Delta SOT - Yet Another Advanced Stat To Chew On (For Forwards)
by Steve Burtch on Jul 29, 2010 9:30 AM EDT
in Maple Leafs Analysis
Email
Print
Tom Awad at Puck Prospectus has been making a habit out of adding interesting statistics to the tool kit with which NHL fans can assess their respective clubs. GVT is a useful counting stat that gives us an idea of a player's value and contribution to the team in terms of goals (which equate to wins in the NHL). Unfortunately, not every player produces points, so GVT is sort of slanted against more defensive players.
Plus/Minus seems relevant to a lot of people but stat geeks are moving away from it because of it's inability to account for individual contributions. It has more value over time but it's hard to see how much sense it makes in the smaller sample size of an individual season.
Corsi numbers are being used more and more to assess puck possession but they don't do a wonderful job of accounting for the quality of scoring chances. Again, we might want to use something slightly different.
Another interesting statistic that Awad has recently sussed out is something he's terming Delta. Basically a statistical combination of the ideas of Corsi numbers and +/- Delta weights every shot on the ice (the Corsi portion) on the basis of the expectation of whether or not said shot will result in a goal (the +/- portion). The shot-quality is determined using models around shot distance and location on the ice as well as the game score. I won't get into a much more detailed explanation but if one is desired it can be found here.
The whole purpose behind the exercise of determining this new statistic is basically adjusting the concepts around puck possession in terms of ability to generate and prevent good scoring chances. I am particularly interested in adjusted Delta Values, which control for the various players on the ice with a given player, as well as the game situation in terms of current score. In the end, we get a fairly solid determination of the quality of a given player.
So let's look at how the Leafs Forwards compare using this new stat shall we?
First let's examine the forwards on next year's Leafs using their combined Delta SOT values in 5-on-5 situations for the past 3 consecutive seasons (Tom Awad has tabulated the Delta values and made adjustments for every NHL player from the past 3 years). The first column (Delta) is the raw Delta score for the player at 5-on-5. A value above zero indicates that the number of expected goals with the player on the ice was in his team's favour rather than for the opposition. The Delta QS (quality of situation), Delta QO (quality of opposition), and Delta QT (quality of teammates) are all adjustment factors used to determine Delta SOT. If the Delta QS, QO, or QT value is positive then that condition was working AGAINST the player, so a +QS means a player had more defensive than offensive zone faceoffs, a +QO means the player faced more difficult opposition than the average player, and a +QT means lower quality teammates on the ice hampered the generation of scoring chances. All of these conditions will depress the raw Delta score observed for a given player, so the Delta SOT is adjusted upwards to compensate.
Negative adjustment values result if a player has more offensive than defensive zone faceoffs, if the player faced easier or less difficult competition than average, or if they played with teammates that are above average. All of these conditions should work to elevate the raw Delta score for an individual, so the Delta SOT is adjusted downwards.
A Delta SOT score of zero would be a perfectly neutral player. They give up and produce chances of identical quality and number. Obviously virtually no players will accomplish this feat, so any player on the positive side of the docket produces more good scoring chances than they allow, while players on the negative side allow more good scoring chances than they produce.
| Player | Delta | Delta QS | Delta QO | Delta QT | Delta SOT |
| Kessel 09-10 | 5.6 | -0.6 | 0 | -3.9 | 1.1 |
| Kessel 08-09 | -1.1 | -0.1 | 1.1 | 0.6 | 0.4 |
| Kessel 07-08 | -0.6 | 0.9 | 0.6 | 1.9 | 2.7 |
| Bozak 09-10 | 0.8 | -0.9 | 0.2 | -1.2 | -1.1 |
| Grabovski 09-10 | 13.4 | -0.4 | -0.3 | -3.7 | 9.1 |
| Grabovski 08-09 | -3.2 | 0.5 | -0.2 | 1.9 | -1.1 |
| Grabovski 07-08 | -1.3 | 0.3 | 0 | 0.8 | -0.3 |
| Mitchell 09-10 | -1.9 | -0.2 | -0.9 | -1.6 | -4.5 |
| Mitchell 08-09 | -11 | -1 | -1.6 | 2.7 | -11 |
| Hanson 09-10 | -2.1 | -0.5 | -0.5 | -0.3 | -3.3 |
| Hanson 08-09 | 0.2 | -0.1 | -0.1 | 0 | 0.1 |
| Orr 09-10 | -3.6 | -0.9 | -0.9 | 0.2 | -5.2 |
| Orr 08-09 | -1.2 | 0.3 | -0.9 | 0.1 | -1.6 |
| Orr 07-08 | -1.5 | -0.2 | -0.7 | -0.4 | -2.8 |
| Sjostrom 09-10 | -2.9 | 0.2 | -1.3 | 0.6 | -3.6 |
| Sjostrom 08-09 | 0.9 | 0.1 | -0.6 | 0.2 | 0.6 |
| Sjostrom 07-08 | -0.3 | 0.4 | 0.7 | 1.1 | 1.8 |
| Armstrong 09-10 | -7.2 | -0.5 | -0.4 | 5.7 | -2.4 |
| Armstrong 08-09 | -3.3 | 0.9 | -0.8 | 4.5 | 1.3 |
| Armstrong 07-08 | -3.9 | 0 | 0.3 | 1.8 | -1.7 |
| Versteeg 09-10 | 11.2 | -0.8 | -0.3 | -6.7 | 3.4 |
| Versteeg 08-09 | 0.5 | -0.8 | 0.7 | -2.1 | -1.6 |
| Versteeg 07-08 | -1 | 0.1 | 0.2 | 0.3 | -0.4 |
| Kulemin 09-10 | 4.4 | -0.5 | -0.6 | -2.6 | 0.6 |
| Kulemin 08-09 | -3.1 | 0 | 0 | 1.6 | -1.5 |
| Brown 09-10 | -3 | -0.5 | -1.5 | 3.3 | -1.7 |
| Brown 08-09 | 0.7 | -0.1 | -0.3 | 0 | 0.3 |
| Brown 07-08 | -0.6 | 0 | 0.1 | 0.2 | -0.3 |
| Caputi 09-10 | -0.5 | -0.2 | 0.1 | -0.6 | -1.2 |
| Caputi 08-09 | 0.2 | 0 | 0 | 0.1 | 0.2 |
Generally speaking, the Leafs forwards next season have limited experience facing up against top competition. They also seem to generally get out-chanced when on the ice. The exceptions to that so far are Mikhail Grabovksi, Kris Versteeg, Nikolai Kulemin, Phil Kessel, and to a very limited extent, Tyler Bozak and Fredrik Sjostrom. Colby Armstrong has exhibited an ability to be productive in the past, but he has been lining up with some horrible teammates, against some difficult competition down in Atlanta, so it hasn't really come out in the wash.
Luca Caputi, Christian Hanson, and Mike Brown appear to be passable NHL players by this measure, and should be respectable from the standpoint of holding a lead.
At the horrible end of the scale are John Mitchell (thus further indicating a selection of very poor shot choices and generally poor defensive hockey sense) and Colton Orr. The Leafs enforcer brings something else that is valued by management to the table on a regular basis, so he gets a bit of a pass, and he doesn't play a lot so it's not a big deal.
Mitchell on the other hand is an exhibit in negligence. He is a generally negative player when on the ice, and based on these numbers, he is consistently out chanced while on the ice. This likely stems from a combination of poor defensive coverage, and his general tendency to stick handle the puck away from anything resembling a dangerous situation from the opposition's perspective.
I will be posting similar data for the Leafs defensive group tomorrow. Feel free to contribute comments below.
264 comments
|
0 recs |
Do you like this story?
Comments
Yet another thing to add to my list of “Reasons to throw John Fucking Mitchell off a goddamn cliff”.
Never liked that big ugly baby.
Fine with him not playing above the 3rd line, though.
"We can't solve problems by using the same kind of thinking we used when we created them."
Albert Einstein
by Say *plan the parade one more time*... on Jul 29, 2010 4:14 PM EDT up reply actions
I won’t get into a much more detailed explanation but if one is desired it can be found here.
Link doesn’t work.
Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.
Fixed
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
That’s a whole lotta analysis to tell me what I already knew. Mitchell sucks.
"I'd walk into the Leafs dressing room to get ready for the day and Harold would be there in his boxer shorts shaving. King Clancy would drop by a little later, play the fool, and then head off to the racetrack." John Brophy
by Mike Pelyk's Hairdo on Jul 29, 2010 9:47 AM EDT reply actions
So to be clear, per this model, Mikhail Grabovski drives offense about nine times better than Phil Kessel?
Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.
Yeah, I’m not a very big Grabovski fan, but that number really stands out. Are there some crazy lucky-linemate shooting percentages feeding that, or what is it? Not enough time spent playing in front of Toskala?
I've been looking at the sky
by Back In Black on Jul 29, 2010 10:12 AM EDT up reply actions
Proof that Grabbs is the best offensive player EVAR IN THE HISTORY OF THE LEAFS AND MAY NEEEEEVER BE TRADED EVAAAAAAAARR!!!1
Got it?
Oh yeah, and no moving him to the fucking wing, either.
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 6:06 PM EDT up reply actions
not a bad idear
Grabovsky on the wing eh? He plays along the boards almost exclusively already. He can’t win a faceoff. This has some traction…
by Johnny Bower's Pokecheck on Jul 29, 2010 6:10 PM EDT up reply actions
Have you been under a rock for the past month, JBP? We’ve (I’ve) argued this ad nauseum. Makes no sense unless you have all of: a centre that is better, no chance of ruining his confidence, and the coaching staff believe it’s the right move.
Until then, terrible idea.
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 7:03 PM EDT up reply actions
Eventually it might make sense. Right now though it won’t happen.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
shooting percentages
are irrelevant to this statistic.
It is a result of SCORING CHANCES… shots from ideal locations on the ice where the average NHL shooter is more likely to score.
Actual shooting percentages don’t come into play.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:53 AM EDT up reply actions
I’m confused by this, then:
Delta weights every shot on the ice (the Corsi portion) on the basis of the expectation of whether or not said shot will result in a goal (the +/- portion)
If you’re using +/- to determine whether or not a shot shoudl result in a goal, then shooting percentage necessarily comes into play. +/- is based on shots that go in, not quality of shots. Over the entire league, it seems perfectly appropriate to evaluate the “expectation of whether or not said shot will result in a goal” this way. For an individual player it seems like luck in shooting percentages and save percentages could affect the results, at least in extreme cases. But I haven’t waded through Awad’s description to figure this out.
Maybe Grabovski was just sneaky-good this year.
I've been looking at the sky
by Back In Black on Jul 30, 2010 9:33 AM EDT up reply actions
No
you’re misunderstanding… when I say “the +/- portion”, I’m making an analogy to the actual +/- stat without invoking the actual numbers.
Goals are not accounted for… it’s the quality of the scoring chances that add or subtract from a player’s score… hence the comparison to the NHL’s +/- statistic. The “corsi aspect” comes from counting ALL shots for and against, whether or not they hit the net.
Shooting percentages and luck and goals and all that jazz you’re discussing have absolutely ZERO impact on this statistic.
I’ll state it more explicitly… EVERY SHOT ATTEMPT is registered in the statistic, and a positive score is assigned to every player on the ice for shots taken by he or his teammates from positions based on how LIKELY the player is to score from that location. If the line on the ice surrenders a shot to the opposition then a negative is assigned based on how LIKELY the opposition is to score from that location on the ice.
The result (goal, miss, block, shot) is not accounted for or registered. Therefore, none of the aforementioned stats affect it.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Aug 1, 2010 12:52 AM EDT up reply actions
Oh and
to clear up likelihood… that is determined based on shot location, and total percentages based on all NHL shooter data recorded for those locations. Thus the percentage for a given location on the ice depend on the AVERAGE result from said position.
Typically shots from right in front are HIGHER percentage shots, while those from the point or outside the zone would be LOW percentage shots. An individual’s shooting percentage isn’t taken into account.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Aug 1, 2010 12:53 AM EDT up reply actions
Thank you
That makes perfect sense. You can maybe understand why I got confused, though, that CORSI was being used but +/- was an analogy.
I've been looking at the sky
by Back In Black on Aug 1, 2010 9:20 AM EDT up reply actions
So, what you're saying is....
Grabovski puts himself into position to score. Whether or not he has the talent to produce goals is a different matter.
but no pop
so did Stempniak
Didn’t work in T.O, but somehow worked in Phoenix (for a while)
"We can't solve problems by using the same kind of thinking we used when we created them."
Albert Einstein
by Say *plan the parade one more time*... on Jul 31, 2010 9:14 AM EDT up reply actions
Yes
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
Exactly
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Aug 1, 2010 12:53 AM EDT up reply actions
And not just himself
but his linemates also.
He ALSO has a positive impact on his LINE preventing shots from scoring positions on the ice. That would also correlate to his Corsi numbers as he does a good job of maintaining control of the puck while on the ice, and pursuing it when it is not in his possession.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Aug 1, 2010 12:54 AM EDT up reply actions
my guess is that since the data for him was compiled over fewer games, it’s not as accurate.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 11:49 AM EDT up reply actions
So Fredrik Sjostrom in 07-08 is roughly three times better at driving offense than 09-10 Phil Kessel?
Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.
Sjostrom also played less games in t hat year than Kessel did this past year….
I dunno, just trying to guess as to why the numbers look wrong.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 1:32 PM EDT up reply actions
or we misunderstand the purpose of this stat
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:23 PM EDT up reply actions
is this stat slanted because kessel plays against stronger opponents?
"We can't solve problems by using the same kind of thinking we used when we created them."
Albert Einstein
by Say *plan the parade one more time*... on Jul 29, 2010 4:15 PM EDT up reply actions
Sjostrom played
5 fewer games than Kessel… and no I didn’t “calculate it wrong”.
I summed the totals from both NHL teams the players played for if traded.
It’s a counting stat, just like +/- is… jeesh.
I rarely add wrong… I might mess up 5vs5 against all situations, and I know I messed up zone start vs. zone finish with Schenn once before… but I try to minimize these things generally.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:56 AM EDT up reply actions
The secret is pineapples, people, finely sliced pineapples. No one knows how or why it works, but it does.
I am Mikhail Grabovski's smirking revenge.
by kidkawartha on Jul 29, 2010 11:52 AM EDT up reply actions
It's not multiplicative
any more than +/- is… so yeah… I won’t venture there… but feel free to go overboard.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:54 AM EDT up reply actions
Just looking at Kessel’s numbers makes me question the numbers a little. If true, Kessel’s teammates last year in Toronto, when the Leafs were the second worst team in the league, were significantly better than his teammates in Boston the year before, when the Bruins was the second best team in the league. Yes, Boston almost certainly overachieved based on outstanding goaltending and the Leafs certainly underachieved based on horrific Toskala goaltending, but even so, this is something that I cannot accept as true, or even close to true.
----------------------------------------------------------------------------------
HockeyAnalysis.com - Taking a Deeper Look at the World of Hockey
As much as I love advanced stats and lean towards beliving objective numbers over subjective evaluations, I have a lot of trouble with any stat that says that Mikhail Grabovski was the Leafs best forward by far.
Also, I thought the book on Versteeg is that his linemates in Chicago were generally poor — but these numbers seem to suggest the opposite.
Down Goes Brown - Unapologetically nostalgic for the past. Brutally realistic about the present. Grudgingly optimistic about the future.
by Down Goes Brown on Jul 29, 2010 10:20 AM EDT reply actions
Yeah, I had the same two thoughts. I don’t see how Grabbo is far and away our best forward, and Versteeg had far better teammates in 09-10 than 08-09.
This seems to me a highly context sensitive stat like corsi that is a useful tool but is not as good at loosely “ranking” players as something like GVT (while admitting GVT has its blind spots).
by The '67 Sound on Jul 29, 2010 10:30 AM EDT up reply actions
I actually don’t like the advanced stats that much although they have some limited use. Too often they are used as a crutch by people who don’t really understand the game that well or who don’t watch very much hockey.
There have been a number of articles/posts I’ve read in the last few months (not on this site exclusively) where a guy’s play has been statistically analysed to death and then the commentator concludes with something like, “thus making Setoguchi the second best left-winger on the Sharks”, proving the commentator doesn’t even actually watch the games or know the position the player plays.
"I'd walk into the Leafs dressing room to get ready for the day and Harold would be there in his boxer shorts shaving. King Clancy would drop by a little later, play the fool, and then head off to the racetrack." John Brophy
by Mike Pelyk's Hairdo on Jul 29, 2010 10:54 AM EDT up reply actions
Too often they are used as a crutch by people who don’t really understand the game that well or who don’t watch very much hockey.
I can’t remember who wrote it but a good rule of thumb I read was that if a stat is completely at odds with what you see with your eyes it probably still needs work. The best tend to confirm what you see.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
by PPP on Jul 29, 2010 11:29 AM EDT up reply actions 7 recs
Generally true. Stats, when used properly, can eliminate bias or help us to look past extreme events that stick out vividly in our minds (i.e. a small number of horrendous turnovers that may make a player look real bad) and instead look at the overall picture (i.e. a players overall effectiveness) but rarely will they tell us something completely different from what we see. So when Grabovski turns up as the best Leaf by a sizable margin or Kessel’s teammates in Boston were significantly worse than his team mates in Toronto, that likely means the statistical analysis is flawed and not that the analysis telling us something completely new.
Generally speaking, shot analysis such as this (or Corsi analysis) has done a very poor job explaining the Leafs performance the past season or two.
----------------------------------------------------------------------------------
HockeyAnalysis.com - Taking a Deeper Look at the World of Hockey
by HockeyAnalysis on Jul 29, 2010 11:39 AM EDT up reply actions
I usually rely on the “I am always right and everyone else is full of crap” analytical tool. Especially with a few drinks in my system.
When I explain how this works people often say, “What a tool!”.
"I'd walk into the Leafs dressing room to get ready for the day and Harold would be there in his boxer shorts shaving. King Clancy would drop by a little later, play the fool, and then head off to the racetrack." John Brophy
by Mike Pelyk's Hairdo on Jul 29, 2010 12:08 PM EDT up reply actions
Yeah, the corsi thing has not been working with the leafs at all. They have been practically a sprinkler system for pucks, firing them totally ineffectively from all over.
I think we see some insane range restrictions in typical NHL data.
It’s sort of like how GRE scores predict nothing about grad school… So you say GRE’s are useless, right? But the people with bad GRE scores are never admitted, so all you can conclude is that the people who score high on the GRE’s don’t do any better than the people who score VERY high.
Maybe it’s the same thing with shooting. It probably really does matter the quality of the shots you take, but all the NHL players are close enough to each other in skill that it LOOKS like shooter quality doesn’t matter… there’s just not enough variance to see what’s really going on.
Food for thought anyway.
On the other hand a guy like Kessel is better at driving offense than Mikhail Grabovski.
Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.
That’s not what the Delta says!
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 10:53 PM EDT up reply actions
I’m pretty sure that Delta counts both shots for and shots against, so even if Kessel is putting up solid offensive numbers, he may be giving opportunities from good locations back.
Only YOU can prevent idiots from commenting!
by Knee high to a duck on Aug 2, 2010 2:49 PM EDT up reply actions
probably
except you guys aren’t actually looking at the correlations with this stat… you’re just riffing off what you think of the data without ANY IDEA what the actual correlation is.
This data correlates more closely with +/- than corsi does… but I bet you generally speaking absorb corsi data and think there isn’t much to disagree with … so why is this meeting the opposition it is?
It’s a measure of scoring chances for vs. against… nothing more, nothing less… why is that hard to believe?
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:22 AM EDT up reply actions
Your comments
are completely at odds with the results we’re seeing.
Last year, the Leafs – according to their corsi – should have DRASTICALLY outscored their opposition and made the playoffs…. and yet they didn’t.
Also, take a look at Kessel’s corsi QoT in Boston… and it’s quite a bit lower than it was in Toronto. 1.892 in Boston in 2008-09 and 3.013 in 2007-08… then it was 7.409 in Toronto in 2009-10… yeah there’s no way in hell it was better in Toronto right? Just more than twice as good as it was two seasons previous, and almost four times as good as the immediate previous year… that can’t be an improvement can it?
OBVIOUSLY there’s no hope in hell that his line mates were superior.
Yeah obviously said statistical analysis is flawed… or else maybe our view point of Boston’s 3rd line in comparison to Toronto’s top line is flawed…. which is it do you think?….
p.s. yes I’m drunk… but wtf people.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:16 AM EDT up reply actions
I find it bizarre
that rather than look at the numbers more in depth… people immediately revert to the whole “gut reaction” thing… that really bothers me actually.
Especially when the majority on here are probably people that consider this “advanced” stuff more than most…
rather than discount it as flawed… why don’t you explore it more before assuming it’s “wrong”.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:19 AM EDT up reply actions
Ah drunk. Explains why I suddenly had a hard time deciphering some of your posts, and the slightly beligerant tone.
yep
both…
and my typing is worse… I used to be a 100 wpm typist while loaded… what happened?? ?
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:37 AM EDT up reply actions
Ah college
You may be taking Jared a little too seriously
by JaredFromLondon on Jul 30, 2010 3:12 AM EDT up reply actions
but not the rest of the NHL
so assuming it’s wrong, because it doesn’t explain the Leafs… doesn’t make a lot of sense.
Why not look bigger picture, rather than myopically exploring our small corner of the NHL?
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:20 AM EDT up reply actions
I’m gonna go ahead and rec this.
Stats are never a replacement for actually watching the game and figuring out what you’re seeing. They’re just a good complementary too.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 11:51 AM EDT up reply actions
Exactly – rec’ed
"We can't solve problems by using the same kind of thinking we used when we created them."
Albert Einstein
by Say *plan the parade one more time*... on Jul 29, 2010 4:17 PM EDT up reply actions
Heard of that too
90% of the time, if a statistic confirms what you know, you’re probably on the right track.
If it tells you everything you know is wrong, it’s probably not worth a damn.
I think it was Bill James who said that or something very close to it.
by Slava Duris #24 on Jul 29, 2010 10:13 PM EDT up reply actions
I like advanced stats when I dissect and understand them and agree with their assumptions. I won’t be writing about Delta SOT any time soon.
Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.
Yeah, or at least (as PPP pointed out) their results make sense.
GVT, by contrast, makes sense to me because it identified Ovechkin as the best player in the league of all time, and Toskala as the worse. It probably overvalues offensive players and undervalues defensive, but then so does the league for the most part.
by The '67 Sound on Jul 29, 2010 1:17 PM EDT up reply actions
of all time...?
better than Gretzky, Lemieux, Howe, Richard etc…?
I have nothing interesting to say.
by blurr1974 on Jul 29, 2010 1:19 PM EDT up reply actions
Sorry, my bad. Just for last year.
I was simultaneously thinking of GVT’s “all time” rankings and it unconsciously slipped into that sentence by mistake.
by The '67 Sound on Jul 29, 2010 1:26 PM EDT up reply actions
I don’t understand GVT very well, but I went through the all-time spreadsheet and saw a lot of rankings I agreed with, a few that seemed odd at first but made sense when I thought about them, and virtually none that made me say “that can’t possibly be right”.
Down Goes Brown - Unapologetically nostalgic for the past. Brutally realistic about the present. Grudgingly optimistic about the future.
by Down Goes Brown on Jul 29, 2010 1:26 PM EDT up reply actions
I should add that I felt this way only after cutting and pasting the words “eleven billionty” into all Wendel Clark’s columns.
Down Goes Brown - Unapologetically nostalgic for the past. Brutally realistic about the present. Grudgingly optimistic about the future.
by Down Goes Brown on Jul 29, 2010 1:27 PM EDT up reply actions 1 recs
Check out Wendel’s 92/93 playoffs though: 6.5GVT in only 21 games. That projects to a Crosby-esque 25 over a full season.
He was a one man wrecking crew that playoffs.
And Gilmour was a 10.9.
by The '67 Sound on Jul 29, 2010 1:32 PM EDT up reply actions
No one here should be surprised about this.
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 6:09 PM EDT up reply actions
that’s the point
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 6:13 PM EDT up reply actions
you’re pointy.
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 7:03 PM EDT up reply actions
Yeah, I did that same exercise, had almost exactly the same reaction, and it’s what really sold me on the stat.
by The '67 Sound on Jul 29, 2010 1:28 PM EDT up reply actions
And I agree, there are some tweaks to be made to GVT (and I bet Tom Awad would tell you all about them) but on the whole you look at it and say “hmmm ok”.
This makes me want to say “lol”.
Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.
Agree. I’m awful with numbers but GVT makes sense to me and seems like an accurate reflection of things I can observe on my own.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:08 PM EDT up reply actions
I wonder if Steve got it backwards because everything makes more sense if a negative number counts against the player.
Grabovski was the best Leafs forward at getting the puck down the ice last year. Since that’s a part of this I’m not surprised that he does well by this measure.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
Nope
I didn’t get it backwards… what I’m saying is just being read incorrectly.
The Delta QS (quality of situation), Delta QO (quality of opposition), and Delta QT (quality of teammates) are all adjustment factors used to determine Delta SOT. If the value is positive then that condition was working AGAINST the player, so +QS means more defensive than offensive zone faceoffs, +QO means the player faced more difficult opposition, and +QT means lower quality teammates on the ice.
Read that and think about what I’m saying. If you have a POSITIVE number in the Delta QT column, that means you are getting added points because that factor was working AGAINST you… i.e. you had crappy teammates, so your score is increased to compensate for the fact that they sucked. If you have a POSITIVE number in the Delta QO column, it means you played more difficult opposition, and you’re again rewarded for having to deal with playing against more difficult competition than the average NHL player.
Having a negative wouldn’t make MORE sense in either case. If I’m facing harder competition and playing with crappier players, why would I be punished further for those two factors than I already am in my Raw Delta Score?
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
In re-reading what YOU said
you’ve got it correct… I’m saying exactly what you are… I’m discussing the CONDITION counting against the player… not the score.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
Ah ok. That is clearer. Maybe some editing is in order.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
shall fix it.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
As much as I love advanced stats and lean towards beliving objective numbers over subjective evaluations, I have a lot of trouble with any stat that says that Mikhail Grabovski was the Leafs best forward by far.
The problem is that all of these advanced stats are subjective. The only objective data is the raw data, and even that is problematic in some cases.
All of these measures are trying to assess value based on criteria that are dependent on the person creating the measure. All of this is basically “intangibles” backed by math.
Leaf, the universe and everything.
Now in year 44 of the 42-year saga.
Excellent point.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 12:42 PM EDT up reply actions
Actually
the stat isn’t any more subjective than any other stat… the whole point is it’s more objective than just saying “hey I like that guy”. Your gut instincts are subjective. People like Mitchell because he works hard…. unfortunately he sucks.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:14 PM EDT up reply actions
Actually I think people like Mitchell because he’s cheap.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:16 PM EDT up reply actions
Yes that too
but that will be slightly different this season… which means he’s less valuable.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:16 PM EDT up reply actions
The main distinction on cheapness
people have made is that he produces a fair number of points for very little cost.
Unfortunately he is a horrible player from a defensive perspective, even when accounting for his weak linemates.
He lets up a lot more chances/goals than he produces… so whether or not he produces points he’s a danger when he’s out on the ice.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:17 PM EDT up reply actions
Oh I still hate him, I just kinda think there’s some defense for him. He’s pretty productive for a player in his salary range.
For instance, I would rather have John Mitchell at $750,000 than Ben Eager at $925,000.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:18 PM EDT up reply actions
Depends what you want from him though
you aren’t paying Mitchell that kind of money to score for you… despite the fact that people use his point production as justification for keeping him around.
You’re paying him that much to play on your bottom 2 lines, and he kinda sucks from a defensive perspective… that’s why I don’t think he’s of value. He sucks at preventing chances, and his +/- is historically atrocious.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:31 PM EDT up reply actions
SOCKS!
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:17 PM EDT up reply actions
Socks!
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 6:13 PM EDT up reply actions
late to the dance. :(
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 6:14 PM EDT up reply actions
I think that was 1967’ers point. Any advances stat – this or any other – has a measure of subjectivity dependent on the person creating the measure.
In any case, much better effort than anything I’ve ever done.
And Mitchell sucks. That’s why I like this.
BS
by MapleLeafMole on Jul 29, 2010 2:20 PM EDT up reply actions
Actually, that’s my point. All of these advanced stats are subjective. Your link supports this.
Leaf, the universe and everything.
Now in year 44 of the 42-year saga.
They’re not subjective, they’re objective. But that’s not the same as “foolproof”. It sounds like a bunch more work needs to be done on validating it as a measure (i.e., showing that it really does pick up what we think it does). If you were doing academic research on this you would spend a fair bit of time having it predict various things, and showing that it does a better job than other similar measures.
But being as this is just a couple of smart people fooling around with the numbers in their spare time, the “hits you between the eyes” test is probably the best one.
Once you manipulate and interpret data, it’s subjective. You’re weighting shots for as more important than, say, PIM, because you believe shots are a better indicator of wins. That last bit is a subjective element.
Just the raw data is objective,* because all is factual and all is equal.
*: Even stat recording has been shown to be subjective. Shots and hits, for example, vary by home rink bias.
Right, there are some subjective elements that go into recording them, and “crap in, crap out” always holds.
But you can get around the fact that choice of manipulations is subjective by trying to fit data against objective criterion data. For example, does your corsi measurement from year X better predict teams wins in year X+! than does your DSOT measure? Does it predict the pattern of same-season wins with closer accuracy?
There’s a bunch of fit stats that people who do modelling for a living use to evaluate the models they construct (and these different stats are basically models trying to concisely describe performance in ways that reveal the useful essence of what is actually happening). Without those tests, yeah, it’s just subjective :)
I'm also in 100% disagreement
with the ‘intangibles backed by math’ statement.
It’s a calculation based on shot distance, game score situation, faceoff location, and shots allowed vs. shots taken… I fail to see how that’s a set of “intangibles”.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:15 PM EDT up reply actions
A Sjostrom shortened season is more valuable than Kessel scoring 30 goals.
Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.
yes 65
vs 70 games…
how shortened.
Also… Sjostrom’s D and Kessel’s O may be quite balanced, and then it’s not really a matter of goals…
I find it baffling how much trouble people are having wrapping their head around this.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 3:04 AM EDT up reply actions
Versteeg's numbers
this past season for line-mates were a -6.7 shows that he played with far better linemates than you might think based on what you’ve read in the general media. Of course, that would be true of most Chicago players using this statistic.
It’s related to scoring chances when he and his team-mates were on the ice. If you compare him to the rest of the Blackhawks, He had the 12th best QoT on the team. Duncan Keith and Bryan Campbell had the best teammates. There were seven forwards ahead of him, including Kane, Toews, Brouwer, Ladd, Sharp, Madden, and Byfuglien.
Basically what I’m saying is… it’s all relative man. Just because the Blackhawks as a team generated more scoring chances, doesn’t mean Versteeg was playing with WAY better players than all the Leafs.
In comparison if you look at what rank Kessel had on the Leafs w.r.t. linemates, Kessel had the best linemates of any forward on the Leafs, while Kaberle had the best linemates of any of the D men.
It makes sense if you look at the entire NHL… which you guys aren’t.
I’d suggest people explore the entire 3 year data set before forming conclusions…
I’m actually quite shocked at how vehement the rejection sentiment has been to all of this…. in a weird way it makes me think the stat is more valuable.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 3:03 AM EDT up reply actions
Yep
but Luke really isn’t there to genereate offense… so he suffers from this stat.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:19 PM EDT up reply actions
Schenn
That’s not really a fair assessment though…Schenn doesn’t generate any offence whatsoever. He’s Drake Berehowsky with more speed. What do you expect?
by Montreal Export on Jul 29, 2010 3:17 PM EDT up reply actions
i dont think berehowsky ever hit anyone with half the enthusiasm schenn does
There is no "I" in team, but there is an "M" and an "E"
by Matt_Roberts on Jul 29, 2010 4:21 PM EDT up reply actions
not without breaking his leg, anyway.
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 6:26 PM EDT up reply actions
Thank goodness, we’ve been waiting for that. Grit Versus Threshold can only take us so far.
Join me on the Hockey Blog Adventure! (or Twitter.) GO BRUINS! (and Wild!)
by Cornelius Hardenbergh on Jul 29, 2010 11:05 AM EDT up reply actions 1 recs
I’m waiting for the truculencometer so we can accurately and fairly evaluate Brian Burke’s moves.
----------------------------------------------------------------------------------
HockeyAnalysis.com - Taking a Deeper Look at the World of Hockey
by HockeyAnalysis on Jul 29, 2010 11:41 AM EDT up reply actions
Puck Prospectus has Patrick O’Sullivan as fourth best DeltaSO in the league. That seems odd.
by Leaf in Habland on Jul 29, 2010 11:47 AM EDT via mobile reply actions
He's
shooting from good locations… but he has a horrible tendency to shoot into the goalie.
Shooting percentages are ignored in this, as are goals.
It’s a measure of CHANCE production vs. allowance… the fact that a lot of goals go in, or don’t go in has nothing to do with it.
It’s just as viable as corsi for measuring puck possession… that doesn’t factor goals in either.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:11 PM EDT up reply actions
I am really leery of these stats, they don’t tell the whole story, or anywhere close to the story. I actually wrote a blog about stats like these (a short one).
http://leafsbattlestation.blogspot.com/
Feedback would be great, also, I don’t know how to get rid of the latin in the “Featured Contents” part of the blog.
"There is no big lie, no system. The universe is indifferent."- Don Draper
Think you’re really off track with that post to be honest. The head in the sand approach isn’t any better than using one single stat to base their analysis upon.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
also
doesn’t Burke pay for advanced stat data for the Marlies?
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 1:44 PM EDT up reply actions
He does, but from what I see, (not just PPP but elsewhere) is that teams are gauged solely on stats, which is why the Armstrong signing was slammed and Burke was criticized for not signing Frolov. I’m not saying obscure stats are bad or am I trying to be rude to the people who use them, but this offseason is flooded with stats that don’t realistically factor into the season
"There is no big lie, no system. The universe is indifferent."- Don Draper
by Killer Sundin on Jul 29, 2010 1:52 PM EDT up reply actions
Oh I know what you’re saying about teams being gauged on stats, I am frustrated by this as well. But I don’t think it means that we should ignore the stats, I just think the discussion should be tempered with an admission that there is a very specific team building strategy happening in Toronto, and from that POV players like Colby Armstrong make sense.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 1:55 PM EDT up reply actions
I think stats discussion with these other factors taken into consideration is really interesting. Take Steve’s statement above:
Colby Armstrong has exhibited an ability to be productive in the past, but he has been lining up with some horrible teammates, against some difficult competition down in Atlanta, so it hasn’t really come out in the wash.
I think this is something we need to consider when evaluating the signing.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 1:58 PM EDT up reply actions
players like Colby Armstrong, just not salaries!
Sorry, couldn’t resist. Ignore me.
by The '67 Sound on Jul 29, 2010 2:09 PM EDT up reply actions
haha
if only there was a metric that helped determine the proper salary for a player based on stats! :P
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:14 PM EDT up reply actions
I HOPE a third liner is the key to next season but my guess is a 30 goal scorer would have been more helpful.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
One of these years our shiny new third liner is bound to be the difference, right? Right?
by The '67 Sound on Jul 29, 2010 5:12 PM EDT up reply actions
Yeah, I think baseball has pretty much proven than relying on “what you see” is a recipe for disaster that makes people think Scott Podsednik and David Eckstein are good baseball players.
No one says ignore the games themselves, but ignoring stats that can increase your knowledge and understanding of the game is—for a GM at least, fans should do what brings them the most enjoyment—foolhardy.
The most obvious example is the work in the statistical community proving that goaltending is a crapshoot and it’s foolhardy to pay significant dollars for all but the most elite of goalies. The NHL has finally caught up to that research this year. Too late for us to avoid trading for and re-upping Toskala a few years back at $4MM unfortunately.
by The '67 Sound on Jul 29, 2010 2:08 PM EDT up reply actions
Exactly, great stats can fundamentally change the way you watch a game and evaluate people.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:11 PM EDT up reply actions
To give an example, I undervalued Parise. I would have thought of him as a Top 30 F in the NHL, no more.
GVT had him 6th this year among forwards. If you think about it, it makes sense. Very good offensively, and one of the best defensive forwards in the game. A latter day Doug Gilmour if you will.
The best stats not just confirm most of what we already believe but also uncover things we didn’t.
by The '67 Sound on Jul 29, 2010 2:19 PM EDT up reply actions
Parise
is currently one of the top 5 forwards in the NHL as far as I can tell by most estimations.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:20 PM EDT up reply actions
Yeah, my point being I wasn’t aware of that from conventional stats and just watching him the few times the Leafs play him each year.
by The '67 Sound on Jul 29, 2010 2:21 PM EDT up reply actions
If your point is GVT underrates him… you’re probably right. 3 of the top 5 are Caps, and Backstrom/Semin probably get a boost from their teammates.
by The '67 Sound on Jul 29, 2010 2:22 PM EDT up reply actions
I think if you asked fans to write a top 5 list of Forwards in the NHL his name would not be on most of them.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:22 PM EDT up reply actions
That's because
fans underrate defensive performance.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:23 PM EDT up reply actions
he’s also in New Jersey, a team which people don’t know much about and often write off as being totally systems-driven.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:24 PM EDT up reply actions
Well that's the other thing
it IS systems driven… but he plays within that system excellently.
Travis Zajac is another great defensive player who plays excellently within that system.
Players that play successfully in NJ for a long time often come out being some of the best defensive players in the NHL… funny that… even if you put them in OTHER systems.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:25 PM EDT up reply actions
Sure, but while a lot of former Devils have their careers defined by the defensive abilities they learned in New Jersey, Parise would probably be incredibly good anywhere he played.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:26 PM EDT up reply actions
Yeah that's more my point
He’s awesome at both ends of the ice… sort of like Brian Gionta or (generally speaking) Scott Gomez… they’re both very solid defensively, and produce decently at the offensive end.
Parise is another level above that.
Other examples include Jason Arnott, Patrick Elias… etc.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:30 PM EDT up reply actions
Exactly. He wouldn’t have been on mine until I delved into this stuff.
by The '67 Sound on Jul 29, 2010 2:23 PM EDT up reply actions
As with physics, the more in depth you go, the more abstract the math becomes. I think the math behind things like GVT or VORP in baseball is pretty straightforward. But I’ve always felt that when you pick apart something to such a minute level, like the percentages of LWs in the NHL scoring from the right side in 4 v 5 situations, you’re losing too much in the process. I think up to a certain threshold, statistical analysis is great, and then it loses its value.
but no pop
You changed the outcome by measuring it!
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:41 PM EDT up reply actions
Advanced stats
make me feel sleepy time. just go watch the game kids.
by Johnny Bower's Pokecheck on Jul 29, 2010 2:02 PM EDT reply actions
Now, I wouldn’t go there. I think there’s a lot to be found within the numbers, I just haven’t seen a measure yet that I find particularly convincing. There are too many assumptions in the models and I can’t tell whether the more reasonable-seeming ones are actually telling me something or whether it’s simply confirmation bias.
I’m much more willing to put stock in advanced metrics in baseball than in hockey. Baseball is built much more around discernable, measurable events than hockey is.
Leaf, the universe and everything.
Now in year 44 of the 42-year saga.
Don’t bother. He’s an old school guy who will never acknowledge there’s any value to using anything other than your gut to judge teams/players.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
me too
but I use my eyes more so than the guts
by Johnny Bower's Pokecheck on Jul 29, 2010 6:19 PM EDT up reply actions
Everyone knows your gut has way more nerve endings than your eyes.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
Ok I'm sorry people have such a huge problem with Grabovski
being superior to Kessel… but there’s a very easy explanation for it.
This is on the basis of the likelihood of shots going in based on shooting location and shooting situation. Obviously Grabovski takes far fewer random outside shots than Phil Kessel.
Grabovski is also a very good player by the corsi metric, he’s a superior puck possessing centre ice man.
The idea everyone is fixated on that Grabo is an inferior player is frankly beyond my understanding. I watch the guy and I don’t get nearly as frustrated as many others on here appear to.
I also don’t think Kessel’s shooting locations are as great as some others might think. He takes a lot of stupid wrist shots from 40 feet out – a la Jason Blake… he just has a better shot so they tend to go in more.
The fact is, Kessel’s decisions to shoot instead of pass a lot of the time are also borne out by this evaluation.
Oh – and this stat has nothing to do with accuracy over time… it’s a measure of shots from a good scoring location for vs. shots from a good scoring location against… it’s like +/-… you don’t rack up a big number randomly in 50 games by sucking and lucking out.
Also the SOT adjustment takes out line mate and opposition and teammate effects.
The fact you guys don’t buy the result 100% doesn’t really change the outcome.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
I'm adding it to the toolkit...
as I originally mentioned in the first line of the posting.
I’m not saying Grabovski is the Leafs top forward… I’m saying this is one thing he does well. That’s not really a huge surprise to you guys is it?
He doesn’t take wild shots from all over the ice. He does a good job of maintaining control of the puck… he’s also good at retrieving it.
His corsi numbers are also very good, so its’ not like the puck possession aspect should surprise anyone.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:13 PM EDT up reply actions
You make a good point: this isn’t trying to measure “how good/effective” is the player (like GVT); it’s measuring a specific skill.
by The '67 Sound on Jul 29, 2010 2:15 PM EDT up reply actions
exactly
He generates more good scoring chances than he allows…
that’s all this means.
Full stop.
No need to deride advanced stats as meaningless if that’s all you take it to mean.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:16 PM EDT up reply actions
This is true, I think what it is is that the actual defintion of a stat often gets lost in the discussion.
I keep Behind the Net’s guide of what each stat is handy, and have to refer to it quite often because I forget what they are.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:17 PM EDT up reply actions
Yes
and the whole “kessel must be better thing” is hogwash.
Kessel scores… sometimes from BAD scoring chances. He also gives up a lot at the defensive end in missed assignments, and he’s soft on the puck, and he bails on coverages sometimes. He gives up almost as many as he gets. I don’t think that’s surprising.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:18 PM EDT up reply actions
Yeah, Kessel’s gift is probably chance conversion, not chance generation per se. He’s got a wicked wrister. He doesn’t have a better shooting % because he lets it go from just about anywhere.
by The '67 Sound on Jul 29, 2010 2:20 PM EDT up reply actions
If you had a player with Grabovski's
defensive pressure abilities, and stickhandling, and gave them Kessel’s shooting ability? You’d easily have a 100 point per year player.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:26 PM EDT up reply actions
I just got a really gross mental image, thanks Steve.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:27 PM EDT up reply actions
really? i just got a boner, actually.
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 6:38 PM EDT up reply actions
this is amazing
I am watching Steve go down with his ship…(of numbers) one comment at a time!
Helpful tip: the “Grabovski is good” argument makes a horrible life jacket
by Johnny Bower's Pokecheck on Jul 29, 2010 2:34 PM EDT up reply actions
Uhhhhh….Grabovski is good. More than one stat demonstrates this.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:35 PM EDT up reply actions
Among other things yes.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:42 PM EDT up reply actions
His
points per 60 minutes numbers are excellent for a 2nd line centre on a team surrounded by average players… not sure why that escapes notice.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:42 PM EDT up reply actions
Plus he goes hard into the corner is willing to take a hit to make a play, and is batshit crazy.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:43 PM EDT up reply actions
Yup. He’s made chicken salad out of chicken shit for two years running.
Not a tasty porterhouse, but chicken salad ain’t bad.
by The '67 Sound on Jul 29, 2010 2:45 PM EDT up reply actions
if steve burtch and I are looking at a bird
and I say “look Steve, a hummingbird”
Steve would say “not so fast Pokey” catch the bird, dissect it, run a blood analysis come back 3 months later after the results are in and state “Pokey, that was indeed a hummingbird”
by Johnny Bower's Pokecheck on Jul 29, 2010 2:45 PM EDT up reply actions
Okey Dokey Pokey Smokey.
(Do you smoke? If not, then my new nickname for you doesn’t work.)
I've always wanted to be a PPP Princess. You see kids, you can be anything you want to be; so long as Jay Leno doesn't also want to be that.
Weirdest. Analogy. Ever.
The thing is, these sort of stats can help us avoid mistakes or temper our expectations.
They would have really come in handy when we signed Jason Blake, we thought we were getting a 30 goal scorer when the stats clearly showed that season was an anomaly. If we would have had a reasonable impression to anchor our opinions with he might not have been so hated, or even more hated for signing him in the first place for that dollar amount and cap hit.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:49 PM EDT up reply actions
You didn’t even need advanced stats to see that Jason Blake wouldn’t score 30 goals, you just needed to look at his shooting % that season vs. his career average.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:50 PM EDT up reply actions
I know, but a lot of people saw his 30 goals and thought that’s what would happen here.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:52 PM EDT up reply actions
Yeah, once upon a time shooting percentage was an advanced stat.
by The '67 Sound on Jul 29, 2010 3:14 PM EDT up reply actions
Jason Blake is unlikeable
look at him
by Johnny Bower's Pokecheck on Jul 29, 2010 2:50 PM EDT up reply actions
Blake scored more than 30 goals once in his career – that 40 goal contract season at age 33. You didn’t have to look past career goals to see it was a poor signing.
BS
by MapleLeafMole on Jul 29, 2010 3:30 PM EDT up reply actions
maybe, but it gave me a good laugh.
Me: Any idiot could see that.
Mrs. P: You're not just any idiot, you're the most special kind of idiot.
by Sergei Puckizin on Jul 29, 2010 6:39 PM EDT up reply actions
Actually
based on this stat Jason Blake is a demon.
He generates a lot of scoring chances.
What you need is the context that goes along with shooting percentages… Jason Blake’s shooting percentages are generally atrocious. He might get a lot of shots, but he rarely if ever converts them.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
Grabovski isn’t a bad hockey player. He’s a good puck possession forward with a low scoring ceiling.
but no pop
he is REALLY good
at holding onto that puck. That I cannot argue
by Johnny Bower's Pokecheck on Jul 29, 2010 2:55 PM EDT up reply actions
Umm
how am I going down? I fail to see an argument that’s setting me up for any sort of failure here. What I’m getting out of this is that some people either don’t entirely understand what the stat is measuring, or what it’s value is as a measurement.
Some people that get what it measures also just don’t seem to care what it’s telling them, which is fine in it’s own right.
Don’t see how any of that ruins the point of what I’m saying.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:36 PM EDT up reply actions
There was some initial misunderstanding of what exactly the stat measured, but there isn’t much to argue with once that was cleared up.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:38 PM EDT up reply actions
I think the problem is for all his puck possession, and creation of scoring chances, if he doesn’t score/convert a lot of people won’t care. Goals matter
Need to emphasize tthat his stat highlights how defensively responsible he is. I think that part of his game is underrated
BS
by MapleLeafMole on Jul 29, 2010 2:42 PM EDT up reply actions
does Grabovski get onto your
penalty killing units?
His defensive game is similar to his offensive game. meh
by Johnny Bower's Pokecheck on Jul 29, 2010 2:56 PM EDT up reply actions
What’s the deal with the Grabbo hate?
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 3:00 PM EDT up reply actions
whats the deal with the "grabbo" love?
I know he is fodder on these boards for internet jokes but really..
by Johnny Bower's Pokecheck on Jul 29, 2010 3:02 PM EDT up reply actions
I wouldn’t say it’s love so much as, he’s a young decent number 2 center who has consistently put up good numbers with bad teammates. Plus we stole him from the Habs.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 3:04 PM EDT up reply actions
who was that 2nd rounder
anyways
by Johnny Bower's Pokecheck on Jul 29, 2010 3:05 PM EDT up reply actions
Not sure.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 3:09 PM EDT up reply actions
holy shat
I just tried looking it up and my head nearly exploded trying to understand where that hot potato ended up
by Johnny Bower's Pokecheck on Jul 29, 2010 3:14 PM EDT up reply actions
That was the 2nd round in the Boston deal for Kessel, so Jared Knight
Montreal traded to Chicago for Lang, and then we traded Calgary’s 2nd and our 3rd to Chicago for it.
Prefers pugnacity to truculence.
Ah, all I could figure out was that Montreal didn’t have it anymore. That second round pick has been useful.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 3:17 PM EDT up reply actions
so?
I know its early but do the beloved Buds not have a 1st, 2nd or 3rd round pick next year?
by Johnny Bower's Pokecheck on Jul 29, 2010 4:55 PM EDT up reply actions
the leafs have their own 2nd next year
You may be taking Jared a little too seriously
by JaredFromLondon on Jul 29, 2010 4:57 PM EDT up reply actions
How does that even relate to whether or not Grabovski is a good player??
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 5:00 PM EDT up reply actions
Probably just acknowledgement of the fact that, unlike most guys on the Leafs, his production is commensurate with his salary.
It’s a refreshing change of pace.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
He doesn’t get on my PK.
Both offense and defense are average. I just initially thought his defensive was below average.
BS
by MapleLeafMole on Jul 29, 2010 3:34 PM EDT up reply actions
It’s funny when someone that can’t follow what’s going on confirms it with the closing to their comment.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
How does the stat differentiate what a “bad” and “good” scoring chance are? Would this bust the old myth of “it’s never a bad play to put it on net”? Is it purely location on the ice?
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:21 PM EDT up reply actions
shot location
and the likelihood of an “average” shooter scoring from a given location on the ice.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:24 PM EDT up reply actions
I’d like to see the data for calculating how much an average shooter scores from a given location on the ice.
Something like that I could see being useful to a coach. IE, Ron Wilson could tell Kessel to shoot less when he’s at this location and more when he’s at this other location.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:25 PM EDT up reply actions
Teams can get that easily I'm sure
in fact scouts track shot location from the press box consistently… so most teams probably have stock piles of data for the entire NHL.
It’s not difficult to figure out if you have the raw data… and I’m pretty sure Wilson et al have that kind of stuff.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:26 PM EDT up reply actions
But I don’t.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:27 PM EDT up reply actions
No you don't.
And as far as I know, that info isn’t easy to rip off the internets… if you want to look, ESPN records all that info for every game… but I don’t have any VB code written to rip the data, so at this point we’re going to have to trust the guys with the energy to actually devote to this stuff (i.e. Gabe Desjardins, Tom Awad, Vic Ferrari, etc.)
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:29 PM EDT up reply actions
sure
I just think that this is one that would be worth doing and am wondering why they haven’t.
The NHL had that interesting graphic of where Crosby, Stamkos and Ovechkin (I think?) scored their goals this year, it’d be interesting to compile that for Kessel, which seems to have already been done in order to calculate this Delta stat.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:31 PM EDT up reply actions
sure
I just think that this is one that would be worth doing and am wondering why they haven’t.
The NHL had that interesting graphic of where Crosby, Stamkos and Ovechkin (I think?) scored their goals this year, it’d be interesting to compile that for Kessel, which seems to have already been done in order to calculate this Delta stat.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:31 PM EDT up reply actions
well that was weird
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:31 PM EDT up reply actions
Yeah
if we got all the raw data we could come up with any sort of stat we wanted and wouldn’t be at the whims of the guys from puck prospectus… but that would to some extent be like open sourcing the whole process.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:33 PM EDT up reply actions
I like Grabovski and think he’s a useful player for the Leafs.
You’re smoking crack if you think he does anything better than Kessel.
Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.
by Chemmy on Jul 29, 2010 6:18 PM EDT up reply actions 1 recs
Zone Shift?
Kessel = -43, Grabovski = +13
Corsi Rel QoC?
Kessel = -0.035, Grabovski = 0.138
BTN Rating?
Kessel = -0.12, Grabovski = 0.51
There’s actually a lot of stats where Grabovski is superior to Kessel… most of which relate to defensive play and puck possession. Kessel is soft on the puck, and his defensive play leaves something to be desired… He needs work to be a complete player. I hope this isn’t news.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
Another reason I posted this info though
is it sort of contradicts all the articles out there about the Leafs great Corsi rating. Sure we outshoot the opposition a bunch, but the Leafs take a lot of shots from crappy scoring locations, and they tend to have a lot of guys that are unlikely to score on those shots taking them.
If the Leafs focused more on having good shooters shoot… and doing it from the right spots on the ice… they’d be a lot further ahead in the long run in my estimation.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
Where is the data that proves the reason the Leafs don’t score on these chances are because they’re from crappy scoring locations?
Am I missing something?
I believe the same about the Leafs focusing on having good shooters shoot more from locations where it’s likely they’ll score, but I’m not seeing the connection in how this stat helps to prove that they don’t do that.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:49 PM EDT up reply actions
I thought Steve’s point was that the Leafs fare very well on corsi consistently underperform vs. expectations on that measure. This Delta stat suggests that while we may take a lot of shots we don’t take terribly good ones.
Which is what a lot of people have been saying for a while, and this stat now confirms.
The Leafs were a historical outlier in terms of defying corsi projections. This helps explain why.
by The '67 Sound on Jul 29, 2010 3:16 PM EDT up reply actions
Oh
I just took that to mean that corsi was somewhat flawed and doesn’t account for goaltender performance enough.
But your explanation makes more sense.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 3:35 PM EDT up reply actions
Yeah, my understanding of Hawerchuk’s work is that over a long enough period of time corsi is a good predictor of team performance—except for the Leafs.
by The '67 Sound on Jul 29, 2010 4:48 PM EDT up reply actions
Except for the Leafs
Which is a real pain in the ass.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
Generally speaking
this data confirms it.
Aside from Grabovski, the average Leaf is getting slightly more chances than he’s giving up from good shooting position on the ice in a given shift. If a Leaf is getting chances from a good location, odds are he’s surrendering the equivalent. The good thing is Versteeg has been playing in Grabovski territory, and Armstrong has been playing with teammates with such low ratings, he might be a solid plus player with decent compatriots.
Overall though, we take a lot of shots, from bad locations on the ice. We have control of the puck, but we waste it consistently.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:43 AM EDT up reply actions
Gabe’s looked at the Leafs’ being outliers in terms of Corsi. It’s partly a function of always being behind in the score.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
this is why I was thinking that Corsi needed to take into account the goaltender’s performance somehow.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 5:04 PM EDT up reply actions
What I mean is, say the Leafs take 30 shots in a game which works out to something like 4 goals in a game. Their opponent would only need to take something like 20 shots on Toskala to get 4 goals… so even though we outshoot them, we don’t really outchance them.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 5:05 PM EDT up reply actions
Yeah, horrible goaltending is part of the equation (why we allow more goals than corsi would predict). But we also score a lot less than corsi figures would predict.
Usually the “but they took weak shots” argument (see e.g. Caps vs. Habs; Canada vs. Swiss) is incorrect but for the Leafs it would appear to be true.
by The '67 Sound on Jul 29, 2010 5:15 PM EDT up reply actions
Right but Corsi is trying to show you which players drive offense, not which players lose because their goalie stinks.
Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.
I mean about team Corsi though.
Individual corsi makes sense to me.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 6:45 PM EDT up reply actions
No...
this is a misconception…
Corsi is showing you which players drive possession and thus shots on goal as a result… not offense in general. You could have a player that excels at taking control of the puck in his own end, but perpetually fires shots from the blue line (Jason Blake?). That doesn’t make him a top 10 NHL player, despite the fact that he’s in the top 5 in the NHL in shot attempts.
Scoring chances and the quality of them in regards to offense is a bit more detailed than JUST corsi numbers… the context, and the more detail we can get on it, means we’ll have a better idea of what we’re building as a team.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:49 AM EDT up reply actions
Corsi
is completely unrelated to a goaltender’s performance.
It just displays possession. The fact that the Leafs have a high corsi number generally speaking just means they have the puck a lot in the offensive zone.
If you’re worrying about the save percentages of goalies on the other side of the Leafs, just look at their shooting percentages. If their shooting percentages are low… they’re not scoring much… which means the goalies are saving them.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
The effects
on this stat based on team score is actually discussed in the longer winded break down by Awad, and it isn’t significant enough to explain all of this away.
Even in Gabe’s detailed analysis, it really doesn’t explain or justify the discrepancy enough to qualify the Leafs season last year. They fucking blew it last season. They didn’t score enough for the frequency of shots they were taking, and they gave up far too many goals for the number they allowed. This stat goes into some level of detail to explain that fact, but it also points out the fact that we have a lot of sub-par shooters doing the shooting for us.
I’d like to see Kulemin, Bozak, Versteeg, and Grabovski do more shooting next year, and perhaps have Kessel distribute the puck more. I’m also curious to see how Kadri does as a top 6 forward.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:47 AM EDT up reply actions
Tom Awad
actually discusses the impacts of the score and game situation when he runs down the idea behind this stat, and he actually checked his data against scoring situations… the full description is in his initial posting on it.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
To me this seems like an intermediary stat that should be a part of calculations for a stat that would be more easily understood and with greater merit.
Sure it’ll say a guy like Grabovski can create chances, but if he never finishes them then where’s the benefit? On the flip side it says that Kessel doesn’t create a lot of quality chances, but we know that he is able to capitalize on the ones he does generate.
If Awad could incorporate scoring here then I’d think he’d be on to something. But even then I’m not so sure that such a stat wouldn’t merely be the equivalent of GVT.
I've always wanted to be a PPP Princess. You see kids, you can be anything you want to be; so long as Jay Leno doesn't also want to be that.
The whole point of the stat
was to find a metric that was more similar to classic +/- than corsi is.
A measure of scoring CHANCES rather than puck possession proxy via shot’s attempted for and against.
If a player fires a million shots from his own blue line that’s a pretty useless shot, but it counts just the same as a shot from 5 feet out in the corsi metric.
It’s just another piece of the overall puzzle. I really don’t see why this is any more complicated than Corsi… you’re adding in the idea of shot location and game situation… is that really that bad?
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:34 PM EDT up reply actions
+/- isn’t at all about chances from what I’m seeing though… it’s about real scoring. Although it doesn’t include things like PP etc which is why it’s not very useful. Neither Corsi nor Delta seem to be good solutions for the problems of +/- as far as I can see.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:36 PM EDT up reply actions
This stat
correlates more closely to actual +/- than Corsi does.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:37 PM EDT up reply actions
….because it’s partially based on +/-. If it didn’t, there would be real problems.
this stat isn’t a five-course meal, it’s a gumbo. You can tell whether it’s good or bad, but it takes a lot longer to figure out the recipe for success than if you simply had the ingredients in front of you.
by Nirbo on Jul 29, 2010 2:39 PM EDT up reply actions 1 recs
no there's no relation between the original stat
and +/- as it’s tabulated.
You’re using shot location in adjusting the basic corsi metric.
Goals are irrelevant, so +/- doesn’t factor in.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:41 PM EDT up reply actions
people
generally use 5 on 5 measurements for corsi and in this case I did so for delta also.
so the +/- vs. corsi vs. delta thing shouldn’t be such an issue since they’re ALL using 5 on 5 data.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:38 PM EDT up reply actions
Also
I don’t think the problem with +/- not being useful is a result of NOT including PP data. It has a lot more to do with it not correlating to success on the ice overall, largely because there’s no adjustment for situational play. Players who play in front of a bad goalie suffer from crappier +/-, while players in front of a good one get higher +/-.
There’s other factors that also go into what is basically a team stat. This is ALSO a team stat… think of it as generation of scoring chances when the player is on the ice, minus how many happen against.
When Grabovski is present, the team generates a lot more good offense than when he isn’t… that’s ALL this means.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 29, 2010 2:40 PM EDT up reply actions
alright
when i said “not includion PP data” I was trying to say “adjusting for situational play” in a very inaccurate manner.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:44 PM EDT up reply actions
alright
when i said “not includion PP data” I was trying to say “adjusting for situational play” in a very inaccurate manner.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:44 PM EDT up reply actions
don’t think the problem with +/- not being useful is a result of NOT including PP data. It has a lot more to do with it not correlating to success on the ice overall, largely because there’s no adjustment for situational play.
The Tomas Kaberle Case [and I bet someone does have actual numbers available but I’m not going to break down game-by-game and an just going to guess]:
Kaberle was -16 last season if I recall.
Kaberle played in front of a crap goalie on a crap team that was often behind.
Leafs gave up 11 empty net goals last season. I’d bet TK was on the ice for about 7 (ie -7) or so of those since it is the type of situation he was always utilized in.
Leafs gave up 7 shorties last year. Likewise I would bet TK was likely on the ice for about 4 or 5 (say, -4) of those because he played huge minutes on the PP.
TK would also be utilized lots when the team was behind late and they were taking high risk chances trying to catchup. I’d guess that would count for about 3 or 4 more goals against (say -3).
So there are somewhere near -14 of his -16 accounted for by situations he was in that a Luke Schenn or a Mike Komisarek would never see. Some of these situations he could not possibly get a + and in some the odds are stacked against him getting anything but a minus. Not to say that some of the SHA and EN were not partly his fault, but maybe he was just not that defensively bad at even strength taking out some of the oddball situations. -2 is not bad on a 2009-10 Leafs team.
"I'd walk into the Leafs dressing room to get ready for the day and Harold would be there in his boxer shorts shaving. King Clancy would drop by a little later, play the fool, and then head off to the racetrack." John Brophy
by Mike Pelyk's Hairdo on Jul 29, 2010 3:15 PM EDT up reply actions
When Grabovski is present, the team generates a lot more good offense than when he isn’t… that’s ALL this means.
Yes. At one point early last year Grabovski was among the top 20 players in getting the puck down the ice and on the opponent’s net. Not for the Leafs, in the entire league.
Pension Plan Puppets: A Toronto Maple Leafs blog and a group therapy session.
Like reading thoughts confined to 140 characters? I'm on Twitter too.
also, he has fabulous hair
You may be taking Jared a little too seriously
by JaredFromLondon on Jul 29, 2010 5:04 PM EDT up reply actions
The only thing I can think of that might affect shot location is the opposing goalies. Maybe teams are told to shoot from a certain side of the ice. For example, I bet opposing teams were told the best place to shoot when facing Toskala was from their own red-line.
The Guess Who sucked, the Jets were lousy anyway
by Plea From A Cat Named Felix on Jul 29, 2010 2:37 PM EDT up reply actions
Glove side high.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 2:38 PM EDT up reply actions
Jaroslaugh Hahalak
I've always wanted to be a PPP Princess. You see kids, you can be anything you want to be; so long as Jay Leno doesn't also want to be that.
spamming, close you are getting to it
You may be taking Jared a little too seriously
by JaredFromLondon on Jul 29, 2010 3:18 PM EDT up reply actions
Kessel was the best player on the team at creating quality chances.
by Slava Duris #24 on Jul 29, 2010 10:15 PM EDT up reply actions
Ok I'm going home and then out for the night
so this discussion will continue without me for a while. Hope everyone has a clearer idea of what the intent behind the numbers was. I’ll check in later.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
smell test?
To paraphrase:
In 2009-10, Jason Blake (+9.2 Dsot on the Leafs) produced more good scoring chances than he allowed, while Ilya Kovalchuk (-4.5 Dsot) allowed more good scoring chances than he produced.
In 2009-10, Mikhail Grabovski (+9.1 Dsot) produced more good scoring chances than he allowed, while David Krejci (-7.7 Dsot) allowed more good scoring chances than he produced.
I can keep going with this; please tell me that Daniel Carcillo was superior on the offensive-defensive aggregate to Joe Thornton. Or Rob Scuderi (11 points) more valuable than Shea Weber (43 points).
There is a counter-argument that this stat is just what it is, and shouldn’t be taken as a “which player is better” number. This is true to an extent, but unlike Corsi or Off%, which are pretty clear about what they represent, DeltaSOT sure seems to be another attempt at a “superstat” which combines player value and situations into one easily-compared number.
Though it does seem to have some value, there is too much noise in this one for me to take it overly seriously. It’s difficult to evaluate without discussion of variance and significant differences. Probably something to bring up at PP rather than PPP, always great to have these discussions started here.
I don’t think it’s a big surprise that Kovalchuk’s lackadaisical defensive play has been costly, although he balances that out by being one of the top snipers in the league. He hasn’t been playing on a very good team most years either so his bad defensive coverage is exacerbated a bit by that too, playing with the New Jersey Devils would most likely hide his weaknesses more effectively.
As far as this Delta SOT stat goes, the thing that looks really skeptical to me, as someone who tracked scoring chances for the Leafs last season, is Grabovski rated so far ahead of Kessel and Kulemin. It all seems sketchy to me, but I’ve kind of soured on shot quality metrics anyway over the past few years. The NHL does a far from perfect job of accurately tracking shot information.
I don’t get the hate/dislike some have for Grabovski however. Maybe it’s the same reason why Stajan, Ponikarovsky, Blake, etc. are supposedly terrible hockey players according to some (the message-board crowd); he’s a pretty good but flawed hockey player who has a quirk or two that annoys some fans, who decide that for that reason this guy must be a detriment to the team and they ought to be rid of him.
by Slava Duris #24 on Jul 29, 2010 10:04 PM EDT up reply actions 1 recs
Having read
your work on scoring chances for the Leafs, I can respect and understand the disdain you have for shot metrics and recording from NHL statisticians.
That being said, the same statisticians are recording the data for all of the players on the same team over the same season no?
I honestly think these sorts of advanced stats are MOST valuable for comparing players on the same team… and that’s what this is an attempt at accomplishing.
To me the indictment here is more of Kessel’s attention to the defensive aspects, and his softness on the puck. Kulemin’s raw Delta number isn’t actually that bad, but his team-mate effects depress his overall Delta SOT rating because he was playing with Kessel and Bozak for a large chunk of the year.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Aug 1, 2010 12:43 AM EDT up reply actions
Ok ... let's think about this
David Krejci… in 2008-09 had a Delta SOT of 6.8… Mikhail Grabovski had a -1.1. Yeah obviously the stat is flawed from a year ago.
In 2009-10 Kovalchuk had a Delta SOT of 3.1 JUST with NJ, but produced a -4.5 with Atlanta… Blake produced a 1.3 with Anaheim, and a 9.2 with Toront… that of course is factoring out Blake’s atrocious shooting percentage and Kovalchuk’s ridiculously good shooting percentage.
All of this is stemming from a misunderstanding of the statistic. It ignores the players ability to score from a given opportunity. It rates scoring chances on the basis of an “average shooter”‘s ability to score from them… since Kovalchuk has regularly shown an ability to score well above an average shooter, and Jason Blake has generally produced at a lower rate, then it isn’t remotely surprising that the numbers work out this way.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 2:32 AM EDT up reply actions
All of this is stemming from a misunderstanding of the statistic. It ignores the players ability to score from a given opportunity. It rates scoring chances on the basis of an "average shooter"‘s ability to score from them…
I do understand that, and to me, that’s precisely the problem with this stat. It tells us whether shots tend to get taken from the “money areas” when a given player is on the ice, but unless you know something about the quality of the shooter, that doesn’t really translate to chances.
If, for example, Bozak tends not to have a lot of shots from the zones you want, a lot of that is likely the fact that his prime job is to set up Kessel, who fires a lot from the outside. Poor percentage chances? For an average shooter, maybe, but Kessel has outside speed and a quick release and can score from places where Colton Orr can’t.
Leaf, the universe and everything.
Now in year 44 of the 42-year saga.
Ok two problems with this logic
The idea that any out of the majority of NHL shooters are better than an “average NHL shooter” is completely mistaken based on the work done by Gabriel Desjardins. He has done enough playing with the numbers that shows that there are roughly 2 shooters in recent history that consistently outperform the shooting percentages that would result from an “average shooter” taking shots from where they shoot from.
For evidence I’ll direct you here, and here.
Those shooters are Ilya Kovalchuk, and Alex Tanguay. For all intents and purposes the rest of the NHL’s shooting is generally within a range of acceptability we term the “average shooter”. The guys who are taking most of the shots in the NHL have roughly equivalent talent to pick corners. Most of the variation we observe in shooting percentages are luck, and over a long time frame, we expect shooters to regress towards their natural shooting percentage.
Obviously Colton Orr is not taking a lot of shots, but Phil Kessel is… this is likely due to the fact that Kessel is superior to Orr in regards to his shooting ability. That does not mean that Kessel is amazingly superior in terms of his shot to a player like Tyler Bozak or Nikolai Kulemin. He may be, but he hasn’t concretely demonstrated it over the course of his 4 years in the NHL.
Either way, I digress. The point is, the ONE thing players actually have a lot of control over is where they shoot the puck from. This is a learned, and repeatable skill. Getting to good scoring areas is something worth doing. The fact that Grabovski does so consistently is very important, because it means he’s MORE likely to generate good offense. Kessel shooting from all over the ice is generally unlikely to result in a higher number of goals, and in fact will probably depress his overall shooting percentage. If he fired a better selection of shots from high percentage areas, he would see higher shooting percentages overall.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Aug 1, 2010 12:37 AM EDT up reply actions
If you follow those links through, he says that
But the way a forward plays – the spots he gets to to take his shots – is barely more important to his shooting percentage than his shooting ability even though it is very consistent even over the course of five seasons.
That is – getting to the good spots is only marginally better than being able to shoot the puck.
Further, of everything he lists as factors, 57% falls under “transient.” There is a lot less noise in the “getting to your spots” stats, but that falls well below basic scoring ability.
Leaf, the universe and everything.
Now in year 44 of the 42-year saga.
I like () Mitchell 6.3 but I hated () Blake +9.2 . I like () Kaberle 3.9. I’m totally ambivalent (/- 0) about Colby “Cheese-Boy” Armstrong.
"I'd walk into the Leafs dressing room to get ready for the day and Harold would be there in his boxer shorts shaving. King Clancy would drop by a little later, play the fool, and then head off to the racetrack." John Brophy
by Mike Pelyk's Hairdo on Jul 29, 2010 4:18 PM EDT reply actions
message fail … I guess I better not use brackets and minus signs in the future … guess I coulda previewed too
"I'd walk into the Leafs dressing room to get ready for the day and Harold would be there in his boxer shorts shaving. King Clancy would drop by a little later, play the fool, and then head off to the racetrack." John Brophy
by Mike Pelyk's Hairdo on Jul 29, 2010 4:19 PM EDT up reply actions
Useful link
Here is Behind The Net’s FAQ about advanced stats. I’m serious about having to refer to it quite often, and know exactly what each stat is supposed to measure is immensely helpful in following the discussion.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Jul 29, 2010 5:28 PM EDT reply actions
Grabovski?
Who would’ve thought he would have the highest Delta. I guess missing 20 games this season has a lot to do with it.
It's a +/-
stat, so to some extent that is a possibility… but if he were perfectly average, then a zero score would result (or something close by).
Barring horribly wild swings in his production, such a high score makes no sense as a result of a lower number of games played.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Aug 1, 2010 12:45 AM EDT up reply actions
I'm going to apologize
post morten for all of my end of night comments. If they come off as harsh, sorry. If they don’t make sense… well that’s where I’m at right now.
I’m sure you’ll generally forgive me, but yeah, I’ll discuss this whole thing in more detail when I show the stats for the D on the team.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
All this statistical wonkery gives me the brain jimmies
You may be taking Jared a little too seriously
by JaredFromLondon on Jul 30, 2010 3:13 AM EDT reply actions
Steve,
First off, it’s great that you and others get these advanced discussions started, so don’t get offended if others bring skepticism to the table. The argument will make all of us wiser.
Asserting that a disagreement on the value of a stat is based on a lack of understanding can be taken as a bit of an Emperor’s New Clothes line.
It also sidesteps the actual criticisms, which seem to summarize to
1) what is it really that you’re measuring?
2) how noisy is this measurement given the complicated model?
Again from the 09-10 even-strength chart:
raw Delta correlates with deltaQT @ r=-.076 and deltaQS @ r=-.048, but what looks like an insignificant .08 with delta QO.
Huge effect of your teammates’ delta, not surprising since it measures what happens when you’re on the ice, not just your own shots. Hence the Grabovski-Blake delta-superstar combo. Curious that who you play against seems to be just noise comparatively speaking.
Grabovski and Blake both seem like the sort to juke this stat pretty badly: they take tons of shots from short and medium distances but bad angles, and what giveaways they do cause are met by a defense that is terrified to give up any shots at all, even bad ones, due to historically bad goaltending.
Ok
you can ignore some of my commentary from last night… drinking will do that.
As for the player impacts – I do agree with you that it’s a bit odd that players you play with have a much greater impact than players you play against, but there’s a logical explanation for that. You tend to play with the same players far more often than you play against the same players. A single player might take over 50% of his shifts with the same 2 other forwards, but there’s no chance if he plays over 30-40 games, that he’s taking over 50% of his shifts AGAINST the same forwards/defense.
As for Grabovski vs. Blake… I’ll buy the point regarding Blake shooting from all over the ice, but I won’t really buy it for Grabovski, and I think the key point this stat is making is both are productive at the offensive end of the ice in terms of generating shots on goal, but both are consistently solid DEFENSIVELY enough to pursue and gain control of the puck while not really allowing many chances against.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Jul 30, 2010 7:37 AM EDT up reply actions
So does this stat imply that grabs is Kessel’s best linemate, because he would be getting him shots in better areas? Or is this a misreading? (not sure that it’s true in reality)
I suppose that would be true
in regards to the fact that this statistic is measuring TEAM scoring chances when a player is on the ice.
You also have to remember the fact that Bozak played half his rookie NHL season, and some of the Grabovski results would stem from the fact that he is defensively aware… probably more so than Bozak.
Over time, Bozak may improve.
I’m not saying that this shows one thing or another consistently, it’s just another interesting set of info to examine with respect to the Leafs forwards.
"Success is the ability to go from one failure to another with no loss of enthusiasm."
- Sir Winston Churchill
I'm pretty sure he's talking about the Leafs.
by Steve Burtch on Aug 1, 2010 12:40 AM EDT up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
In Association With
Latest Puck Daddy - NHL from Yahoo! Sports
- Fight Video: Ilya Kovalchuk KOs Flyers’ Brayden Schenn with nasty right hook
- Penguins’ Brooks Orpik launches Daniel Paille with massive hit (Video)
- Penguins, Bruins face their challenges in Eastern Conference showdown
- Olli Jokinen’s slump-buster is a Big Mac meal from McDonalds (Video)
- Friday’s Three Stars: Brassard wins it for Blue Jackets in overtime; Halak blanks Kings
SBNation.com Recent Stories
StatCounter
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
