Pre-Game Talk: Killed in Kowtown

[Drivesaitl]

That is just small sample size illusion, 32 teams and they have 25-30 games both in home and away. It is pretty likely that there are couple anomalies. No team is better in away games if you look at big enough sample size. And if you want to put a lot of weight on small sample sizes, Calgary has won 11 out of their 12 last home games.

The oilers are better on the road, and its no short sample, its the whole bleeding season. There is no home advantage when you're the Edmonton Oilers. The bolded is just silly, because people are speaking of the Oilers record. not a bunch of other clubs.

A number of posters have mentioned possibilities for why the Oilers struggle at home. Its quite plausible that a team due to unique roster, circumstance, will struggle at home. One factor even is the Oilers even reported preferring playing in front of crowds on the road and disliked playing in empty, or half empty rinks in Canada. I mean the players have flat out said it.

That said it doesn't explain the Montreal game. That was essentially a road game played at Rogers place. jk..

 

[CycloneSweep]

That is just small sample size illusion, 32 teams and they have 25-30 games both in home and away. It is pretty likely that there are couple anomalies. No team is better in away games if you look at big enough sample size. And if you want to put a lot of weight on small sample sizes, Calgary has won 11 out of their 12 last home games.

By this logic you can’t say Arizona sucks. It’s only 50 some games

 

[Drivesaitl]

Expected goals for percentage for every game they have played. Lots of green means that Oilers have created more scoring chances, red is the opposite.

Simply put:
Green is good
Red is bad

Does any of the statistical methodology factor in the scoring probability of the actual players getting the scoring chances?

For instance I would expect the Oilers to generate far more scoring chances than goals, with respect to other clubs. The reason is that vast majority of plays that McDrai setup are not finished. Due to relatively weak wingers. Conversely, McD and Drai get so much focus from opposition that their chances are impacted as well by tight checking. We don't have a one shot winger scorer in the whole lineup, with the exception of Kane, who is too new here to impact the stats.

Its the old proviso I state all the time. Not every scoring chance is similar. One thing is it depends who's getting the chance. If its players like Yama or Foegele getting chances I don't care how many chances they get, probability of it resulting in any goal is remote.

 

[Drivesaitl]

By this logic you can’t say Arizona sucks. It’s only 50 some games

Lets play 100more. Calgary will be bad again by that time..

 

[CycloneSweep]

@McCombo hete is a bigger sample size
Last 4 seasons Edmonton is 65-54-10 at home .543p% -14

on the road
72-50-14 0.581 +10

 

[Mez]

True, but when that first line is outscoring it's opposition 50-18, or 68-28 with any of the top three on the ice, the rest of the team being outscored 62-58 isn't the end of the world.

Right? Holland had an easy job and fked it up. All we needed was good enough depth to keep us even while MCDrai score.

 

[Drivesaitl]

@McCombo hete is a bigger sample size
Last 4 seasons Edmonton is 65-54-10 at home .543p% -14

on the road
72-50-14 0.581 +10

not long enough, hurr durr..

lol

But yeah, thats where most people die on hills on stats. Instead of looking up specific case examples they say something like thats impossible according to other teams data..

The Oilers say hold my beer.

thanky for doing the work for me

 

[McCombo]

The oilers are better on the road, and its no short sample, its the whole bleeding season. There is no home advantage when you're the Edmonton Oilers. The bolded is just silly, because people are speaking of the Oilers record. not a bunch of other clubs.

A number of posters have mentioned possibilities for why the Oilers struggle at home. Its quite plausible that a team due to unique roster, circumstance, will struggle at home. One factor even is the Oilers even reported preferring playing in front of crowds on the road and disliked playing in empty, or half empty rinks in Canada. I mean the players have flat out said it.

That said it doesn't explain the Montreal game. That was essentially a road game played at Rogers place. jk..

That is the difference between us. 25-30 games is a small sample size for me. You just don't get the point. 32 teams, all teams have around 25-30 sample sizes in home and away. It is much more likely that there are teams that have better win percentage in away games, compared to that every team would be better at home.

And definitely playing with no crowd reduces the home ice advantage. That is well documented at least in bigger european soccer leagues during Covid. Most likely in other sports too, but don't have statistics. But that just makes it even more likely to have more points from away than home games, when they played to empty arena in Edmonton at the start of the year.

Pittsburgh at home 15-8-5
Pittsburgh on the road 19-6-4

Washington at home 14-11-5
Washington on the road 16-7-4

Boston at home 16-10-1
Boston on the road 18-8-3

Nashville at home 15-10-0
Nashville on the road 16-10-4

LA at home 14-12-2
LA on the road 17-7-5

There are some crazy good home teams in the same company as Oilers. That is perfect example for small sample sizes, people try to explain anomalies with made up stories when it all is just randomness of a small sample size. You can go dig those teams winning percentage (home vs away) from last 5 seasons for example and we will see if these teams truly are better on the road.

 

[Drivesaitl]

That is the difference between us. 25-30 games is a small sample size for me. You just don't get the point. 32 teams, all teams have around 25-30 sample sizes in home and away. It is much more likely that there are teams that have better win percentage in away games, compared to that every team would be better at home.

And definitely playing with no crowd reduces the home ice advantage. That is well documented at least in bigger european soccer leagues during Covid. Most likely in other sports too, but don't have statistics. But that just makes it even more likely to have more points from away than home games, when they played to empty arena in Edmonton at the start of the year.

Pittsburgh at home 15-8-5
Pittsburgh on the road 19-6-4

Washington at home 14-11-5
Washington on the road 16-7-4

Boston at home 16-10-1
Boston on the road 18-8-3

Nashville at home 15-10-0
Nashville on the road 16-10-4

LA at home 14-12-2
LA on the road 17-7-5

There are some crazy good home teams in the same company as Oilers. That is perfect example for small sample sizes, people try to explain anomalies with made up stories when it all is just randomness of a small sample size. You can go dig those teams winning percentage (home vs away) from last 5 seasons for example and we will see if these teams truly are better on the road.

A poster above just posted that the Oilers have had a far better record on the road, than at home for the last 4 seasons combined.

Again its you that doesn't get it.

I start off saying the Oilers are better off playing on the road. You immediately say thats wrong, not the case, no way thats the case. People will respond to indicate the OILERS ARE WORSE at home. You continue to state that "I don't get it" then spew off again about how OTHER teams due, when the discussion is actually about the Oilers.

 

[McCombo]

Does any of the statistical methodology factor in the scoring probability of the actual players getting the scoring chances?

For instance I would expect the Oilers to generate far more scoring chances than goals, with respect to other clubs. The reason is that vast majority of plays that McDrai setup are not finished. Due to relatively weak wingers. Conversely, McD and Drai get so much focus from opposition that their chances are impacted as well by tight checking. We don't have a one shot winger scorer in the whole lineup, with the exception of Kane, who is too new here to impact the stats.

Its the old proviso I state all the time. Not every scoring chance is similar. One thing is it depends who's getting the chance. If its players like Yama or Foegele getting chances I don't care how many chances they get, probability of it resulting in any goal is remote.

There are good and bad players in every team. This stat doesn't take it into consideration. Also there are problems if a really good build up doesn't end up in a shot attempt. So in small sample sizes, you need to be more careful. But the bigger the sample size gets, the more these things even out. And you need to take a lot more things into consideration than xG, but there isn't really a better stat imo. The top9 in the league includes these teams via NTS: Florida, Calgary, Boston, Toronto, Tampa Bay, Carolina, LA, Pittsburgh and Colorado. LA is the only one who doens't really belong in the top8/9.

Bottom end includes: Arizona, Buffalo, Columbus, Chicago, Rangers (all hail Shesterkin), Ottawa, Montreal and Philly. Elite teams create more scoring chances, that's why they are good. Bad teams give up more than they create, that's why they suck.

 

[Drivesaitl]

There are good and bad players in every team. This stat doesn't take it into consideration. Also there are problems if a really good build up doesn't end up in a shot attempt. So in small sample sizes, you need to be more careful. But the bigger the sample size gets, the more these things even out. And you need to take a lot more things into consideration than xG, but there isn't really a better stat imo. The top9 in the league includes these teams via NTS: Florida, Calgary, Boston, Toronto, Tampa Bay, Carolina, LA, Pittsburgh and Colorado. LA is the only one who doens't really belong in the top8/9.

Bottom end includes: Arizona, Buffalo, Columbus, Chicago, Rangers (all hail Shesterkin), Ottawa, Montreal and Philly. Elite teams create more scoring chances, that's why they are good. Bad teams give up more than they create, that's why they suck.

One of the mistakes I see commonly made in citation of hockey data is people confusing it with randomly occurring data or what would generate normal curve distributions. This is not the case with data that could be heavily impacted by several or endless extraneous third variables. The Oilers are not a random data point which conform to normal data curve. There would be many reasons why, and some have been stated, why the Oilers are better on the road, than at home, and it isn't the case of being simply outlier data that is not conforming to normative distribution due to limited sample.

 

[McCombo]

@McCombo hete is a bigger sample size
Last 4 seasons Edmonton is 65-54-10 at home .543p% -14

on the road
72-50-14 0.581 +10

not long enough, hurr durr..

lol

But yeah, thats where most people die on hills on stats. Instead of looking up specific case examples they say something like thats impossible according to other teams data..

The Oilers say hold my beer.

thanky for doing the work for me

Yeah. No/limited crowds makes this 4 season sample size a lot less accurate when there are full arenas again. That is also roughly 3,5 seasons worth of games. I wouldn't be surprised if there still is a team or two with better road record in that span.

Oilers records in McDavid era

Home

128-104-20
276 points in 252 games. 1.09 PPG.

Away

123-109-27
273 points in 259 games. 1.05 PPG.

Regardless of that it was good piece of information. Thanks Cyclone.

 

[McCombo]

By this logic you can’t say Arizona sucks. It’s only 50 some games

Trying to miss understand everything I say on purpose. If you look at Arizona's point totals, goal differential, xG, corsi and the material they are operating, it all suggest that they suck.

 

[McBooya42]

But we don't. If Russell comes back and Kassian and Archibald do to, our cap space is all gone. If Russell doesn't come back than we would have like...700-800k in cap space?

I heard Russell isn't coming back this year...forget where.