Team defense (11-12 Comparison)

[True Blue]

Except, there's an actual reason to inflate our stats. Are we going to take away 10 randomly bad games from all teams that just had a slump? We're not taking away a slump, we're taking away a transition period. The Flyers are another team that has a great excuse and I HATE the Flyers. The Wings and Pens also have good excuses but for different reasons. If you can't see the difference between 10 games where our team was transitioning to a new coach and 10 games where the Bruins had a slump, for example, then you're just being stubbornly pessimistic.

Did I read this correctly? There is a reason to inflate stats?

 

[SnowblindNYR]

That's fine, my point was that ranking a tampered stat against untampered stats produces a meaningless number. It's a fallacy. Although I'm sure if it was -31- and not me who pointed it out, they'd all be singing his praises.

It's a bigger fallacy to leave those games in is all I'm saying. It just gave us a number which would put us 8th. Doesn't mean we're the 8th best offense. Unless you want to sit there taking out games for other teams using 8th as a guide is decent enough. Probably close to 10-12. 16th is pretty damn misleading is all I'm saying.

 

[True Blue]

We're not, I would be delusional to ever say that. But it's not a logical fallacy. A logical fallacy is including a 10 game stretch of a team that was adjusting to a coach. If you actually think that that wasn't a legit excuse for the horrible stretch then I don't know what to tell you.

Do adjusted stats somehow count more than unadjusted?

Let's stay in the real world. What does the actual documentation show? Adjusted or unadjusted?

The fallacy is indeed present because once you start to inflate stats, you are no longer dealing with things that really happened. Not apples to apples anymore. More like apples to pears.

 

[True Blue]

aufheben.

i bet if you ran the stats readjusted to removing the first 9 games of every team in the NHL, it would still be pretty much the same thing...our EVERYTHING improves.

But that is not quite accurate either. You are then picking the worst time to exclude the Rangers stats and equalizing the rest of the league. To make the analysis normative, you would need to exclude other teams worst 9 games.

 

[SnowblindNYR]

Did I read this correctly? There is a reason to inflate stats?

Yes, context. Following stats blindly, rather than interpreting them is the smart thing to do, any idiot can read numbers and see which one is bigger. Otherwise you wouldn't need analysts, just get some computers. There's a reason Wall Street analysts get paid big bucks, not because they can put something into a computer.

Let me ask you something, you're taking a serve in Tennis. You take a serve once a day for 81 days. The first 10 games your arm is injured and hurts like a mother****er. Your serve is in the 80s. Day 11, your arm feel like nothing happened, since then your serve is in the 90s-100s and once in a while in the 80s but not consistently for 10 days like before. Do you throw out the first 10 serves? Or are should those be included in the average? The latter is how ridiculous people that think the first 10 games should have counted sound.

I'm not surprised all of the pessimists "magically" can't understand this concept.

 

[SnowblindNYR]

But that is not quite accurate either. You are then picking the worst time to exclude the Rangers stats and equalizing the rest of the league. To make the analysis normative, you would need to exclude other teams worst 9 games.

No, because most teams didn't have to deal with that. There are a handful of teams that WOULD be true about, but I don't think anyone here wants to go through this painstaking analysis. Hell, the "experts" don't.

 

[Inferno]

Yeah, too bad they're not sitting in their zone all game blocking shots. That's a TRUE defensive team.

never said that either.

Their issues are clearly they have scrambly moments against teams with movements.

simply switching to a zone coverage, but aggressive zone coverage, would resolve that issue.

i dont like the pack it in Torts D. I prefer to attack the points and force defenseman to make good passes under duress in ALL zones.

playing man to man makes movement tough to handle.

 

[SnowblindNYR]

Do adjusted stats somehow count more than unadjusted?

Let's stay in the real world. What does the actual documentation show? Adjusted or unadjusted?

The fallacy is indeed present because once you start to inflate stats, you are no longer dealing with things that really happened. Not apples to apples anymore. More like apples to pears.

Except it's a ******** reality. The first 10 games were clearly a team adjusting to a radically different system and doesn't represent this team in the slightest. It wasn't just normal seasonal fluctuation like the December slump probably was.

 

[Inferno]

But that is not quite accurate either. You are then picking the worst time to exclude the Rangers stats and equalizing the rest of the league. To make the analysis normative, you would need to exclude other teams worst 9 games.

Disagree to a point.

I think teams that have long term continuity, with players, and coaches, will benefit the least, and teams with new coaches and heavy turnover will benefit the most.

Really, if you want to exclude the worst 9 games, rangers could easily take out our homestand of 9 and would be in the same boat...we've had 2 terrible 9 game stretches, as most teams do. However most of us arent saying take out BOTH of those, we're saying, take out the first 9 games because of all the circumstances that revolve around it.

all teams have circumstances that cause them to lose games.

The flu.

Outdoor games and 24/7 hoopla

injuries

etc.


Rangers, however, seemed to have several confluences at the same time. On the road for a bajillion games.

totally new system.

Henrik Lundqvist playing like Kevin Weekes.

Lots of terribleness to go around in that 9 games to start the season.

If someone wants to re-adjust the standings for all 30 teams, GF, GA, etc.

I bet we improve in every statistical category...

I also think the eyeball test explains it.

Rangers were transittioning from a completely different system and had great difficulties in wrapping their heads around it.

Rangers had to start on the road against some of the best teams in the West.

Lots of crap to overcome when you do that.

I think it's foolish to simply ignore these facts. If you dont want to readjust the stats, that's totally fine. But I think you need to be able to look at this team and say...you know what, since they started to get this system...particularly in the 2nd half of the season. Theyve been one of the best teams in the NHL. Theyve also been very consistent in their effort level, offensive levels, and defensive levels...and of course Henrik getting his head out of his ass is a huge reason for that as well.

 

[SnowblindNYR]

never said that either.

Their issues are clearly they have scrambly moments against teams with movements.

simply switching to a zone coverage, but aggressive zone coverage, would resolve that issue.

i dont like the pack it in Torts D. I prefer to attack the points and force defenseman to make good passes under duress in ALL zones.

playing man to man makes movement tough to handle.

Ok, but if you exclude the 9-2 loss that rational people realize was a team adapting, 3 of the best puck possession teams and 3 of the best offenses in the league in the Sharks, Hawks, and Bruins, in 6 games we allowed 1, 1, 2, 2, 3, 6 goals. That's 15 goals in 6 games. 2.5 GAA, not terrible at all against the best in the league. That 6 goal game is more of an exception than the norm too.

 

[Inferno]

Ok, but if you exclude the 9-2 loss that rational people realize was a team adapting, 3 of the best puck possession teams and 3 of the best offenses in the league in the Sharks, Hawks, and Bruins, in 6 games we allowed 1, 1, 2, 2, 3, 6 goals. That's 15 goals in 6 games. 2.5 GAA, not terrible at all against the best in the league. That 6 goal game is more of exception than the norm too.

fair enough. Rangers have played well against the better teams. I just dont like what I see when we dont have the puck. Way too much open areas of the ice, way too many guys getting caught trying to catch someone that could easily be picked up on a switch.

im just not a fan of man to man defense...no more than i am a fan of the 6 goalie system.

frankly, a neutral zone trap with our ability to hold the puck would probably be the best way to defend..though I'd hate to watch that kinda boring ass hockey.

 

[SnowblindNYR]

Not to mention that saying things like "most teams did not have to deal with what the Rangers had to deal with" is nothing but an excuse to use creative statistical methods.

No...it's reality.

 

[SnowblindNYR]

fair enough. Rangers have played well against the better teams. I just dont like what I see when we dont have the puck. Way too much open areas of the ice, way too many guys getting caught trying to catch someone that could easily be picked up on a switch.

im just not a fan of man to man defense...no more than i am a fan of the 6 goalie system.

frankly, a neutral zone trap with our ability to hold the puck would probably be the best way to defend..though I'd hate to watch that kinda boring ass hockey.

I think we've been inconsistent in our zone. We were more consistent under Torts because when in doubt fall on the ice. But, I think it's been way more good than bad. The slot IS more open now, but when Lundqvist is on his game I'd take a slot chance over a screened point shot or one deflected off a D-man because we're collapsing around him so much.

 

[McRanger]

My problem isn't that you're excluding our stats, it's that you're leaving EVERY other team untouched.

That's simply not true.

At least from me, stats run from the middle of October (or the beginning of November or January first or the Olympics...) include the stats of all the other teams accumulated over those time frames. It is a direct comparison of numbers.

It's not "excluding" games. Its showing the teams stats over a set time frame of the season. The last 25% of the season. Or 75% of the season. Or 90%. Its called looking at trends. Or progress, in the Rangers case.

Excluding games would be wacking out a stretch in the middle of the season without context.

Frankly I find it comical that Torts entire comparative sample is solely the '11-'12 season (not his most recent season, but certainly his best) while at the same time anyone only looking at the most recent 90% of AVs tenure (putting aside the road trip, the new coach, the injuries etc) is somehow being intellectually dishonest.

 

[True Blue]

YesLet me ask you something, you're taking a serve in Tennis. You take a serve once a day for 81 days. The first 10 games your arm is injured and hurts like a mother****er. Your serve is in the 80s. Day 11, your arm feel like nothing happened, since then your serve is in the 90s-100s and once in a while in the 80s but not consistently for 10 days like before. Do you throw out the first 10 serves? Or are should those be included in the average? The latter is how ridiculous people that think the first 10 games should have counted sound.

I will throw this back at you. An potential investor comes to me and we say "Look at these awesome returns that we have generated in the last 3 months. We are outperforming the field"

"But your performance does not match your competitors when taken as a whole".

"You need to ignore that and only look at the last 3 months"

Let's play the real world game. Does such an investor rush to give me money to invest?

I'm not surprised all of the pessimists "magically" can't understand this concept.

I am surprised that the optimists cannot fathom the concept of living in the real world.

 

[Thirty One]

That's fine, my point was that ranking a tampered stat against untampered stats produces a meaningless number. It's a fallacy. Although I'm sure if it was -31- and not me who pointed it out, they'd all be singing his praises.

I did mention this:

http://rangersunlimited.com/2014/02/05/just-how-good-are-the-2013-14-rangers/

"Now, is it fair to simply remove the Rangers early seasons struggles? Probably not. They were adjusting to a new coach, but so were other teams who performed better. Still, it’s a sample of 47 games where the Rangers rank among the top teams in the league. Hardly negligible."

But does that mean we can learn nothing from the results 70+ most recent games played?

I understand applying context to stats, my problem is that you guys only do it to the Rangers to make them look better.

I find this ironic. During St. Louis' first 4-6 games you, Bleed Ranger Blue and True Blue were quick to draw conclusions about St. Louis and the Rangers as a team. But using the 70 most recent games is outrageous?

Your interest in the integrity of a sample seems to depend on the sample.

 

[True Blue]

No, because most teams didn't have to deal with that. There are a handful of teams that WOULD be true about, but I don't think anyone here wants to go through this painstaking analysis. Hell, the "experts" don't.

All teams deal with different things. The teams that were injury ravaged all year, do not get excuses. They do not extrapolate states of what only happened with they had a full line up.

Rangers had a coaching change. Other teams have injuries. BTW, I find it laughable that the same people who are out to make excuses for "what the Rangers had to deal with" are largely the same people who did not allow last year's team the excuse of having a 7 day training camp.

 

[True Blue]

Except it's a ******** reality. The first 10 games were clearly a team adjusting to a radically different system and doesn't represent this team in the slightest. It wasn't just normal seasonal fluctuation like the December slump probably was.

It is only a reality if you go to all other teams and exclude their worst 10 game stretches.

 

[True Blue]

I think it's foolish to simply ignore these facts. If you dont want to readjust the stats, that's totally fine. But I think you need to be able to look at this team and say...you know what, since they started to get this system...particularly in the 2nd half of the season. Theyve been one of the best teams in the NHL. Theyve also been very consistent in their effort level, offensive levels, and defensive levels...and of course Henrik getting his head out of his ass is a huge reason for that as well.

Fair, but we can take extrapolation to a degree here. Did they get better or did the competition let up? Was their second half easier than the first?

My only point is that when you start to go down the slippery points of extrapolating stats, that opens up the door to a whole set of other circumstances that tend to be left out of your analysis.

 

[aufheben]

I did mention this:

http://rangersunlimited.com/2014/02/05/just-how-good-are-the-2013-14-rangers/

"Now, is it fair to simply remove the Rangers early seasons struggles? Probably not. They were adjusting to a new coach, but so were other teams who performed better. Still, it’s a sample of 47 games where the Rangers rank among the top teams in the league. Hardly negligible."

But does that mean we can learn nothing from the results 70+ most recent games played?


I find this ironic. During St. Louis' first 4-6 games you, Bleed Ranger Blue and True Blue were quick to draw conclusions about St. Louis and the Rangers as a team. But using the 70 most recent games is outrageous?

Your interest in the integrity of a sample seems to depend on the sample.

I never doubted that MSL would find his groove, I was just frustrated with how he was playing, and concerned that we would missed the playoffs because of it.

My problem with the sample is that it's ranking a tampered team with 29 untouched teams. I don't have a problem with people saying we're a better team since October, I have a problem with stat skewing.

For example, the link you just posted shows EVERY teams record since October 28th. That's fine.

 

[Thirty One]

I never doubted that MSL would find his groove, I was just frustrating with how he was playing, and concerned that we would missed the playoffs because of it.

My problem with the sample is that it's ranking a tampered team with 29 untouched teams. I don't have a problem with people saying we're a better team since October, I have a problem with stat skewing.

My bad.

The possibility that St. Louis had fallen off a cliff permanently was something that was brought up by the "I hate everything, everything is terrible" crowd.

 

[GoAwayPanarin]

aufheben.

i bet if you ran the stats readjusted to removing the first 9 games of every team in the NHL, it would still be pretty much the same thing...our EVERYTHING improves.

Slow Friday so I actually took the time to run the numbers in this scenario (for just goals for: I used 10 games, since its a nice round number.)

1. Chicago 3.31 GPG
2. Boston/Anaheim 3.24
4. Colorado 3.07
5. Pittsburgh 3.01
6. Philadelphia 3.00
7. Dallas/St Louis 2.95
9. Ottawa 2.91
10. Tampa Bay/San Jose 2.87
12. New York 2.86


Rangers would sit 12th in GPG if this were to be done.

So yeah they wouldn't be 8th, but they wouldn't be at 16th either and the difference between them and a team in the top 10 would be entirely negligible (one goal.)

 

[Thirty One]

My problem with the sample is that it's ranking a tampered team with 29 untouched teams. I don't have a problem with people saying we're a better team since October, I have a problem with stat skewing.

For example, the link you just posted shows EVERY teams record since October 28th. That's fine.

OK. I don't see how it makes a difference in this case, though. For every game played one team wins and one team loses. It's not as if the rest of the NHL could uniformly struggle from October 5 - October 28.

Side note: I think my next statistical study will be how many time you edit each post. Jesus.

 

[aufheben]

OK. I don't see how it makes a difference in this case, though. For every game played one team wins and one team loses. It's not as if the rest of the NHL could uniformly struggle from October 5 - October 28.

Because at least it shows all teams equally.

That's simply not true.

Yes it was.

 

[McRanger]

Yes it was.

No. It's not. At all.

All the stats are from fixed starting times and cover all the teams stats over that period. It was not just deleting the Rangers first month and comparing those stats to the current numbers of rest of the league.

All the teams Points, GF, GA and games shift. That is how you track trends. That is how you can tell whether a team is improving. And, maybe, try to figure out how they will perform in the future. Which is really all that matters as we approach the playoffs.

For example, this is the Rangers breakdowns over some set periods:
-------------------------PPG------GFG-----GAG----Goal Differential
Entire Season---------1.173----2.691----2.370----0.321 (9th in NHL)
Since November------1.227----2.879----2.273----0.606 (5th in NHL)
Since January--------1.324----3.081----2.108----0.973 (2nd in NHL)
Since deadline-------1.368----2.947----1.842-----1.105 (2nd in NHL)
and for comparisons:
'11-'12 Season-------1.329-----2.756------2.280-----0.476 (6th in NHL)

These are the stats compared to all the other NHL teams. This is a direct comparison of the Rangers numbers versus the rest of the league numbers over that same time period. Points per game. Goals for per game. Goals against per game. And goal differential. In another thread I broke it down this way and by individual month. No team is untouched. At all. In any way.

The November one includes the Rangers awful December run. The final two include the end-of-the-world 3 game losing streak in March. And the entire season includes all of the above, and of course the Rangers horrific first few games.

The purpose of these stats is to see try and predict how a team is trending as to see how they might perform in the future. And in that way people can interpret these stats however suits them.