This is a debate between RAIF and PJ, please don't intervene.

You are allowed to answer him; however, please do so concisely (i.e. in a single reply).

Threads like these are always tough, because they usually get heated and result in a headache for the moderation team, but this one has been a pleasant surprise so far, thanks guys, keep it up (civil) ?

Banks, please don't interrupt. We're having a debate here.

Well i looked at HERO charts to figure this out ... they show that Klefbom is a great player however not a top 10 defenseman in the league ... also, as much as WAR tells a lot, i really think his career +/- and his team's SV% when on the ice add to that and say that as much as he's very good, not a top 10 d-man. McDonagh, for example, had a .933 even strength team SV% when on the ice while Lundqvist's was .918. Also, doesn't Klefbom play with Larsson? McDOnagh played with Girardi most of the year, which is not to say Girardi isn't a good player, but he's not at the same level as Larsson, which also says that McDonagh is helping out his team more defensively than Klefbom is ... whether McDonagh's better than Klefbom or not, there are at least 10 defensemen better than Klefbom.

+/- is one of the worst indicators of skill set. My question that I want you to consider, is that if Klefbom is such a severe SV% killer, why is that not reflected in the correlating WAR rankings?

Klefbom's Defensive Point Shares Is 4.3. McDonagh's Defensive Point Shares is 4.8. Their total Point Shares are 7.6 and 7.8, respectfully.

In this visual aid (http://hockeyviz.com/fixedImg/wowy/1617/NYR/mcdonry89/wrap) comparing to this one (http://hockeyviz.com/fixedImg/wowy/1617/EDM/klefbos93/wrap), the amount of players who play worse without Klefbom is higher than those who play worse without McDonagh.

If, by your own admission, Ryan McDonagh is better than Oscar Klefbom, why do the visual aids, HERO, WAR, etc., not correlate with your statement? Klefbom is either equal or better than McDonagh in HERO ratings, and if HERO ratings are the foundation of your contention, why was McDonagh listed as better than Klefbom in an earlier admission?

McDonagh's PDO is measured on 5v5, while Lundquivst's SV% is measured including all scenarios, hence the substantial difference in McDonagh's 5v5 SV% compared to Lundquvist's SV%.

Forget about McDonagh, isn't WAR more about shot blocks, shots, takeaways, etc? Because Klefbom is also not as much of a puck-mover than a lot of defensemen, and remember once again, his good WAR could be partly from playing with Larsson, who is coming off a really good season.

WAR measures individual value, while taking into account virtually every analytics available. Larsson has very little to do with how Klefbom's WAR is affected, seeing as how the metric incorporates every quantitative metric possible, and compacts them into one metric, WAR or Wins Above Replacement, based on individual value to one's team.

We cannot forgo McDonagh. You made the claim that he is better than Klefbom, primarily due to his HERO rating, while Klefbom has a better HERO rating than McDonagh. If other claims are fraudulent like that of McDonagh's, which a claim you made based on stats that I have countered with reasoning to them, what other claims do you have that are fraudulent?

You have yet to coherently rebuttal my most recent claims, specifically my 4th bullet point in my previous comment. "If, by your own admission, Ryan McDonagh is better than Oscar Klefbom, why do the visual aids, HERO, WAR, etc., not correlate with your statement? Klefbom is either equal or better than McDonagh in HERO ratings, and if HERO ratings are the foundation of your contention, why was McDonagh listed as better than Klefbom in an earlier admission?"

I said forget about McDonagh, this is nothing to do with him ... but Adam Larsson's is .923 which is better than Klefbom's. So it seems to be that Talbot is better when Klefbom isn't on the ice than he is when Klefbom is on the ice ... and i'm not saying that's true, but then look at some other stats. Klefbom is also top 40 in giveaways ... now i know the Norris winner was 1st but this tells something ... his points were pretty good but not super good ... his shooting percentage isn't great ... and his expected +/- is above his +/- which i believe also is not great? So there are a lot of stats that say he's very good, but also a lot that show he's not great ... i think he's a great player, but i don't think stats say he's a top 10 D in the league.

McDonagh is relavant because you claimed that he is better than Klefbom, with the tranche of your support coming from HERO ratings. I have disproven that McDonagh is better than Klefbom in HERO ratings, yet you still have not directly addressed this point.

+/- is the worst indicator of talent. Claude Giiroux's +/- is (-15). Jake Guentzel has a +/- of (+7). Claude Giroux is substantially better than Jake Guentzel.

I will ask you to answer this question directly the next time you respond. If Klefbom is a supposed liability compared to Adam Larsson, why is that not reflected in the WAR rankings?

Just going to say this.

Ovechkin had a -29 5 years ago despite scoring 50 goals

Shhhh

Whether McDonagh is better or not, take a look at this list: https://www.nhl.com/news/drew-doughty-nhl-network-top-defenseman/c-281334346

Whether McDonagh is better or not, there are a lot of very good NHL defensemen right now. If you look at all the stats of these players, there are so many good ones ... let's take another guy: Duncan Keith. His HERO chart is better, his team SV% when on the ice is better, he has a better CF%, and a better +/- with a lower expected +/- which i believe is better. And then the list of players goes on and on ... another one: Drew Doughty. He had a better team SV% when on the ice, a better HERO chart, a better +/-, etc. You're looking more at one stat.

Also, Klefbom's never got a top 5 norris trophy vote ... ever. even Beauchimen has.

That list is invalid, seeing as how it was published about a year ago.

Ducan Keith and Oscar Klefbom basically have the same HERO, except that Klefbom plays two percentile points lower than Keith in TOI.

Your fixation of +/- constantly reassures me that you have not been able to comprehend my reasoning to why it is not a good indicator of talent, and multiple other professional advanced analytics folks agree as well.

You have yet to directly respond to my question "f Klefbom is a supposed liability compared to Adam Larsson, why is that not reflected in the WAR rankings?" and "If, by your own admission, Ryan McDonagh is better than Oscar Klefbom, why do the visual aids, HERO, WAR, etc., not correlate with your statement? Klefbom is either equal or better than McDonagh in HERO ratings, and if HERO ratings are the foundation of your contention, why was McDonagh listed as better than Klefbom in an earlier admission?""

You have shown nothing that inclines be to believe that you have looked at the graphs I have provided. You have time and time again failed to directly answer my question(s). I would ask upon the jury to hold Rangersandislesfan in contempt if he fails to address any of my points that I have made.

Norris Trophy voting means nothing. Lebron James doesn't win the MVP every season, but he is always the best player in the NBA.

Please address the questions that I have imposed and the graphs that I have provided a few pages back. If you don't, I will ask the jury to dismiss you of the debate, granting me the victor.

It's cuz I haven't intervened yet.

1st question: Because WAR doesn't tell everything.
2nd question: That's fair, maybe Klefbom is better than McDonagh, but that doesn't matter, there are still at least 10 d-men better than Klefbom.

And about the Norris trophy voting, i'm saying he's never received a single Norris trophy vote at all. It's not because he's never won it, he's never even received a vote. And about the list being a year ago, here's the thing: that list of the top 20 defensemen from a year ago not having him on it means he must have had a really good season to be considered a top 10 d-man in the league ... and not even one person voted him in the top 5 for this year ... one year doesn't tell everything ... for example, if he was the 5th best d-man this year, that makes him a top 10 d-man in the league but not top 5 if he wasn't top 20 a year ago ... 21 defensemen received one or more votes this year, including Orlov, Shattenkirk, etc. If 21 d-men received at least one vote and he didn't, i don't see how he'd be top 10 in the league if this list a year ago didn't have him top 20 either. If i made a top 10 defensemen in the league list, 8 of them received at least one Norris vote this year and the other 2 were on the list from last year.

But wouldn't you concur that the WAR and how one performs on the stat sheet, correlate? You claim that they don't, and the answer of "Because WAR doesn't tell everything" is not a sufficient answer to my question.

You claim that there are 10 better NHL defensemen better than McDonagh, but you still haven't been able to make a strong case, other than points that I have countered, which you have not countered.

Norris Trophy is given to the best NHL defensemen. Klefbom isn't a top 3 NHL defensemen, so why would someone vote him as one? By your logic, every single defensemen could receive a vote, other than Klefbom, and you would consider them to be better than him, because of where biased analysts rank them.

Jesus Christ, it's hard not to jump in.

You're telling me haha.

PJ and RAIF, please make a closing argument (in ONE post), as to why Oscar Klefbom is or is not a top 10 defenseman. This should summarize your previous argument in a coherent, concise manner.

Voters: is this my last post about this then?

phillyjabroni: Here's the thing: 21 defensemen received at least one Norris trophy vote in the top 5, not 5 ... there are so many different opinions so all the people voted differently ... also, about the WAR, why do other stats look different then? Klefbom is better than a lot of these guys in WAR, but not in other stats, so they're definitely different things ... so Klefbom is better in some stats, and these other guys are better in other stats. In the end, i think the fact that Klefbom, unlike 21 other defensemen, not 9, did not receive a single norris trophy vote, tells the answer, especially when that list from last year didn't put him in the top 20 ... even though that was that year, he would have had to have a GREAT year to be a top 10 d-man in the league, and still didn't even get a norris trophy vote. If the stats are even, that's what i think we'll look at.

The Closing Statement by phillyjabonri

WAR is derived from stats such as shot suppression, shot blocks, point shares, etc., basically any analytic that evaluates an individual's value. What this means is that the stats that you claim have nothing to do with WAR, actually do, thus making them correlated.

Norris Trophy voting has little to no value comparatively to that of WAR, HERO, etc. Simply because one voter feels that Dmitry Orlov is better than Oscar Klefbom, doesn't conclusively determine that Orlov is better than Kelfbom. If that was the case, then you would theoretically support if Andrew MacDonald received a vote, which would be more than Oscar Klefbom, you would make the contention that MacDonald is better, based on the foundation that an analyst ranks him higher.

As sited in this Google Doc (https://docs.google.com/spreadsheets/d/1U5gF9iypiqQ4sfNiwI0XEpaWRCYiSQ9ri5cQqUIwKxg/edit?usp=sharing), I evaluate that Klefbom has a better HERO subsection/TOI than most players you claimed to be better than him, with HERO ratings as your primary tranche.

As sited in these graphs (http://hockeyviz.com/player/klefbos93/EDM/1617), Oscar Klefbom enhances his teammates play, generates more shots 5v5 compared to when he is not on the ice.

Jury, I ask you to vote based on who's side you were persuaded to more. The defendant, Rangersandislesfan, hasn't presented a case that Klefbom isn't a top ten defensemen, rather he has presented that there are at least ten defensemen better, strictly based on Norris voting and analytics that I have contented to be inaccurate. I ask if you cannot conclusively determine that Klefbom isn't a top ten NHL defensemen, then you must vote that he is.

Please rid yourselves of any bias that you have towards Oscar Klefbom, Rangersandislesfan, or myself before voting. Please examine the debate in its entirety before reaching a verdict.

thank you - phillyjabroni

Thanks for your concluding statements. This debate is now over. The jury will be discussing here who they feel made the better argument.

Other jury dudes, whatcha think about this fiasco

I say he automatically loses for spelling his name wrong

Actually relatively happy it wasn't a fiasco. Only took me a small bit of time to review all the arguments.

What impresses me is that very few stats arguments were used properly.
- +/- has been proven by multiple generally appreciated analysts. While I did not like the method of choice phillyjabroni used to demonstrate that +/- is not a good stat (as picking and choosing in any circumstance normally can prove just about anything), his point is well taken.
- HERO was used in this arguement as HERO/TOI, which is not the appropriate way to evaluate it. HERO uses rate stats, dividing by per hour rather than per game. What this means is that dividing by TOI is completely unecessary.
- oZS%, while helpful in showing what situations a coach is deploying his players, often doesn't have a demonstrable influence on the results a player produces.
- Neither contender used information like power play or penalty kill statistics in making their argument, which I would say have a noticeable effect on HERO, WAR, and many other values. A top 10 defenseman would be effective in at least one of those situations, no?

In evaluating the qualitative arguments, I think it was dismissive of phillyjabroni to immediately disregard leadership and educated community voting of awards. While we cannot quantify the effects of leadership, we can say that they are indeed a positive trait to have, and so they must be at least marginally considered. Furthermore, awards voting is done by an educated community (PWHA). While they are not perfect, they watch and follow hockey significantly more than the vast majority of the population; their collective opinion is a significant sample and cannot be immediately discarded.

That being said, I evaluate debates not only on the content provided, but the ability for the user to contextualize and make relevant, along with rebutting other arguments.
Content provided: phillyjabroni
Contextualize: phillyjabroni
Rebutting arguments: Neither? They both didn't do a particularly good job at it.

I would give phillyjabroni the win on the debate.