Six Degrees of Separation

Earlier this week, I was listening to NHL radio on XM/Sirius. They’re running old playoff series during the summer, and this particular day it was G1 SCF 2006. I turned the radio so quickly it startled the passengers. I don’t know that I’ll ever be able to watch or listen to G1 SCF 2006. However, there are some things to be learned from that hockey club.

The 2006 Edmonton Oilers had a lot of very good players before the trade deadline. Pronger and Peca arrived in summer, and they had an immediate impact on the team. The deadline (adding Roloson, Spacek, Tarnstrom, Samsonov) marked Oiler management’s high water mark for the decade and we all enjoyed that wonderful team and their terrific run to the Stanley cup finals. Many look back on that team’s regular season (barely won 8th in the conference) and suggest it was a weak team that got hot at the right time. Rubbish. It was an outstanding team in search of goaltending until the deadline. Anyone who watched that spring knew this team could win a playoff round or more.

It would be a mistake to give too much credit to the hired guns that season. The 2005-06 Edmonton Oilers were blessed with 6 stunning "value contracts" and their presence gave the team exceptional depth (especially up front) and more than one line capable of outscoring the best opposition. Here are the players, their contracts and their accomplishments that season (in alpha order):

  • Marc Andre Bergeron ($931k). Bergeron played 1600 minutes in the 05-06 season, 350 of that on the powerplay. He delivered 2.74/60 with the man advantage but was pretty solid at EVs (1/60) and his 15-20-35 for the season was exceptional for the price.
  • Ales Hemsky ($901,740). Hemsky played 1375 minutes in the 05-06 season, slightly over 400 of them on the powerplay. His PP/60 number (6.17) was very nice, his EV number 2.25 was a little better than he managed this past season (2.09); that PP number helped him lead the team in scoring (19-58-77) and he delivered 6-11-17 in the playoffs too. A wonderful payoff for less than a million, a season to remember.
  • Shawn Horcoff ($1M). Horcoff played almost 1600 minutes, almost 300 on the PP and 225 on the PK. In 05-06 he went 3-3-6 on the PK (about 1.6/60), went 4.82/60 on the PP and then 2.44 at EVs and this was against the other team’s good players. A very underrated season when all was said and done.
  • Fernando Pisani ($611,800). In 05-06 he was excellent in the regular year and ridiculous in the postseason. Pisani played 1100 minutes in 05-06, 150 on the PP. He did a lot of heavy lifting at EVs and still managed to score 1.84/60 and 3.59 on the PP. Pisani was Guy Lafleur in the post season, 14-4-18 in a run I will never forget. At $611,800 he was ridiculous value.
  • Jarret Stoll ($501,600). Stoll played 1500 minutes in 05-06, 410 on the PP and 200 on the PK. He was a pretty valuable hockey player. On the PP he was 4.53 and at EVs he was 2.35 on the way to 82gp, 22-46-68 totals. At the price, he was extremely valuable.
  • Raffi Torres ($875,000). Raffi played 1100 minutes in 05-06, 224 of them on the PP. He’s pretty famous for wandering out there but his results have always been solid. His EV number in 05-06 (2.07) and his PP number (2.95) were very good considering he spent little time on the club’s top line or #1 PP. Torres’ biggest moments in the season came during the playoffs when he made some massive hits (one of which had an impact on the SJS series). At this price, he was a bargain.

Taylor Hall is a huge part of the Edmonton Oilers future, but his entry level contract (with bonuses) is $3.75 million dollars a year. Since the club will no doubt start burning that entry level deal this fall, chances are that Hall won’t over-deliver on that contract based on expectations for a teenager in the NHL. A young man like Tyler Pitlick–should he stay in junior and sign at age 20–has an excellent chance of being a "value contract" at a time when the Oilers should be pushing for a deep run into the playoffs (2012, 2013, 2014).

Which players could deliver the most value this season? Colin Fraser (.825M); Smid (1.3M); Dubnyk (.800M). RFA’s Sam Gagner, Andrew Cogliano and Gilbert Brule could also deliver more than their contracts, but we don’t know their cap number yet. The rookies mentioned here are unlikely to outperform their cap hits as rookies, but in year two and three it is entirely possible.

With Pääjärvi (1.525M), Eberle (1.158M) and Linus Omark (.875M) joining Hall as possible rookies, years two and three of their entry-level contracts offer a real opportunity for creating a list similar to the 2006 group above (although Hall will have his work cut out for him). Add Vande Velde, Peckham, Lander, Plante, Petry and a host of young pro level kids, plus Pitlick, Marincin and a few other quality picks (should they develop) and this team should have a nice group of value contracts in the next few seasons. We can only hope for a group as strong as the 2006 six.

This is the most important area for the team. Big name free agents, high cost offer sheets, even #1 overall picks bring their own cap hits and a team must pay 100 cents on the dollar. Those value contracts–like the ones we saw in 2006–will allow the Edmonton Oilers some separation from the rest of the NW division.

  • OB1 Team Yakopov - F.S.T.N.F

    If Ales Hemsky has a healthy season, 82points or more could put his contract in the “value” category.

    Then again it’s doubtful since he misses as many games as Souray the last few years.

      • 1st, Hemsky has a good contract.

        I’m suggesting that being a “value” contract would entail exceeding expectations. Hemsky should be a point a game player if healthy…

        @ 4.25 it doesn’t exactly sneak up on you, it’s still costly.

        • You are suggesting that to be a value contract at 4.25 he needs to score 82 points.

          If he is not value then please tell me how many guys making his money outperform him. Hemsky may not be exceeding YOUR expectations but he is exceeding the value of his contract.

          • Correct. If he were on the open market, he’d be in the $5.5M range… for what he already DOES. A point/game at 5.5 would be the expectation. SO… if he puts up a point/game, he’d be providing services at $1.25M cheaper than market value, OR… a “good value” contract.

            In other words, its (relatively) easy for Hemsky to outperform his current $4.25M contract. It’d be (relatively) that much harder for him for him to outperform a $5.5M contract (without him taking a big step passed the 90+ point barrier).

          • rindog

            You’re both exactly right, BUT…

            Here’s some hockey economics in my opinion..

            Missing 96 games x 4 years, that’s 4.975 …

            Hemsky is essentially a 5.49 player based on when he’s really IN the lineup..

            I don’t wanna argue this because it’s the best angle I could come up with..Hemsky’s contract isn’t a problem, it’s good…

  • So it sounds like Todd Nelson will be the Head Coach Of Ok City. As per twitter. Hmmmmm. Best option? I don’t know enough about him. Stats so no IMO but maybe he is ready.

  • OB1 Team Yakopov - F.S.T.N.F

    Sometimes I get the feeling Oiler fans would prefer a 20th place team with the lowest payroll in the league over a 10th place team with the highest payroll in the league.

    • Ender

      While your point (I assume it’s that 10th is always better than 20th) has some merit, I think to a lot of people it’s about hope for the future.

      Let’s take your scenario to the extreme. What if this year, you could build a team that was chock-full of expensive contracts, was pretty much guarenteed to finish in the playoffs (say 5th in the conference using your analogy), and was also pretty much guarenteed to need to be completely rebuilt from scratch in the off-season with no decent picks or prospects in the system (because you traded them all away to get the expensive players for this year).


      You could build a team that was full of picks and blue-chip prospects that was close to the cap floor that you knew you could keep together for at least the full length of all their entry-level contracts, several years the way you had it structured. This team would probably only be good for 10th place in the conference this year, finishing out of the playoffs.

      In scenario one, you better get it right, because if you miss the cup with that one shot, you are out of the race for a LOT of years. (Hello Calgary Flames)

      In scenario two, you are pretty much sacrificing this years playoffs for a chance to be in the playoffs for the next several years and maybe a serious and powerful contender for a couple of those years. (Hello Edmonton Oilers)

      I know which team I’d rather be on.

  • I thought we agreed never to speak of that terrible game ever again.

    And if anyone is to blame it’s MacT for dressing Ty Conlkin instead of Markannen* as the back-up. MacT cost us the Cup.

    *I already forgot how to spell his name

  • On the Bergeron hit on Ladd that took Roli out…. it’s easy to say it’s a stupid play on Bergeron’s part. Hockey unfortunately is a game that comes down to split second decisions. Ladd beat Greene, and Bergeron saw that Ladd was about to have a pretty damn good opportunity on Roloson.

    Some say that it was dumb of him to throw the hit, but what is he going to do, let him skate in cleanly? Obviously his intention was to body Ladd away from Roloson completely, but momentum didn’t quite allow that. Bergeron isn’t going to stop and think “hmm, Ladd is travelling at X Km/h at xxx angle, and I’m travelling at y km/h at yyy angle, I wonder where he’ll end up when I hit him…”. No. Like anyone, his thought would have been “holy f—ing s—, Ladd’s in all alone”, and he did all he could do to stop a goal.

    The alternative to the play… watch Ladd go in all alone. Roli would have made the save cause Ladd lost the puck, but Bergeron didn’t know that. If he does that though, he looks just as bad because he’s now doing nothing to stop a guy from getting in clear on Roloson.

    Honestly, he’s damned if he does, damned if he doesn’t on that play. It’s not a play where he had enough time to think of a safe way out. Arguing that he lost the Cup for us on that play is incredibly narrow-minded.

  • Since we’re off the rails anyways, I just headed over to capgeek to fill out the Oilers roster (I’ve got 2 extra F, and an extra D above the 23-man roster). Here’s what I came up with…

    Taylor Hall ($3.750m) / Dustin Penner ($4.250m) / Ales Hemsky ($4.100m)
    Magnus Pääjärvi ($1.525m) / * Sam Gagner ($1.750m) / * Gilbert Brule ($1.975m)
    * Jochen Hecht ($3.525m) / Shawn Horcoff ($5.500m) / * Mike Comrie ($1.975m)
    Ryan Jones ($0.975m) / Colin Fraser ($0.825m) / Zack Stortini ($0.700m)
    Jean-Francois Jacques ($0.615m) / Steve MacIntyre ($0.500m)
    Jordan Eberle ($1.158m) / * Andrew Cogliano ($0.975m)
    Ryan Whitney ($4.000m) / Tom Gilbert ($4.000m)
    Ladislav Smid ($1.300m) / Kurtis Foster ($1.800m)
    Jim Vandermeer ($2.300m) / * Theo Peckham ($0.725m)
    Jason Strudwick ($0.725m) / Shawn Belle ($0.600m)
    Nikolai Khabibulin ($3.750m) / Devan Dubnyk ($0.800m)

    ROSTER: 26; CAP:$59.4m; CARRY-OVER PENALTY: $0.354m;
    PAYROLL: $54.370m; CAP ROOM: $8.818m; BONUSES: $3.788m

  • Lowetide

    Actually, Hecht had a nice RelCorsi (2nd on the team among regular forwards) while facing tough opposition (and he played with good players).

    He was pretty good at 5×5/60 mins (2.02) and his Offensive zone start was 51%. He’s still a good hockey player.

    Good call, Ross Creek.

    • Better than Sourays. Souray for Hecht anyday!
      Good call RC. The question I think is not if the Oilers are interested, rather if the Sabres are. I think they will try for Kaberle first. If we are lucky Souray second.

        • Like I said lucky if Souray was second. Beiksa will go before Souray for sure. With the cap room I could see something with Chicago to help them out cap wise. Just a theory, but what do you think of Souray and younger cheaper need in CHI. (not sure if they’d want Cogs) for say Campbell’s anchor of a contract and say maybe Bolland. Is there room to work this in to some kind of package?

          • Can’t see the Hawks moving Bolland or Sharp. Maybe they’d move Skille or Beach in order to move Campbell, but I can’t see more than a couple teams having interest in his contract… maybe Atlanta (since the Hawks have done them so many favors) or NYI (just to get to the cap floor). And Campbell has a NTC, so chances are he’s not willing to waive for either of those destinations.

  • DSF

    Once again the point is missed.

    Lowetide and many other Oiler observers assess the Oilers’ future as if they are playing a vacuum. They aren’t.

    It’s likely the Oilers will need to finish first or a second in the NW division to even make the playoffs never mind be successful.

    Virtually every team in The WC has a much better group of veteran talent but many also have the “value” contracts being discussed here.

    You have to be successful in both areas to win.

  • OB1 Team Yakopov - F.S.T.N.F

    Buffalo won’t trade within the division, and probably won’t trade a legit prospect that TO is looking for in exg for Kaberle.
    Buffalo probably won’t deal with us either because of the “Vanek incident”.
    Making a deal for a D with Van for Bieksa makes more sense.

    No no friends, if we want to deal Souray we’ll have to get ugly.
    To NJ for Zubrus (4 more years $3.88M avg) or
    To CBJ for Huselius (2 more years at $4.75).
    … if we’re lucky. Yikes.

      • Nice thanks man. Yeah I read the stats and agree with them until I read about a player I think sucks and the stats show different….Weird hey.

        I’m glad Lowetide is now part of the Nation he seems to put lots of effort into getting us an article everyday.i checked out his site but find it harder to browse than the Nation. Good work Lowetide

          • I couldn’t agree more. The advanced stats are somewhat irrelevant if not outright delusive.

            Nothing like the school of duck. If it looks like, sounds like, and walks like… it’s a duck. Hockey players and there roles and importance are only calculated through observation.

          • I agree. Stats can give you a quick glance at someone, but there is no substitute for good old fashioned watching games.

            Some people get far too wrapped up in players stats, whether it be the old box cars, or the more advanced stats. As much as I dislike when people quote stats like they’re the law though, they can help support an argument, to some extent. But generally people overanalyze stats and often twist them to completely explain an argument, rather than use them sparingly to help support an argument (I do like though if someone uses them briefly to support an argument more so than driving their whole argument with stats).

            OB1 – Team Hall wrote:

            Kind of strange that real live NHL teams use them then. I guess the anti-math crew on the net know more about evaluating talent though.

            Are you trying to tell us all that if say the Penguins were looking for a left winger, they would scroll through the stats and then pick their guy? I’m afraid not. They use the stats as a quick glance, and truly pick the right player by evaluating video tape as well as watching them in person. Stats aren’t useless, but they’re only one part of it. Watching the games is a far more important part of evaluating talent.

          • Sorry this whole stats thing has been a bit of a sore spot for me lately. I do agree that teams use advance stats (and all forms of stats, really). I think some people are just arguing that there is no substitute for watching a game. Stats can tell you a little bit about a player, but you really have to watch them to see the whole picture.

            But I’m definitely not one who believes stats are useless. I just believe they are less important than some people make them to be. Or perhaps I should they they’re just to widely open to interpretation for the average person to use them to tell the whole picture well enough. As I mentioned in the post you quoted, I think stats should be used to add a bit of support to an argument, but they shouldn’t be the whole basis (which unfortunately, many people too often use them in that way).

          • What percentage of NHL players/NHL prospects have you seen play live? How many of them have you seen play live more than once? More than twice?

            I would be willing to be a very large sum of money that a more complete picture of a players potential or ability can be had through statistics than through seeing them play 3 or 4 times.

            Watching a player tells us what he seems like he should be able to do, but stats are able to quantify the results of those observations. In short, watching tells us who should be batter, but stats can verify which player actually has better results.

            I would rather have both tools in my belt, but if I have to choose I’d rather know the practical effects a player has on the game than simply a vantage point to observe which skills a player could potentially choose to utilize.

            Here is a small challenge for you. Tell us who is your favorite player and explain why using no stats.

          • So you’re asking me to use no stats when I’ve said that stats should be used to add a bit of support to an argument? Like I said, stats aren’t useless, so I shouldn’t have to argue without stats, but I will.

            Hemsky is my favorite player because I’ve seen him with my own eyes go into corners fearlessly for pucks. As well, his stickhandling and vision are superior. I’ve seen some pretty incredible highlight reel dekes from him as well as numerous seeing-eye passes. That makes him a pretty tremendously gifted playmaker as well as a danger as a scorer.

            No stats 😉

            As far as seeing players “live”… I’m not sure if you mean in person or on TV, but I believe that seeing them on TV is all you need to do. Anyways, I watch as much Junior television as well as occasional College hockey (thank you Big10 Channel) to familiarize myself with players..

            That said, I would never say that I know enough about most players to know who we should draft, and this is why *I am not a scout*.

            Don’t mistake my comment that stats are less important than people make them to be as saying that stats are useless (at least 2 people have now responded to me in the same way, so either I didn’t make myself clear very well or a couple of you jumped to conclusion on that – but I’m not trying to be rude about it, so sorry to be blunt). Stats are very good in helping you get a quick view of a player. You can really dig pretttttty deep with advanced stats. I’ve used them to support arguments before, so I’m familiar with them too. But I’m not going to entirely base an opinion on a player in stats alone. So what I’m saying is it takes a combination of the stats as well as seeing player to get a true opinion of how a player is.

            However, I find that most of the stats guys formulate opinions almost completely on stats alone and pick and choose the stats to use, as well as who to use them on.

            Anyways, for a scout, a team is going to use those stats to identify players who they should take a look at. Then they’re going to go in person and go watch those players and really see what they do in a game situation.

            There is no stat that tells you that Horcoff got kicked out of the circle on a phantom hand pass, is there? This is just one small example, but all these unmeasurable factors add up in the course of a game/season. There’s reasons that certain things happen, and that’s what stats can’t reliably tell you.

            We both agreed that it works best with both tools in the belt, and that’s where the argument should have ended. However, I feel I can judge a player by seeing them enough times more than I can judge a player by stats without ever seeing them (in most cases).

          • OB1 Team Yakopov - F.S.T.N.F

            “Watching a player tells us what he seems like he should be able to do, but stats are able to quantify the results of those observations. In short, watching tells us who should be batter, but stats can verify which player actually has better results.”


            A great example to back this up:

            If you were to remove the name off the back of the jersey of say Chad Kilger and Luc Robitaille and simply watch both guys play a couple of games (without looking at the score sheet after) I’d bet almost everyone would walk away thinking Kilger is and will be the far better player.

          • Chris.

            Did anyone see this awful TSN “documentary” Oil Change about the draft? Starring the shining personnalities that are the Oilers senior management team?

            Maybe the strangest moment in the show is when Kevin Lowe plunks down at his hotel room computer to read the stats page from the World Hockey Championship.

            “Eberle had an assist. That’s good. Magnus had a goal!” K-Lowe uttered, then popped a metamucil and took a cat nap.

            Teams SHOULD use advanced stats, but they are not statisticians. I’m pretty sure they don’t use regression analysis tools, and don’t compare player stats against assumption violations. They are also promoted into positions as managers who oversee 8-figure budgets, however usually have no formal management training and experience.

            This serves to explain things like Wade Redden’s contract or the fact that the Sutters have the league’s worst draft record.

          • cableguy - 2nd Tier Fan

            many teams use a version of advanced stats that a former GM (Mike Smith i think?? will double check) is spear heading. The stats are used for everything from player comparison, to UFA and RFA evaluations, to arbitration cases.

            obviously, it wont be the only variable in trying to build a roster, but it is taking on a larger role within the NHL

          • OB1 Team Yakopov - F.S.T.N.F

            Kind of strange that real live NHL teams use them then. I guess the anti-math crew on the net know more about evaluating talent though.

          • Oil Kings 'n' Pretty Things

            Stats can be skewed depending on your point of view or what you find relevant.

            When you’re watching a game at ice level, you can see you keeps up, who players avoid, and who looks the other way in fear.

            Cameras only cover so many angles,(HD included)

          • Oil Kings 'n' Pretty Things

            Stats don’t lie.

            Interpretation of stats is what you’re getting at.

            If Player X scores 20 goals in a year and 8 are on the power play, that is a statistic. It’s a fact. It’s an objective measure of an event or quantity.

            Where it gets fuzzy is when someone like Willis comes in and explains why Player X is a good or bad player based on those facts. It’s an opinion based on measurements. The opinion can vary because it’s not objective, just like your opinion of a player can vary based on things you see when you’re at the game.

            They’re different ways of observing the same thing.

            Let me put it this way. I weigh 195lbs. That’s all the information you have about me. How do you decide if I’m overweight or not?

            1) You could ask my friends and family and find out if I shop at the big and tall store.

            2) You could take more measurements – shoulder, waist, and hip diameter, height, etc.

            Odds are, either method is going to lead you to the same conclusion. You used pretty different methods, but they’re both entirely valid.

          • book¡e

            A person’s memories will likely be more skewed towards their point of view or what they find relevant.

            Average fan is going to remember his favourite player’s best moments rather then dwell on the negatives. Likewise, he is going to remember every single mistake made by his whipping boy, and gloss over the good done by him. That’s human nature.

            At least with stats you can come up with substantial reasons for why things happen. For instance Horcoff’s 50 points in 53 games season. In this season he faced relatively easy competition (Stoll-Who was billed as future captain by many got the hard matchups and had a terrible year prompting a lot of fans to turn on him). Horcoff also had a high shooting percentage.

            Average fan’s reason for his quality season was that Horcoff found some magical sticks in Mexico.

          • cableguy - 2nd Tier Fan

            problem is, 99% of the fans have little clue how the pro game actually works.. We havent played pro, havent coached, havent sat in the room etc etc etc. watching something alot doesnt really make a person an expert.. does it?

            for example, could you go out and build a house after watching alot of “holmes on homes”? or, does it give you enough info to really make you dangerous?

  • OB1 Team Yakopov - F.S.T.N.F

    If I knew how to give props I would…

    Can you imagine the tail Old Spice Guy & Doseki’s Guy could pull if they hit the clubs together?

  • Lowetide

    I think math and statistics can tell us a lot, but also believe there is more than one way to reach a conclusion.

    For instance, NHL teams have been scouting potential draft picks for almost 50 years and have as a group done a splendid job making sure almost all of the elite talent is off the board early.

    On the other hand, math/stats can tell us a tremendous amount about context. Generally speaking, if we can agree that previous performance is the best predictor of future performance then there’s no real issue.

    Age/injury, context (how a player is being used), there really are very few things that will vary performance year over year.

    If I say “Robert Nilsson could have a 35 point season in 2010-11 if used properly” it would be my assumption most people would agree. Nilsson HAS in fact posted a 41-point season and two more (29 & 27) in the range.

    He’s mid-20s, should have a fire lit under his ass and should be affordable to several teams who can sign him and throw 4th line minutes his way. If he takes advantage, he could have a season that resembles his 07-08.

  • Chris.

    Andrew Lang Wrote: He uses statistics as a drunken man uses lampposts—for support rather than for illumination.

    I quoted this to Willis a couple years ago after a heated exchange… but I still respect Jon and the value of stats to provide a fresh starting point for discussion. Cheers Nation!

    *pouring yet another stiff drink while balancing debt in anticipation of paying for another great year of Oiler season seats*

  • Ross Creek wrote:

    I believe Lowetide is the “original” Willis. At least I think Willis has referred to himself in that manner – Lowetide being the reason Willis got into blogging, IIRC.

    Lowetide was the first blog I ever read, and the reason I got into commenting and blogging. He also got me thinking about the game in different ways.

    I owe him a great debt for that.

  • @ Chris:

    You and I have had a few of those over the years (is it really years?) and I’ve always appreciated the fact that we could disagree on things without drifting into some of the pointless internet bickering that goes on here and elsewhere.

    Speaking of which, the more I look at the draft, the more I come around to your POV of the Prendergast draft years.

    • Chris.

      Cheers Willis!

      I’ve been drinking… have missed your daily offerings as you’ve changed careers; and have been struggling with the need to tell you, “I told you so!” *wink*

      Thing is: even Brownlee was wrong about Sequin; and though I think you were overly optimistic about some Oiler draft picks… there may have been some development issues.

      It takes more courage to write a blog and defend your positons than to just post a few comments attacking those positions. Lord knows I’ve made more than a few mistakes along the way!

  • Leopard

    I have a Entry Level contract question. Do the contracts only count in the number of contracts allowed if the players play in the NHL?

    Thanks in advance if anyone knows the answer.

    • This is from memory, but I’m fairly sure I’m correct here:

      There’s a 50 contract limit, which includes all the players with an NHL clause in their deal except (and this is the part I’m a little hazy on) the junior age players whose deals can slide a year if they don’t play professional hockey (for instance, if Taylor Hall goes back to junior, his three year Entry Level deal won’t kick in until next season, and he wouldn’t count against the contract limit).

      There’s another limit – 100, I think – of players whose rights a team can own, a limit that includes RFA’s and drafted players who have yet to sign a contract but also have yet to be eligible to re-enter the draft.

      • Leopard

        Thanks Johnathan,

        Is there a reason does all their calculations with a 23 contracts per team? Is it only because that is how many players are allowed on a teams roster at once?

        Thanks again.

  • book¡e

    If it were possible to measure every factor in a game, stats would be everything.

    If it were impossible to measure any factor in a game, stats would be worthless.

    Guess what, it is possible to measure quite a few factors in a game.

    Stats are an important way of viewing the game. Observation, which relies upon complex mental processing of observed data, is another way of viewing the game. Both have value. Both have weaknesses.

  • You guys do know that when people watch players play, they don’t close their eyes when a player scores a goal, right? 😛

    It would be incorrect to say that people who are pro-“game watching” don’t use any stats at all in determining how good a player is (really it’s natural human behaviour to quantify things). I’m not so sure it can be said that some of the stats guys have ever seen some of the players they rate via stats, however.

    Really, for best results, it takes a combination. Stats have come a long way… I use advanced stats as well as normal stats all the time too, but it’s only a small portion of how I determine how good a player is. Really though, if I can’t see first hand (i.e. on TV, or in person) how good a player is, I might dig up wads and wads of stats, but chances are I’ll read a scouting report on a player to really make a determination. The scouting report would be an account of another person who watched the player play… they base these mostly on watching a player “in person” as opposed to reading stats later.

    Another example. Let’s get granular here, and let’s exaggerate things a bit. At a game between the Ducks and Oilers, Joe Smith for the Oilers has 1G, 0A and is a +1 with 10:00 TOI. Dave Madeupguy of the Ducks has the same stats, and also faced the same quality of competition, while playing with the same quality of teammates. How do you know which player is better in that game? Let’s say I watched the game. I could tell you that Joe Smith was better in that game because on the goal he scored, he had to battle his way out of the corner, and then deked and dodged through defenders to score the goal, whereas Dave Madeupguy scored his goal by staying parked at the side of the goal and tapped it in (or I could tell you something different… he was parked at the side of the net, but he stickhandled in a phone booth to score his goal). We could look at the hits stat and see that Madeupguy threw 5 hits to Smith’s 2. However, I could tell you by watching the game that Madeupguy’s 5 hits were pretty soft hits, or hits on guys that were out of the play, or I could tell you that Smith’s hits were huge, and one of them caused the recipient of the hit to slow down for the rest of the game.

    Stats can get very fine, but people using stats in an argument are only going to go so fine. And unfortunately, that’s where stats then become a rough estimation of player performance vs. a real account of performance.

    Contrary to popular belief, saying something like “there is no substitute for watching a player play” doesn’t mean you ignore how many goals they scored in a game or how many hits they threw though. I think what we’re saying is, I’m not going to dig up some numbers in a paper to judge how good a player is. But I might watch a game, see Hemsky’s wizardry with the puck and remember how many goals / assists he got in a game. But I might not see any goals or assists in that game, but I might see some crazy effort there though that still makes me feel he’s a good player. Anyways, I’m rambling now.. so time to end it.

    • You do know that just because a person prefers statistics do reach a conclusion it doesn’t mean they close their eyes during play right?

      I’ve seen him with my own eyes go into corners fearlessly for pucks. As well, his stickhandling and vision are superior. I’ve seen some pretty incredible highlight reel dekes from him as well as numerous seeing-eye passes. That makes him a pretty tremendously gifted playmaker as well as a danger as a scorer.

      Sounds an awful lot like Robert Nilsson during one of his good stretches. If only there was some way to differentiate between them…

      Your example of Smith and Madeupgay works more often the other way. Stats are intended to be used within the framework of the largest sample size possible. Watching games cannot match that unless you are the player’s mother.

      It doesn’t end at you should use both tools, because you are arguing that watching gives MOST of the picture and stats adds a small part. Stats actually gives to much much more of the picture, and watching the player fills in the blanks on how his stats are generated.

      I’ll repeat the key phrase again from last post; watching tells us who should be better, but stats can verify which player actually has better results. That is not a “small part” in analysis. It is the most important part, otherwise you could end up with a team of Nilssons and never understand why you keep losing.

  • OB1 Team Yakopov - F.S.T.N.F

    Futuristic move by Oilers i might be interested in . Talk of Souray and the Islanders who are way under the cap this season in Journal today .

    Souray , Horcoff , Khabibulin , Cogliano and 3rd round pick next season for Yashin and Roloson . Sounds rediculous but economically wise ! Gives team incredible cap space to work with .

  • Goals and assists are obviously a useful stat.

    But so are shots and shot location, goals for/against while on ice, strength of opponents/linemates, shifts starting/ending in defensive zone etc.

    And then there are subjective stats from assigning errors from video, to scouts giving games scores on all subjective measures of interest to the team.

    Every scout, gm, and agent wants to know if the results are equal to the shifts given and are likely to repeat.

    Every objective stat and subjective stat and subjective opinion is in competition to see what contributes to assessing the results players are likely to deliver going forward.

    Hockey stats are harder than baseball stats but in a world where every shift in on video everything of value that can be plotted down will be evaluated for predictive power. Side by side with the assessment of the various subjective scouting measures.