Schultz contract and analytics

Justin+Schultz+Edmonton+Oilers+v+San+Jose+8WY780gMOPql

The Oilers signed Justin Schultz to a one-year, $3,675 million deal. Craig MacTavish had very high praise for the 24-year-old from Kelowna.

“I think he’s going to be a great player and a great Oiler. I think he has Norris trophy potential,” said MacTavish.

Fans have been waiting for “bold” from MacTavish, and there is no debating that is a bold statement.

Personally, I think Schultz has a ways to go before he’s a Norris Trophy candidate, but I’d agree that Schultz has a lot of potential, most of it untapped thus far. I believe Schultz has better long-term potential than Jeff Petry, mainly because he has already excelled in one area, offensive production.

He’s only played 122 NHL games, so of course he can improve away from the puck and in his own end, but I don’t believe he’s as bad as his analytics suggest. MacTavish took that even further when discussing Schultz’s play in his own end.

“I disagree with the perception that he’s weaker on the defensive side of things,” said MacTavish.
” He showed me at the end of last year that he was managing the decisions
on when to go, when to probe offensively, when to get back. I know he’s
going to be a player that can be counted on in both ends,” continued the GM.

Schultz has been getting crushed by staunch supporters of analytics, and based on his stats you can see why statisticians would be concerned. The numbers at behindthenet.ca aren’t flattering for Schultz. 

After MacTavish’s high praise of Schultz I asked him his thoughts on Schultz’s analytics and how he compares them to what he sees when watching Schultz in the game.

“I think the analytics on players is very important, but I disagree with it on the way it assesses our group a lot of times. But, it does lead us to ask different questions than what we would normally ask of the tactics we deploy and the icetime we would give players. At the end of the day, I’m evaluating our team based on watching them play 65-70 times a year, and I’ve got a pretty good idea in my mind where the value is,” said MacTavish.

An interesting statement from the GM. He respects the stats, and likes that they bring up new questions and ideas, but I got the sense he still believes in what he sees from his spot in the pressbox.

I followed up with another question.

I asked MacTavish if he had dug into analytics more this summer and if there is anything he has seen that he likes, or that disagrees with his own evaluation of players.

“I think we are on the cutting edge of teams in the NHL that deploy these types of numbers. Without getting too far into it, I believe in the possession numbers. I think there are anomalies in individual player’s possession numbers, but collectively within your team you have to drive possession. It’s directly correlated to points in the standings. You have to be careful on an individual basis. It leads you to try and identify value in outliers or guys who may have unrecognized value,” MacTavish said.

This response seemed to be all about Benoit Pouliot. He had very good possesion numbers, although his QofC was never that high. He can be a very solid 3rd line player, but we’ve yet to see what he can do against top lines on a regular basis. We likely will see that at times this season.

It was interesting that MacTavish mentioned he liked how stats reflected well on a player like Pouliot (he didn’t say Pouliot specifically, but he fits in that category), but he disagreed how they portayed players like Schultz. Hmmmm

MacTavish finished our short conversation about analytics with this.

“I believe possession is a measure of your team’s collective ability to make plays with the puck and to me that is skill. So it is a measure of skill, it’s not a measure of toughness or character, but it is a measure of your team’s ability to make plays.”

I agree completely. It can illustrate one key aspect to helping you win, but it does not measure other intangibles that are also important.

I’ve always been suprised how much flack Schultz has recieved, despite his limited NHL experience. I still feel $3.675 is too much based on Schultz’s experience and overall game, but it is clear that MacTavish is a huge fan of Schultz and believes he will only improve.

Time will tell, but as a fan of offensive hockey and players who can move the puck, I can understand why MacTavish gave Schultz this contract.

Now it is up to Schultz to make his GM look smart.

Recently by Jason Gregor:  

  • Jordan McNugent-Hallkins

    Bit of an overpay, but for one year it isn’t going to sink the team. If he figures out the defensive zone and learns when to rush and pinch, he’ll be an excellent threat from the back end – something we haven’t had since the Rake.

  • ooj

    I personally couldn’t care if they paid him 2m or 4m this year.. I think after a season and a half he’ll start playing better individually in his own end especially with a more solid possession team as a whole around him…then we can probably appreciate him for what he is…a solid offensive dman.

  • 916oiler

    Norris Trophy potential???? WHOA there MacT. I’m hoping he lives up to it, but then again he won’t be providing a discount to the team in any way shape or form.

  • oilabroad

    I like MacT but I have a hard time with that statement about not buying the analytics on Schultz. If you only believe the analytics when they back up your preconceived ideas, what is the point??? That doesn’t sound very ‘cutting edge’ to me… If he hadn’t been signed in the fashion he was, he would be the one on the farm this year, not Klefbom.

    • A-Mc

      When i watched the presser, the message i got from what MacT was saying was that its hard to judge 1 guy too critically based on possession analytics when really possession numbers are a collective team stat. Possession is based on corsi, and corsi is based off shots for and against. If a couple guys are 100% doing their job but the other scrubs on the ice are failing, Shots wont be taken in the O-zone and they likely wont get blocked in the D-zone. You can only be accountable for your part of it, which is 1/12th of what’s happening on the ice.

      The entire team was garbage defensively, and the stats agree with that conclusion; but the team is not made up of garbage players. Some? yes. But not all.

    • '68 Fire Chicken

      Mac Ts statement was kind of dumb. I agree. But then you try saying if Schultz wasn’t signed the way he was Klefbom would be in the NHL? Schultz has already done more then Klefbom in the A.He has also done more in the NHL. If you were to trade one today Schultz would get back more,not a lot but more.

      He isn’t going to be a defensive specialist any time soon, if ever. That’s Kelfboms job and it takes longer to learn. Hence why Schultz is in the big show and Klefbom is still marinating.

      • oilabroad

        well Klefbom may be a stretch but the point I was trying to make is Justin Schultz is barely an NHL defenseman at this point. If there were conflicting stats on Schultz I might be more convinced by what MacT was implying, but quite honestly I haven’t seen any.

  • knee deep in it

    Once again Mac looselips is great with the “bold” statements but can’t seem to deliver the bold moves he promised. I like how analytics are well received when it supports a fact he likes but when it paints a bad picture they are dismissed. That’s going to work really well. I would think you either buy in or you don’t. If you just sit on the fence and cherry pick the message you want to hear you might as well not use them.

  • knee deep in it

    one important thing to remember is that analytics are useful to evaluate past performance. Future performance is based on current ability plus natural growth/improvements in players.

    It is reasonable to expect the young kids to outperform last years’ analytics even if all they do is get a year older. If they work hard on strength, conditional, skills, and mental factors, then next year will be much better. If mact expects that to happen, then he is fair in his statement “ut I disagree with it on the way it assesses our group”

    In regards to rapidly improving or declining players, past performance analytics has to be only a part of the evaluation.

  • knee deep in it

    I agree that Schultz needs to know when and when not to pinch, but it is also up to the other players on the ice to recognize when he is going as well. Getting caught rushing up ice is only half his fault most of the time, other times it’s his teammates missing the coverage boat.

  • BC BOY

    I would like to see what he can do with a better power play and a better d partner. Let’s face it every dman is gonna have bad possession numbers when paired with Andrew Ference or Nick Schultz. That said the oilers should of done a better job on finding a mentor for Jschultz when he came into the league. Finally if anyone can improve his overall game it would be Craig Ramsay. I expect Ramsay will have a huge impact on Jschultz this year and when needed will be able to give impartial opinion on him.

  • Given what the Montreal fans just went through with Subban, I’m not sure us fans are able to see a player in the correct lens.

    What I mean is Subban in Montreal seems to be a rather polarizing figure. Here in Edmonton we’d kill to have Subban, but not every fan was on board. And then it came time to give him his due and the organization bumbled it. I suppose point being the relation between the fans and a player clearly seems to be different than the one they have with the organization, especially when it comes to value.

    I liked this article because it brought up a good point about analytics. It seems that even though they promote a certain view on paper, Mac T clearly belies his experience is going to trump what analytics could be saying about a team or a player, and I agree with that.

  • '68 Fire Chicken

    does this signing hurt the oilers this year or next? are we at cap ceiling? is he still an rfa next year? would we be worse without him signed? does he still have to prove his worth to sign again next year?

  • I agree completely. It can illustrate one key aspect to helping you win, but it does not measure other intangibles that are also important.

    I keep waiting for someone to explain how these invisible traits help a team win if they don’t show up in the numbers. If a team doesn’t generate or prevent more chances because of a trait then the trait is meaningless.

    If a player’s “heart”, “grit”, or “leadership”, improves the team it WILL improve the numbers. How is this such a difficult concept?

    • oilabroad

      I agree with Tiger.

      Many of these “intangibles” are also “immeasureables”. In order for a character player to be making a real didfference, there should/ would be results to reflect it. (Better 3’s for him and/or linemates, better team result’s with him in lineup.) And if these results are shown, then they no longer become intangibles.

    • A-Mc

      Story Time

      The Oilers are down by 1. The captain pulls the guys together and says look, get your heads out of your arses and lets get this damn goal. The team pulls together, scores the goal and wins in the shootout.

      The analytics tell you the Oilers won 3-2 and that their possession numbers on the night were poor (Based on shots for/against) and the goalie kept it close. No where do they tell you anything about the captain and what he had to say to get his guys to perform for the remaining few minutes of the game. My personal belief is that because analytics are based on a larger period of time, it can’t tell you accurately what’s going on at any 1 time.

      If you broke down possession stats accurately on a shift by shift basis, you’d probably be able to isolate exactly what is happening after certain events take place. In this instance you’d see a shift in the green shortly after the captain rallied his troops; thus bridging the gap you’re inquiring about. From there you could then create a new stat where you’d record shifts after certain events based on character,grit, leadership, etc and formulate a picture of how these events are impacting the team

      • Toe-Dragger

        Lol at this ^^

        This literally doesn’t matter at all. If you have ever played any level of hockey and are down by 1, you will try hard to score. No matter what the captain tells you. This should be worth zero dollars to an NHL team.

        • A-Mc

          You mean like how Toews was losing his SH!T during the playoffs and Seabrook had to come settle him down in the penalty box? Ya a good talking to never does any good.. i concur.
          /sarcasm

      • From there you could then create a new stat where you’d record shifts after certain events based on character,grit, leadership, etc and formulate a picture of how these events are impacting the team

        This is entirely unnecessary because the events are already being recorded. Luckily we have the ability to know when events occurred as well as what the game situation was when they occurred. We also know about score effects. A captain who regularly inspires his team in “clutch” situations would be spotted quite easily because the team would regularly perform better “in the clutch” than they do at other times.

        The intimidation source is likely not on the ice at the same time, but is instead beaking from the bench. This is something analytics does not measure.

        Is he in a suit on the bench? If he is in the lineup and has any effect at all the bump in numbers will be there. Does the team play better with the goon dressed? Easy to see.

        • A-Mc

          I think you’re missing the point.

          I’m not saying that intangibles dont show in the numbers, because they do. I’m saying all you see are the end result numbers as a collection of things that lead to getting them. you don’t know what portion of the numbers are a direct result of the intangibles.

          In my crappy Intimidation scenario: the Analytics would show the performance benefit for Nuge because that is the result of the intimidation of the other team. The numbers dont show the actual Intimidation part of it. At the end of the night, you dont know if nuge just had a good game or if he played well because his teammates afforded him the time and space.

          Do you get what i’m saying?

          PS: I’m not trying to make a counter argument to what you’re saying. I’m trying to contribute to a discussion about how the effects of “intangibles” are hard to isolate with our current analytics system.

          • I get what you’re saying. I just think you are very wrong.

            There are many limitations to current analytics, but failing to account for invisible characteristics is not one of them, but you have to know how to use the numbers. You rcomment about only seeing “the end results” tells me you don’t know how to use them.

          • A-Mc

            Your comment about only seeing “the end results” tells me you don’t know how to use them.

            Please enlighten me by providing web resources so that i can understand what you’re talking about. I’ve never seen a resource that breaks down every shift and assigns positive/negative values for each as well as categorizes them based on events.

            NHL players all say Intimidation is a real part of the game. Question: How would you use analytics to measure or quantify Intimidation?

          • A-Mc

            If you could see in the analytics that a player was ineffective until leadership talked to him, then you could quantify an intangible. To do this you need shift by shift analytics.

          • 1. Shift by shift stats are already recorded. We know who did what on a given shift.

            2. We do not need to track by shift for this scenario because a leader who is actually capable of this would improve the results of his team.

          • Randaman

            Did you ever consider that there are people that think analytics are useless, boring and misleading when it comes to individual performance in a team sport??
            Play the game and win or lose! That’s the bottom line isn’t it?
            Just sayin…

          • Analytics are unnecessary. Many people dismiss them for that reason. Now I know the stats nerds are going to come screaming after this post but save it…your arguments that everything is measurable is stupid and does not account for the human element.

          • Serious Gord

            Stupid is as stupid does.

            In every meaningful human endeavour – from war to business to sports to gambling to online dating – statistical analysis improves ones success rate. And that analysis almost always involves quantifying and qualifying human elements.

            To believe that is not the case is to be a true Luddite. A throwback to the Middle Ages as they were.

          • Serious Gord

            Could it be that there is actual middle ground here? I am relatively new to this site but a lifelong Oiler fan. At the risk of sounding reasonable, is it all that crazy to believe that the two go hand in hand and should not exist solely without the other.

            All I have seen in the last few weeks following this site and comments is the mathletes trying to prove that analytics are the end all be all and the ‘i saw it good’ guys trying to decrease the overall value of analytics.

            I guess the back and forth provides for entertaining and at times interesting debate but anyone in a position of power will use both to get a complete picture and gain greater understanding to make informed decisions.

            Real progress is found in embracing both sides and utilizing it all to your full advantage

          • Serious Gord

            I very much think that currently they are both necessary. And that the eyeball method will never go away completely as the stats evolve and improve. And advanced stats don’t account for off-ice activities that may not impact past or current performance but will in the future.

            In the extreme – for example cattle breeding – stats are all that is needed. Yet every semen catalogue I have ever seen has a picture of the bull accompanying the statistical tables. I doubt very much a human endeavor like hockey will ever get that highly distilled down to just statistics.

            There are just too many variables to completely eliminate instinct and intuition.

          • There are no “mathletes” who believe analytics are the “end all be all”. This mythical “you should ONLY use stats” person does not exist.

            What you are seeing is people defending analytics against an inexplicable stream of “stats are useless and dumb”.

            In short, your position is exactly the same as that held by proponents of analytics. We agree that EVERY tool in the chest should be utilized. That is the entire point of collecting more data. The only people who think it has to be one or the other are those who have decided that they don’t like analytics.

    • A-Mc

      If intimidation is at play, and an opposing team is reluctant to be as aggressive with Nuge as they would otherwise. How do you measure that?

      Nuge would play better because he’d have more time and space, but then the numbers are reflected in his analytics eventhough the intimidation has nothing to do with him. How would the intimidation be accounted for in his teammates analytics?

      *shrug*

      The results show in the anayltics but what portion of them are directly a result of “Heart”, “Grit”, or “Leadership”

      • A-Mc

        The numbers are reflected in the fact that Nuge does better when said intimidator is in the lineup/on the ice, then when he is not. These are the EXACT kind of things that advanced stats can measure!

      • Serious Gord

        Heart, grit and leadership are all secondary attributes when evaluating a hockey player.

        There a likely several people who read the posts on this site who have more of those three things than many who play in the NHL.

        What they lack is skill, physical ability and talent.

        And as TUG points out, those secondary attributes will largely bleed through into the advanced stats. And as the field matures the next generation of stats will account for them even more and more accurately.

      • You certainly like to invent arguments to tackle. No, there is plenty of evidence that those things help a team win, and they do so by helping a team to improved results.

        What I am suggesting is that if a player with a ton of “heart”, for example, makes his team better then the team will have better results, ie. stats.

        If the results don’t change then, even if his “compete level” is off the charts, it did not make the team better.

        Take two players with similar skills. The one who works harder will have better results than the one who works less hard. On the other hand, a bad player who tries really hard is still a bad player even if he brings lots of “heart” and “grit”.

  • A-Mc

    I wonder how this signing will impact the likes or Torey Krug and Tyson Barrie. Barries’ stats are definitely better looking than Schultz’s. I’ve got to thinks that Barrie will now be looking for something over $4M for a similar term.

  • A-Mc

    Starting 7 DMen

    Fayne 26 … 242 NHL games, 16.16 TOI (4th)
    Nikitin 27 …206 NHL games 15.33 TOI (4th)

    Petry 26 … 236 NHL games 17.41 TOI (3rd)
    Marincin 22 … 40 NHL games 16.19 TOI (4th)

    Ference 35 ….831 NHL games 17.41 TOI (2nd)
    Shultz 24 ….122 NHL games 18.55 TOI (1st)

    Aulie 25 … 136 NHL games 9.24 TOI (297 in NHL) …. Fraser played 13.09 TOI

    (rank on last years team)

    I hope Ference drops to #6 and looks like we can upgrade Aulie in January with the best OKC Dman.

  • Quicksilver ballet

    Okay, so all we need now is a 2nd line center, a couple top pairing d’men and maybe even a goaltender.

    4.2 million left to accomplish this……no problem.

    • A-Mc

      Clearly if a couple top pairing D men are coming in from outside, then there will be a couple of D-men heading the other way (Petry and tbd). I maybe wrong, but I’d think Mac-T understands the math

    • Serious Gord

      And if something goes to pieces in net or on d or at center there is no cap room to do anything about it. Eakins and MacT should have gone to patriots camp – not Dallas’ they would have actually learned something useful.

  • Quicksilver ballet

    Be interesting to see which of the deals this summer wears the set of goat horns at seasons end. Least bang for their buck is probably the Nikitin deal. He could win back to back seasons. That’s a far worse deal than todays.

  • A-Mc

    people who only believe in advanced stats when it aligns with what they thought before hand, do not believe in it enough for it to make a difference. You have to trust it every time, and over the long haul you will win more than you lose.

    • Serious Gord

      Exactly. That is why Billy beane is one of the few who succeeds using them in the MLB – he trusts them. I bet he would be a killer horse handicapper.

        • Serious Gord

          By far and away the As are the best team for the dollar and have been ever since beane has been the GM. In a league without salary caps where top flight starting pitching is so expensive that is the true stat to use when evaluating success.

          • Who cares?

            One is a poor team that still wins a lot of games due to smart decisions. The other is the richest team in the league and the center of the hockey universe” who loses constantly due to bad decisions.

            You don’t think one of these options is better than the other?

            If you want people to take anything you say seriously you have to avoid taking wildly irrational stances on things like this just because the rational answer doesn’t help your cause.

          • Serious Gord

            Tony Gwynn never won a World Series and was the best hitter since ted Williams (who also never won a series).

            Sometimes the circumstances that prevent you from winning are not controllable by you. Beane has had to deal with having among the lowest salary limits in the league yet his team is perennially competitive and often leads the league. Dollar for dollar he is THE most successful GM in baseball.

            Getting to a World Series title correlates almost exactly with the capacity to spend lots of money over a very long period of time (it’s how the jays got their two titles and it’s a very big reason why they haven’t been inTo the postseason in over twenty years) I don’t have the stats in front of me but the richest teams have dominated the winners list going back to the days of Ruth. There are very rare exceptions to the rule – the marlins and diamondbacks in recent years – and perhaps the As will succeed someday, but not winning it all should not detract from the excellence of beane at his job.

        • Serious Gord

          cost-effectiveness. I bet he has more wins per dollar spent since he started than any other team during that same span. When you don’t have the budget that other teams do, it is your only option. Other teams with more money dont employ the same tactics as there is not as much need to do so, they can overspend and they get talent via the shotgun approach. Throw money at everything and see what sticks.

  • Serious Gord

    First a comment on advanced stats:

    A few months back I attended a seminar on the use of social media in marketing/advertising. One of the speakers compared the discussion around social media by businesses to that of kids in high school talking about sex:

    Everyone says they are doing it; few actually are and those who are, aren’t doing it very well.

    The same applies to advanced stats in hockey. I think MacT and to a lesser extent Eakins talk a good game about the use of stats, but when it gets right down to it they ignore them and go with their gut – using stats selectively when they support their decision. Managing that way is worse than having no stats guys at all as it provides cover for bad decisions and breaks the chain of accountability.

    On MacT: he’s an idiot when comes to being a GM.

    He talks like a player. He forgets that these high praises – grossly exaggerated and emphatic beyond reason (“… I know he’s going to be a player that can be counted on in both ends,”) weaken his ability to negotiate with the player at contract time. He needs to keep his mouth SHUT. But his ego is only a bit smaller than Klowes and thus he just can’t help himself.

    The oil are already chronic over payers for manifold reasons. MacT’s chronic reckless talk only makes it worse.

    And as for Shultz:

    he’s 24. Not 20 or 22. The next 24 months will be the most critical of his career. He either makes some significant breakthroughs – and makes this contract look like a steal – or he improves only slightly and thus is not only a bust but a loss of three years of development effort by the oil to make him a stud #1 d.

    I think he lacks the size/power, aggressiveness and defensive sense to become the former. But, that said, the oil have three years invested in him, he’s still unproven one way or the other, so another year and 3.65 to find out seems a bit high but ‘grimace’ is acceptable.

    • Serious Gord

      Listened a bit to lowetide’s show and his guest – Vollman (?) a master stats guy quietly disparaged the Shultz signing: “…if that’s the type of player the oilers are looking for…” Is a rough quote of one of the phrases he used.

      As an illiterate at advanced sports stats – yet a believer in the use of statistical analysis – such comments like yours and his (and presumably mr dellow’s) are cause for significant concern regarding how good Shultz actually is.

      • Zarny

        I’m privileged to know Rob Vollman and have a habit of listening to his interviews including the one you reference on Lowetide’s show. That is certainly a rough quote but more importantly it’s taken out of context.

        Lowetide asked Rob if he was surprised with the contract for Schultz (1y @ $3.675M). He said no, there is a market value for players like Schultz and that’s about it. He then added that the question to ask from an analytics perspective is do you want that type of player?

        What type of player is Schultz? Well, earlier in the show Rob compared him to Keith Yandle. A Dman who plays a lot of minutes but doesn’t do the heavy lifting defensively. He then added that no one questions the effectiveness of Keith Yandle.

        So I think the answer is yes, you want the type of player that Schultz is on your team. The key is using him properly which should be anyone’s conclusion by eye or by number.

        Interestingly enough, when asked if the Oilers have a top pairing Dman Rob said yes, they have two. Fayne and Petry.

  • Kr55

    The main issue with Schultz is that almost every Oiler had better possession numbers without him on the ice. Couple that with how Schultz didn’t play against top competition and it’s concerning. He really has a lot to prove next season, if he can be more than a d-man playing 4th forward that is pretty poor at getting the puck out of our zone.

  • Zarny

    I think it’s obvious that MacT is being diplomatic. If he really disagreed with the analytics he would have given Schultz 7 yr @ $5M+. He also has to negotiate with the player again 10 months from now.

    What he is talking about is context. Why did the analytics portray that? With Schultz, it was clear by eye or by number the problem was usage. When you are part of the decision to play Schultz above his head I suspect you are more apt to see “anomalies”.

    It’s a mistake to say analytics, specifically possession, is the be all end all and the only thing that matters. There is also goaltending, shooting, special teams, etc. There is no one element of the game that is the magic bullet.

    Possession also isn’t a new concept. People have always known possession is key. To win games you have to score goals; to score goals you have to have the puck. What’s new is we now have a way of counting and measuring possession.

    It doesn’t make everything else obsolete; it’s one more tool in the tool-belt to analyze players, teams and situations. Often it confirms what the eye sees. When it doesn’t you probe deeper to ask why?

  • Zarny

    I have a question about possession. When it relates to a player such as Hemsky who was often criticized for not being decisive enough with a puck, isn’t possession sometimes a negative. For example, Hemsky would often hold onto the puck for far too long and squander a chance. Now let’s speak about Ryan Smyth, maybe not in his later years but when he was a legitimate difference maker. He was never a strong possession player, he often got rid of the puck immediately such as by deflections and hard nosed plays around the net.

    If we are strictly looking at analytics wouldn’t there be flaws with possession numbers when it relates to players who hog the puck and squander chances (good possession players) and players that are very decisive and direct the puck to the net on a whim (poor possession players).

    • I think you have a bit of a misconception regarding possession numbers. They have little to do with a player carrying the puck around a lot.

      A player who makes good decisions with the puck without having it on his stick a lot could easily be an excellent possession player because good decisions generally result in possession.

      The idea that players who “hog the puck and squander chances” are deemed good possession players and “players that are very decisive and direct the puck to the net on a whim” are bad possession players has little to do with the reality of the numbers.

    • Kr55

      How possession is typically measured is by counting shots directed at the net, not an actual tracking of where the puck is at every given moment. So, the typical complaints about Hemsky (holds the puck too long, doesn’t shoot, gives the puck away too easily) are actually all things that would translate to poor possession stats. Hemsky actually has good possession stats because, despite the obvious mistakes he makes that people like to jump all over, a lot of his plays actually do result in chances and also he is good at recovering pucks to prevent the other team from getting shot attempts on us.

      In Schultz’s case, I think we do generate a decent about of shot attempts when he’s out there, but the problem is we get a heck of a lot more against us when he’s in the ice. The risk of having him out there far outweighs the reward at this point of his career. And the stats do show that the majority of his team mates are better possession players without him on the ice which illustrates that Schultz can be a negative factor out there. He has a lot of improving to do in his own end, which it probably completely obvious to anyone that has watched him the last year but the possession stats definitely back it up.