WTG Posted January 12, 2017 Share Posted January 12, 2017 2 minutes ago, xereau said: Corsi this, plus minus that -- Sutter is all about the intangibles. He is a bargain at his current contract (thanks Benning!) and it was a very good move signing a foundation type player that brings it at all levels. Sutter's family pedigree is probably the best in the entire league. He just oozes humble leadership, and is a stabilizing force on a team that has the potential to fly apart at the seams. He is best at center, and now that Bo is emerging as a star, Sutter can be the best 3rd line center in the league, exactly where he belongs. Why is Sutter a bargain? Why does Sutter's family matter? Why is he a stabilizing force? Why is he the best 3rd line center in the league? The only argument I see there is the leadership argument. There is history behind it, but you still have to back that all up with actual logical arguments. If I didn't know about his past showcases of leadership then I would question it too. If you are going to make claims then you have to back them up. Link to comment Share on other sites More sharing options...
WTG Posted January 12, 2017 Share Posted January 12, 2017 1 minute ago, coastal.view said: i have absolutely no interest at all in getting into any debate over analytics but you are delusional (and uneducated) if you think those are facts as in scientific facts putting together some analytic theory involving numbers does not make the outcome facts at best these approaches are weak hypotheses and as many on here are noting there is need for more testing and revision to these hypotheses the analytic theory may move closer to being factual in time but even in baseball, where this approach has more history, i doubt many would call these numbers facts Fine, you can call them weak hypotheses, but they are the best non-anecdotal analysis of a player we got. We have to work with what we got. Look, I'm willing to concede that if you do not believe in statistics that's fine. But there are some statistics in there that are just facts. Like, we allow a bunch of unblocked shots in the slot when Sutter is on the ice vs when he's off the ice. When Sutter is on the ice, we only score 35.6% of the goals even strength. Reject stats like corsi or fenwick all you want, but there are just stats that are facts. Link to comment Share on other sites More sharing options...
coastal.view Posted January 12, 2017 Share Posted January 12, 2017 6 minutes ago, WTG said: Fine, you can call them weak hypotheses, but they are the best non-anecdotal analysis of a player we got. We have to work with what we got. Look, I'm willing to concede that if you do not believe in statistics that's fine. But there are some statistics in there that are just facts. Like, we allow a bunch of unblocked shots in the slot when Sutter is on the ice vs when he's off the ice. When Sutter is on the ice, we only score 35.6% of the goals even strength. Reject stats like corsi or fenwick all you want, but there are just stats that are facts. we can do a dance about this all day analytics is a new model being developed to assess player values it is extremely limited at present it is often wrong no team uses them the way posters on here do yet people on here argue about them as if they reveal truth (or are factual) they simply are not they are just a new model that contains many many flaws and is quite elementary still (edit: it does appears florida and arizona were slow to realize this ) Link to comment Share on other sites More sharing options...
apollo Posted January 12, 2017 Share Posted January 12, 2017 41 minutes ago, oldnews said: Not sure if serious. Bo was -30 last year... Minus -22 from start of the season to early Jan 2016... he wasn't playing well. He turned his game around big time and was only minus 8 the rest of the way... he was +6 the last 6 games of the year... (I had to look at the game log to support my argument, but even from memory, it was obvious he was playing poorly first half of last year and the minuses were pretty much due to his play... you can blame it on vrbata to an extent too) It was obvious he had started to turn things around. What was more obvious was his work ethic this year to get in better shape and make him a better player. Also... I ran into him and told him good luck this year. I think that probably put him over the top to become ELITE Link to comment Share on other sites More sharing options...
oldnews Posted January 12, 2017 Share Posted January 12, 2017 2 minutes ago, WTG said: But if you want to claim things you typically have to back it up. You claimed all of his defensive zone starts ect... I showed that he only started 10% of his shifts in the defensive zone. You don't post "facts" wadr - you post coloured lines. Here are Sutter's zone starts: 29.7% offensive zone starts 36.8 % defensive zone starts 33.5% neutral zone In other words, 43.3% offensive zone starts relative to defensive. What makes the context of Sutter's play - and his possession numbers more complicated - and contradictory - that most players is his dual role. On the one hand he centers a shutdown/matchup line. On the other, he's plays wing on the Sedin line. The Sedins enjoy some extremely opportune ozone starts - 60.5% (Hank) 57.8% (Daniel) Corsi Sutter 46.2 Hank 48.8 Daniel 47.4 Now, when you take Sutter's ozone starts when playing with the Sedins to be in the 58-60% range - but overall at 43.3%, what does that tell you about the context of his play when he's not playing with the Sedins. Whether with them - and facing opposition shutdown lines/pairings or without them and playing a shutdown role against the opposition's best, Sutter is facing strong quality of competition regardless - and always has, despite the fact that the context of his play changes significantly. Nevertheless, his corsi remains within 1.2% of Daniel, and 2.6% of Henrik. But, but, he's the worst center on the team, he's a horrible possession player, etc, etc. If people are satisfied with those kind of elementary and uninformed misconceptions about the player - based on extremely weak 'analytics' - that's your prerogative. Link to comment Share on other sites More sharing options...
coastal.view Posted January 12, 2017 Share Posted January 12, 2017 Just now, apollo said: Bo was -30 last year... Minus -22 from start of the season to early Jan 2016... he wasn't playing well. He turned his game around big time and was only minus 8 the rest of the way... he was +6 the last 6 games of the year... It was obvious he had started to turn things around. What was more obvious was his work ethic this year to get in better shape and make him a better player. Also... I ran into him and told him good luck this year. I think that probably put him over the top to become ELITE and part of his improvement is directly related to sutter being healthy and taking all the most difficult matchups (except when other team has last change and can avoid it ) so that horvat has less defensive responsibility and can focus his energies on other areas and burrows gets some credit too as when he joined that line he freed up bo even more and that is when bo began to become more of an offensive presence this year bo is not doing it by himself . . the coach has helped his development, burrows has, sutter's presence has changed his game, our improved defence has had an impact...etc Link to comment Share on other sites More sharing options...
oldnews Posted January 12, 2017 Share Posted January 12, 2017 2 minutes ago, apollo said: (I had to look at the game log to support my argument, but even from memory, it was obvious he was playing poorly first half of last year and the minuses were pretty much due to his play... you can blame it on vrbata to an extent too) Nonsense apollo. Horvat wasn't the x factor. Baertschi waking up at the half point was. Vrbata checking out every time he wasn't with the Sedins was. There are few things more misleading and less signficant than Horvat's +/- last year - simple as that. You can beat it to death all you want, or you can elect to figure out how irrelevent +/- can be, particularly when you're talking about a player - in the absence of Sutter - whose line was stepping up over it's head, and whose linemates were passengers, one for half the season, the other for the entirety of it. Anyhow, my apologies but that's all the time I can waste on plus/minus wadr. Link to comment Share on other sites More sharing options...
coastal.view Posted January 12, 2017 Share Posted January 12, 2017 3 minutes ago, oldnews said: You don't post "facts" wadr - you post coloured lines. Here are Sutter's zone starts: 29.7% offensive zone starts 36.8 % defensive zone starts 33.5% neutral zone In other words, 43.3% offensive zone starts relative to defensive. What makes the context of Sutter's play - and his possession numbers more complicated - and contradictory - that most players is his dual role. On the one hand he centers a shutdown/matchup line. On the other, he's plays wing on the Sedin line. The Sedins enjoy some extremely opportune ozone starts - 60.5% (Hank) 57.8% (Daniel) Corsi Sutter 46.2 Hank 48.8 Daniel 47.4 Now, when you take Sutter's ozone starts when playing with the Sedins to be in the 58-60% range - but overall at 43.3%, what does that tell you about the context of his play when he's not playing with the Sedins. Whether with them - and facing opposition shutdown lines/pairings or without them and playing a shutdown role against the opposition's best, Sutter is facing strong quality of competition regardless - and always has, despite the fact that the context of his play changes significantly. Nevertheless, his corsi remains within 1.2% of Daniel, and 2.6% of Henrik. But, but, he's the worst center on the team, he's a horrible possession player, etc, etc. If people are satisfied with those kind of elementary and uninformed misconceptions about the player - based on extremely weak 'analytics' - that's your prerogative. you said it better than i ever could have given my aversion to analytics until the models are improved and more reliable Link to comment Share on other sites More sharing options...
aGENT Posted January 12, 2017 Share Posted January 12, 2017 3 minutes ago, oldnews said: You don't post "facts" wadr - you post coloured lines. Here are Sutter's zone starts: 29.7% offensive zone starts 36.8 % defensive zone starts 33.5% neutral zone In other words, 43.3% offensive zone starts relative to defensive. What makes the context of Sutter's play - and his possession numbers more complicated - and contradictory - that most players is his dual role. On the one hand he centers a shutdown/matchup line. On the other, he's plays wing on the Sedin line. The Sedins enjoy some extremely opportune ozone starts - 60.5% (Hank) 57.8% (Daniel) Corsi Sutter 46.2 Hank 48.8 Daniel 47.4 Now, when you take Sutter's ozone starts when playing with the Sedins to be in the 58-60% range - but overall at 43.3%, what does that tell you about the context of his play when he's not playing with the Sedins. Whether with them - and facing opposition shutdown lines/pairings or without them and playing a shutdown role against the opposition's best, Sutter is facing strong quality of competition regardless - and always has, despite the fact that the context of his play changes significantly. Nevertheless, his corsi remains within 1.2% of Daniel, and 2.6% of Henrik. But, but, he's the worst center on the team, he's a horrible possession player, etc, etc. If people are satisfied with those kind of elementary and uninformed misconceptions about the player - based on extremely weak 'analytics' - that's your prerogative. Much better (Your first reply was lazy ) Link to comment Share on other sites More sharing options...
oldnews Posted January 12, 2017 Share Posted January 12, 2017 19 minutes ago, mll said: Even Strength Line Combinations Freq Line Combination 34.3% COMEAU,BLAKE - MALKIN,EVGENI - SUTTER,BRANDON 25.6% DOWNIE,STEVE - SPALING,NICK - SUTTER,BRANDON 24.4% BENNETT,BEAU - DOWNIE,STEVE - SUTTER,BRANDON 11% DOWNIE,STEVE - SUTTER,BRANDON - WILSON,SCOTT 4.7% LAPIERRE,MAXIM - SUTTER,BRANDON - WINNIK,DANIEL - See more at: http://dobberhockey.com/players/brandon-sutter/2014-2015/playoffs#sthash.Bya1rnak.dpuf That's a very misleading graph you've posted - from a playoff sample of 5 games. Here are his last 3 full seasons in Pittsburgh: 2014-2015 Line Combinations & Production Even Strength Line Combinations Freq Line Combination 32.5% DOWNIE,STEVE - SPALING,NICK - SUTTER,BRANDON 21.9% BENNETT,BEAU - SPALING,NICK - SUTTER,BRANDON 19.8% BENNETT,BEAU - DOWNIE,STEVE - SUTTER,BRANDON 14% COMEAU,BLAKE - MALKIN,EVGENI - SUTTER,BRANDON 11.7% DUPUIS,PASCAL - MALKIN,EVGENI - SUTTER,BRANDON 2013-2014 Line Combinations & Production Even Strength Line Combinations Freq Line Combination 27.3% ADAMS,CRAIG - GLASS,TANNER - SUTTER,BRANDON 22.1% CONNER,CHRIS - SUTTER,BRANDON - VITALE,JOE 18.7% GLASS,TANNER - MEGNA,JAYSON - SUTTER,BRANDON 18% GLASS,TANNER - PYATT,TAYLOR - SUTTER,BRANDON 13.8% CONNER,CHRIS - EBBETT,ANDREW - SUTTER,BRANDON 2012-2013 Line Combinations & Production Even Strength Line Combinations Freq Line Combination 65.7% COOKE,MATT - KENNEDY,TYLER - SUTTER,BRANDON 11.1% COOKE,MATT - MORROW,BRENDEN - SUTTER,BRANDON 8.4% COOKE,MATT - NEAL,JAMES - SUTTER,BRANDON 8% COOKE,MATT - GLASS,TANNER - SUTTER,BRANDON 6.8% COOKE,MATT - DUPUIS,PASCAL - SUTTER,BRANDON Link to comment Share on other sites More sharing options...
oldnews Posted January 12, 2017 Share Posted January 12, 2017 8 minutes ago, J.R. said: Much better (Your first reply was lazy ) No, this one was redundant. But, but Sutter's corzzi with the Sedinz was 33%!! Link to comment Share on other sites More sharing options...
mll Posted January 12, 2017 Share Posted January 12, 2017 9 minutes ago, oldnews said: You don't post "facts" wadr - you post coloured lines. Here are Sutter's zone starts: 29.7% offensive zone starts 36.8 % defensive zone starts 33.5% neutral zone In other words, 43.3% offensive zone starts relative to defensive. What makes the context of Sutter's play - and his possession numbers more complicated - and contradictory - that most players is his dual role. On the one hand he centers a shutdown/matchup line. On the other, he's plays wing on the Sedin line. The Sedins enjoy some extremely opportune ozone starts - 60.5% (Hank) 57.8% (Daniel) Corsi Sutter 46.2 Hank 48.8 Daniel 47.4 Now, when you take Sutter's ozone starts when playing with the Sedins to be in the 58-60% range - but overall at 43.3%, what does that tell you about the context of his play when he's not playing with the Sedins. Whether with them - and facing opposition shutdown lines/pairings or without them and playing a shutdown role against the opposition's best, Sutter is facing strong quality of competition regardless - and always has, despite the fact that the context of his play changes significantly. Nevertheless, his corsi remains within 1.2% of Daniel, and 2.6% of Henrik. But, but, he's the worst center on the team, he's a horrible possession player, etc, etc. If people are satisfied with those kind of elementary and uninformed misconceptions about the player - based on extremely weak 'analytics' - that's your prerogative. Sutter often takes D-zone draws for other lines so it skews a bit his numbers. So does Horvat. They take the draw then get off the ice. Do you have the numbers for Horvat. 1 minute ago, oldnews said: That's a very misleading graph you've posted - from a playoff sample of 5 games. Here are his last 3 full seasons in Pittsburgh: 2014-2015 Line Combinations & Production Even Strength Line Combinations Freq Line Combination 32.5% DOWNIE,STEVE - SPALING,NICK - SUTTER,BRANDON 21.9% BENNETT,BEAU - SPALING,NICK - SUTTER,BRANDON 19.8% BENNETT,BEAU - DOWNIE,STEVE - SUTTER,BRANDON 14% COMEAU,BLAKE - MALKIN,EVGENI - SUTTER,BRANDON 11.7% DUPUIS,PASCAL - MALKIN,EVGENI - SUTTER,BRANDON 2013-2014 Line Combinations & Production Even Strength Line Combinations Freq Line Combination 27.3% ADAMS,CRAIG - GLASS,TANNER - SUTTER,BRANDON 22.1% CONNER,CHRIS - SUTTER,BRANDON - VITALE,JOE 18.7% GLASS,TANNER - MEGNA,JAYSON - SUTTER,BRANDON 18% GLASS,TANNER - PYATT,TAYLOR - SUTTER,BRANDON 13.8% CONNER,CHRIS - EBBETT,ANDREW - SUTTER,BRANDON 2012-2013 Line Combinations & Production Even Strength Line Combinations Freq Line Combination 65.7% COOKE,MATT - KENNEDY,TYLER - SUTTER,BRANDON 11.1% COOKE,MATT - MORROW,BRENDEN - SUTTER,BRANDON 8.4% COOKE,MATT - NEAL,JAMES - SUTTER,BRANDON 8% COOKE,MATT - GLASS,TANNER - SUTTER,BRANDON 6.8% COOKE,MATT - DUPUIS,PASCAL - SUTTER,BRANDON Thanks for correcting. I thought I had changed it. Link to comment Share on other sites More sharing options...
apollo Posted January 12, 2017 Share Posted January 12, 2017 12 minutes ago, coastal.view said: and part of his improvement is directly related to sutter being healthy and taking all the most difficult matchups (except when other team has last change and can avoid it ) so that horvat has less defensive responsibility and can focus his energies on other areas and burrows gets some credit too as when he joined that line he freed up bo even more and that is when bo began to become more of an offensive presence this year bo is not doing it by himself . . the coach has helped his development, burrows has, sutter's presence has changed his game, our improved defence has had an impact...etc You make some solid points as do many posters in this thread... Regardless, if it's game 7 SCF, and I have to pick 3 of our forwards to guarantee to play it would be Bo, Hank and Danny. I love Sutter but he's not top 3 yet, nor is he #2. Link to comment Share on other sites More sharing options...
Wise Owl Posted January 12, 2017 Share Posted January 12, 2017 28 minutes ago, WTG said: Fine, you can call them weak hypotheses, but they are the best non-anecdotal analysis of a player we got. We have to work with what we got. Look, I'm willing to concede that if you do not believe in statistics that's fine. But there are some statistics in there that are just facts. Like, we allow a bunch of unblocked shots in the slot when Sutter is on the ice vs when he's off the ice. When Sutter is on the ice, we only score 35.6% of the goals even strength. Reject stats like corsi or fenwick all you want, but there are just stats that are facts. Do these facts take into account that your job is to make sure McDavid, Getzlaff and Kopitar don't score from a faceoff in your own end, and that other players get to defend against less dangerous players and perhaps score so your team wins the game? Link to comment Share on other sites More sharing options...
WTG Posted January 12, 2017 Share Posted January 12, 2017 12 minutes ago, oldnews said: You don't post "facts" wadr - you post coloured lines. Here are Sutter's zone starts: 29.7% offensive zone starts 36.8 % defensive zone starts 33.5% neutral zone In other words, 43.3% offensive zone starts relative to defensive. What makes the context of Sutter's play - and his possession numbers more complicated - and contradictory - that most players is his dual role. On the one hand he centers a shutdown/matchup line. On the other, he's plays wing on the Sedin line. The Sedins enjoy some extremely opportune ozone starts - 60.5% (Hank) 57.8% (Daniel) Corsi Sutter 46.2 Hank 48.8 Daniel 47.4 Now, when you take Sutter's ozone starts when playing with the Sedins to be in the 58-60% range - but overall at 43.3%, what does that tell you about the context of his play when he's not playing with the Sedins. Whether with them - and facing opposition shutdown lines/pairings or without them and playing a shutdown role against the opposition's best, Sutter is facing strong quality of competition regardless - and always has, despite the fact that the context of his play changes significantly. Nevertheless, his corsi remains within 1.2% of Daniel, and 2.6% of Henrik. But, but, he's the worst center on the team, he's a horrible possession player, etc, etc. If people are satisfied with those kind of elementary and uninformed misconceptions about the player - based on extremely weak 'analytics' - that's your prerogative. https://puckplusplus.com/2015/01/15/how-much-do-zone-starts-matter-i-maybe-not-as-much-as-we-thought/ Give that a read. Also: Link to comment Share on other sites More sharing options...
oldnews Posted January 12, 2017 Share Posted January 12, 2017 4 minutes ago, mll said: Do you have the numbers for Horvat. Thanks for correcting. I thought I had changed it. Horvat: 29.2% ozone 34.2 dzone 36.6 neutral In other words, 46.1% ozone starts. Link to comment Share on other sites More sharing options...
oldnews Posted January 12, 2017 Share Posted January 12, 2017 1 minute ago, WTG said: https://puckplusplus.com/2015/01/15/how-much-do-zone-starts-matter-i-maybe-not-as-much-as-we-thought/ wadr, yawn. Link to comment Share on other sites More sharing options...
WTG Posted January 12, 2017 Share Posted January 12, 2017 Just now, oldnews said: wadr, yawn. Lol, seriously, stop trying you are going to lose. Statistics aren't on your side. Link to comment Share on other sites More sharing options...
aGENT Posted January 12, 2017 Share Posted January 12, 2017 1 minute ago, WTG said: Lol, seriously, stop trying you are going to lose. Statistics aren't on your side. 1 hour ago, SID.IS.SID.ME.IS.ME said: This is a great article for understanding just how poor "advanced stats" are in measuring individual performance and making year to year predictions for players: http://hockeyanalysis.com/2016/03/03/evaluating-player-evaluation-metrics-and-expected-goal-models/ Here's an excerpt: To summarize, Corsi can at best explain approximately 50% of a players offensive talent and somewhere around or a little less than 1 year (500 minutes) is where the sample size is large enough that GF60 is a more reliable measure of a players offensive performance And while goals data certainly provides more value than Corsi over samples >500 minutes, it still doesn't really give you good numbers until you get into multi-year samples. Similarly, the article shows that those "expected goals" models du jour don't really do better than Corsi. And the dirty little secret in analytics? None of the models work for defensemen. They only have predictive value, meagre as it is, with forwards. For defensemen, the "advanced stats" really don't tell us anything. http://hockeyanalysis.com/2011/07/12/how-i-evaluate-players-and-why/ http://thesportjournal.org/article/goal-based-metrics-better-than-shot-based-metrics-at-predicting-hockey-success/ So remember the above articles any time somebody tried to use Corsi to tell you how good or bad a dman is. Have to give a ton of credit here to David Johnson of Hockey Analysis (and the same guy who runs the invaluable resource stats.hockeyanalysis.com). He's been something of a rebel in the analytics community in taking people to task about their assumptions and consistently pointing out how crap "predictive analytics" are in hockey. Here's a great piece from a couple weeks ago: http://hockeyanalysis.com/2017/01/02/predictive-analyics-have-failed-hockey-analytics/ And, to move things back on topic a little, here's Johnson calling BS on the online analytics community saying Brandon Sutter is a "replacement level" defensive forward: http://hockeyanalysis.com/2016/07/15/the-brandon-sutter-litmus-test/ Johnson's take on Sutter: I am going to propose that claiming Brandon Sutter is a below replacement level defensive player be a litmus test for whether you are doing hockey analytics correctly or not. Cheeky stuff. Love it. Link to comment Share on other sites More sharing options...
WTG Posted January 12, 2017 Share Posted January 12, 2017 4 minutes ago, Wise Owl said: Do these facts take into account that your job is to make sure McDavid, Getzlaff and Kopitar don't score from a faceoff in your own end, and that other players get to defend against less dangerous players and perhaps score so your team wins the game? Sutter's QoC shows that he's not actually facing top players most of the time. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.