Official 2015 tank thread IV

[Zip15]

Don't be too excited. We have lost a lot of games lately and been on a bad streak and we are still SIX points up on Edmonton....our only hope is that we at least finish 29th and either we win the lottery lol:) or the Oilers retain #1. I really despise Edmonton getting any of McEichel in any scenario but if it means we get Eichel I'll be fine with them getting McDavid undeservingly. Crazy That the big win streak where Enroth completely overachieved is gonna **** us for the next decade if we finish 4 or worse (or is it "better?Anyway picking 4-30)

At the same time, one could say that Edmonton has only won 2 of their last 24 games and yet we are only six points up on them. This is still very much a ballgame.


MOD EDIT: Link to previous thread http://hfboards.mandatory.com/showthread.php?t=1793983

Click to read more...

 

[UnleashRasmus]

Someone said it was Draisaitl's decision not to play in world juniors, not sure if that was accurate but wouldn't surprise with the bad german team.

And the oilers are still bad, but at least they look better than 10 days ago, when everything seemed hopeless. Maybe they can build some momentum from this coaching change, or it's only a short effect, who knows.

Okay, well if that's accurate, then I just blew a bunch of hot air. I appreciate that note if it is correct.

 

[haseoke39]

Edmonton's on Pace for 52 points, which is what we did last year. Totally sustainable record overall, IMO.

 

[dirk41]

Poker should help people when it comes to variance:

Poker is a game of skill and luck. Look at the wild swing between expected results and actual results. This happens in hockey, even if it's to a smaller degree.

The idea that things even-out over a season is just a variant of the Gambler's fallacy. A team could have worse luck than average (run below expectation) for an entire season. This happens in baseball, too, with pitchers over an entire season. Some have really great BABIP others have really terrible BABIP.

 

[dirk41]

Edmonton's on Pace for 52 points, which is what we did last year. Totally sustainable record overall, IMO.

It was sustainable for Buffalo because Buffalo was dead last in corsi-for close and Fenwick For percentage close. Edmonton is 20th in each. Edmonton is bad, just not as bad as their record.

 

[Husko]

I'm going to start reporting every single post that talks about variance and probability, my God move on already

 

[Namejs]

Poker should help people when it comes to variance:

Poker is a game of skill and luck. Look at the wild swing between expected results and actual results. This happens in hockey, even if it's to a smaller degree.

The idea that things even-out over a season is just a variant of the Gambler's fallacy. A team could have worse luck than average (run below expectation) for an entire season. This happens in baseball, too, with pitchers over an entire season. Some have really great BABIP others have really terrible BABIP.

As a former semi-pro poker player, I know exactly what you're talking about.

While everything you said is true, variance decreases with increased sample size.

Yes, there's no reason to assume that luck will necessarily even out over the course of the season. No one knows that. It's unpredictable.

But if we are making a future projection, we have to establish the fact that variance exists and that all of the indicators (CHIP, PDO, OT/SO losses, 1-goal-games) point to the fact that the Oilers have underperformed/have been unlucky and therefore we have to assume that they are likely (which doesn't mean that it is going to happen, it just indicates probability) to produce points at a higher rate than before.

It's quite simple, really.

According to sportsclubstats.com, there's currently a 42% chance that the Oilers won't finish 30th, which is substantially higher than the 0% suggested by MayDay.

 

[Namejs]

It was sustainable for Buffalo because Buffalo was dead last in corsi-for close and Fenwick For percentage close. Edmonton is 20th in each. Edmonton is bad, just not as bad as their record.

A voice of reason at last.

 

[TehDoak]

the problem with all this variance and luck talk is that the sample size of a regular season is relatively small. Teams can defy odds (see Toronto a few years back, Colorado last year, when Carolina won the cup in 06) over long periods. If the game was broke down to advanced stats was the end all be all, it'd be dumb and un-entertaining.

We have 43 games left. What will happen? Nobody knows. You can quote variance, luck, etc as a general guide to why some teams are doing certain things, but overall, the difference between picking 2 and picking 5th might only be 4-5 pts. Thats a few shootout wins, thats a hot goalie, thats so many things that the Sabres can really control.

Buffalo should play to move the franchise forward, and let the chips fall where they fall.

Sure, we'll move out a few pieces at the deadline and sure, we aren't going to actively look to improve for this year, but there is still a chance we'll finish last overall and a chance we'll finish outside the top 5. And there's a chance we win the lotto, regardless of where we finish. Getting into these little petty arguments about things that frankly, no one can control, is counter productive.

As always, remember your ABC's and 123's

Anybody
But
Carolina

1...2...3 1st overall picks is enough already Edmonton, STOP IT.

 

[Namejs]

the problem with all this variance and luck talk is that the sample size of a regular season is relatively small. Teams can defy odds (see Toronto a few years back, Colorado last year, when Carolina won the cup in 06) over long periods. If the game was broke down to advanced stats was the end all be all, it'd be dumb and un-entertaining.

We have 43 games left. What will happen? Nobody knows. You can quote variance, luck, etc as a general guide to why some teams are doing certain things, but overall, the difference between picking 2 and picking 5th might only be 4-5 pts. Thats a few shootout wins, thats a hot goalie, thats so many things that the Sabres can really control.

Buffalo should play to move the franchise forward, and let the chips fall where they fall.

Sure, we'll move out a few pieces at the deadline and sure, we aren't going to actively look to improve for this year, but there is still a chance we'll finish last overall and a chance we'll finish outside the top 5. And there's a chance we win the lotto, regardless of where we finish. Getting into these little petty arguments about things that frankly, no one can control, is counter productive.

As always, remember your ABC's and 123's

Anybody
But
Carolina

1...2...3 1st overall picks is enough already Edmonton, STOP IT.

Well, you're exactly right. But while obviously no one knows what will happen, I think it's pretty damn important to understand that the tank is still alive and that we can finish 30th. Also, Murray's decisions matter. It's definitely beyond our control though, unless he's reading this.

As for finishing outside of the top5, the odds of that happening are currently <2%.

 

[Zip15]

Of the bottom five, we've spent the most home games so far (22). Carolina has only played 17, Edmonton has played 20, Arizona has played 18, and New Jersey has had 17 home games. That's useful. Excluding Edmonton, those other teams are around .500 clubs at home, so they should be making up some points on us as they catch up in home games, especially considering how bad we are on the road.

 

[Namejs]

By the way, if anyone's interested, I tweaked the 'State of the Sabres Tank' spreadsheet a bit and now it auto-updates itself, meaning that anyone can see our chances of drafting McDavid, Eichel or Hanifin in real-time. Yay!

Here it is: https://docs.google.com/spreadsheets/d/17hn4UYpEQ8YNg7nnXu57yiNL21ezJMKehag6FfX--fg/edit?usp=sharing

 

[Milos*]

As a former semi-pro poker player, I know exactly what you're talking about.

While everything you said is true, variance decreases with increased sample size.

Yes, there's no reason to assume that luck will necessarily even out over the course of the season. No one knows that. It's unpredictable.

But if we are making a future projection, we have to establish the fact that variance exists and that all of the indicators (CHIP, PDO, OT/SO losses, 1-goal-games) point to the fact that the Oilers have underperformed/have been unlucky and therefore we have to assume that they are likely (which doesn't mean that it is going to happen, it just indicates probability) to produce points at a higher rate than before.

It's quite simple, really.

According to sportsclubstats.com, there's currently a 42% chance that the Oilers won't finish 30th, which is substantially higher than the 0% suggested by MayDay.

It's too bad with all that said, you still follow that fallacious way of thinking.. (You assume if they've been bad for so long, they're bound to have a good stretch). And that's not the case.

 

[Heraldic]

Have many of you actually watched Oilers this season?

I have watched majority of the consended games, and I think that Edmonton is a team which game has no connection to it's corsi or fenwick numbers. I mean, they make almost as many individual bonehead plays in a single game as our tank commanders in our whole season. And I'm not joking here. And their d-zone coverage is a lot worse than Rochester's. And I'm not kidding here either.

I just think that these guys who think that Edmonton will start to rise because of their corsi and fenwick numbers are going to be disappointed.

 

[Namejs]

It's too bad with all that said, you still follow that fallacious way of thinking.. (You assume if they've been bad for so long, they're bound to have a good stretch). And that's not the case.

Nice straw man. I've never said anything like that.

 

[Milos*]

Nice straw man. I've never said anything like that.

"we have to assume that they are likely to produce points at a higher rate than before.". You seem to be lacking very basic logic for you keep on repeating the same fallacy that's been pointed out time and time again... Meanwhile claiming the opposite! Lots of contradictions.

 

[misterchainsaw]

It's too bad with all that said, you still follow that fallacious way of thinking.. (You assume if they've been bad for so long, they're bound to have a good stretch). And that's not the case.

That's not true. If he thinks they've underperformed relative to their actual ability(say he thinks they should produce points at a .8 pts/game clip), then they should be expected to produce at a .8 pts/game clip for the 2nd half of the season - which is markedly better than the .64 pts/game pace they performed at for the first half of the season. And many of the advanced metrics do say that the Oilers have been unlucky.

If one believes that Buffalo and Edmonton are similarly bad teams, however, then there's no reason to believe that the current 6 point gap between the two teams will narrow. Certainly it'd still be possible, but its just as likely that the gap will widen as well.

 

[Milos*]

That's not true. If he thinks they've underperformed relative to their actual ability(say he thinks they should produce points at a .8 pts/game clip), then they should be expected to produce at a .8 pts/game clip for the 2nd half of the season - which is markedly better than the .64 pts/game pace they performed at for the first half of the season. And many of the advanced metrics do say that the Oilers have been unlucky.

If one believes that Buffalo and Edmonton are similarly bad teams, however, then there's no reason to believe that the current 6 point gap between the two teams will narrow. Certainly it'd still be possible, but its just as likely that the gap will widen as well.

Or if you believe they are similarly bad teams, then one would believe that the 6 point margin would stay around the same, rather than the Oilers finishing the season at around 12 points below the Sabres.

 

[misterchainsaw]

Or if you believe they are similarly bad teams, then one would believe that the 6 point margin would stay around the same, rather than the Oilers finishing the season at around 12 points below the Sabres.

Yes. That's what my 2nd paragraph says. There's no reason to believe that the gap should systematically widen or narrow. That also doesn't mean it can't happen.

Either way, your issue with namejs's point shouldn't be the probability discussion - he is absolutely correct that IF Edmonton was unlucky to post such a dismal 1st half record, they should be expected to improve over the 2nd half of the season. The debate really is about whether Edmonton IS that bad or not (or if Buffalo themselves overachieved during the 1st half, which is where my feelings lay).

Also coming into play is a phenomenon the Sabres have seen quite a bit over the last few weeks. Edmonton is likely to see a lot of backup goaltenders over the 2nd half of the season.

 

[Bps21*]

The difference between the teams is shootout wins.

They lost like 23 of the last 25. We won what was it 9 out if 11 or something, and the difference...is shootout points.

Neither their losing rate or our winning streak is likely to be repeated.

 

[Namejs]

"we have to assume that they are likely to produce points at a higher rate than before.". You seem to be lacking very basic logic for you keep on repeating the same fallacy that's been pointed out time and time again... Meanwhile claiming the opposite! Lots of contradictions.

What fallacy? Saying that they're likely to get more wins doesn't mean that they're bound to have a good stretch.

You're way, way out of your element.

 

[misterchainsaw]

The difference between the teams is shootout wins.

They lost like 23 of the last 25. We won what was it 9 out if 11 or something, and the difference...is shootout points.

Neither their losing rate or our winning streak is likely to be repeated.

The difference isn't entirely shootout points. Even if you knock Buffalo's SO record to 3-4 and Edmonton's up to 2-2, we'd have 2 more points. And it'd be more than that if EDM didn't get points for losing in overtime which they've done six times already.

And while certainly the type of ratio you point out is unlikely to continue, evidence does exist that shootouts aren't really a 50-50 proposition. The records are too diverse. Some variability is expected, but you really shouldn't have this many teams with 6-1, 1-6, 1-7 records in the shootout if it were truly a coin flip.

 

[Bps21*]

I didn't say it was a coin flip. I said it's ****ing stupid that being good at a skills competition makes what should be a gap that could be made up by this time tomorrow into months of grinding. The two point system has always been asinine. It artificially props up a worse team when they succeed at something that doesn't have anything to do with being a good team.

 

[Milos*]

What fallacy? Saying that they're likely to get more wins doesn't mean that they're bound to have a good stretch.

You're way, way out of your element.

And you don't have an element, for if you did, you would not be trying to predict how a team performs in the second half of the season utilizing their first half of the season as a reference point. That is fallacious and downright stupid.

1. Oilers Schedule had the easiest stretches early on in the season.
2. Countless times has a team started out great only to collapse later, or started out poor only to have a strong end to the season. Many times a team will have a linear progression. You're trying to tell people that this 'luck' thing will lead them to have a better record, and assuming their play won't change. What if they perform better? Worse? How many teams perform the same at game 38 as they do in game 82? There's no reason to believe things are going to change for the better just as there is no reason to believe things are going to change for the worse, but you keep pretending there's reason.

Your argument is just really stupid and requires too many assumptions, and the fact is, no one can know how the season will go, and you're not being sensible by trying to suggest otherwise.

 

[Zip15]

In case anyone didn't see, Roy drew the penalty that led to their PP tally that tied the game at 1 last night. He is our secret agent.

Have many of you actually watched Oilers this season?

I have watched majority of the consended games, and I think that Edmonton is a team which game has no connection to it's corsi or fenwick numbers. I mean, they make almost as many individual bonehead plays in a single game as our tank commanders in our whole season. And I'm not joking here. And their d-zone coverage is a lot worse than Rochester's. And I'm not kidding here either.

I just think that these guys who think that Edmonton will start to rise because of their corsi and fenwick numbers are going to be disappointed.

I get that you're dusting off the "watch the games, stat nerds" defense, but I have watched as many Oilers games as anyone in here--of that I am quite confident (thanks, pregnant wife who's asleep by 9:30 every night!)--and I reach a different conclusion than you. While they have their share of bonehead plays, I see little to no discernible difference in the amount of bad plays they make to other bottom-10 teams.

The biggest difference is that, until the past week or so, they couldn't get even average goaltending to save their lives. Their goaltenders weren't keeping them in games, and, on most nights, were playing them out of games. It's no coincidence that they haven't been defeated in regulation over the last three games when Scrivens gives them a solid night of work.

It's an unpopular opinion around here, but I think we are worse than the Oilers. I believe that if we played in the WC and the Oilers in the EC--against which they have a very respectable 6-4-1 record--we'd be trailing them by (at least) six points. Heck, we have a worse goal differential in a far easier conference.

Let's count 'em up after 82 and decide who was the worse team after this season.