his career he's done that. In fact, he's only eclipsed 20 goals in 2 of his other 12 seasons. He turned it into a contract worth a $5.8M cap hit, which was 8.4% of the cap (about $6.3M in 17-18 numbers). Is he worth that now? Probably not (his hit on his new contract is 50k less, but it's a smaller percentage of the cap). This is useful information to have.

The mere existence of PDO is nonsensical, though. Why are we combining shooting percentages and save percentages into one number? It’s as silly as including SHG in +/-.

 

[geehaad]

Cool your buns, kev, for once I wasnt trying to pick an argument. Just a bit unsure what point you *were* making.

You detected a unusually dismissive tone.
I detected a standard dismissive tone.

 

[My Special Purpose]

What conclusion are we supposed to draw from this? Staal-Slavin and friends have awful goal differential, and it's because a combination of crap goaltending and low shooting percentage?

Also, shooting percentage is "luck" in the sense that there are factors involved in first generating a shot on goal, and then that shot on goal turning into a goal, that are outside of the player's individual control. The best example being who is present on the ice at the time of the attempted shot on goal.

You detected a unusually dismissive tone.
I detected a standard dismissive tone.

What about the dismissive tone of the first post? Or does it only count when I do it?

Cool your buns, kev, for once I wasnt trying to pick an argument. Just a bit unsure what point you *were* making.

But I'm the jerk.

 

[Finlandia WOAT]

What about the dismissive tone of the first post? Or does it only count when I do it?



But I'm the jerk.

I didn't write the first post with dismissive intent. You posted an out of context chart and paragraph from a site with a paywall. I just wanted a more specific explanation of what point you were trying to prove with the chart. I don't disagree with that point either, at all, in fact.

 

[The Faulker 27]

Everyone here is a jerk. It's part of the reason I can't stop coming back. Jerks.

 

[Lempo]

So you come for the jerks to a place that prides itself for the sucking.

 

[Canes]

I like a lively discussion on statistics. Reminds me of arguments I had and read about with sabermetrics back when Moneyball was just being popularized.

 

[tarheelhockey]

Dicks.

 

[The Faulker 27]

So you come for the jerks to a place that prides itself for the sucking.

I guess I keep hoping things are gonna change.