Thursday, May 11, 2006

Ron Wilson


Several years ago I went to a Capitals game at the MCI Center in Washington DC, I don't remember exactly when, but pre-Jagr for sure, maybe 1998. A surprisingly knowledgable crowd, at least around where I was sitting, and a beautiful building in a shopping centre kinda way.

Anyhow the guy next to me was a hardcore Caps fan and a decent guy, he told a story about Adam Oates' unwillingness to take shots. Oates was still a very useful NHLer then, but as we all know he was a 'pass first' guy to a fault. According to the Caps fan; Ron Wilson had made a video of all Oates' goals, edited in with a bunch of nearly identical plays where Adam had made the pass instead, trying to convince him to take more shots. The stats at hockeydb.com tell us that this didn't work, but I love that type of stuff, so I've always sort of followed what Wilson was up too since then.

He was a hardass with the media too, long before that. And not just to be a jerk, but pointing out why a question was stupid in a constructive way (though I doubt that guys like Spector, Jones, etc will ever print a kind word about the guy because of it :D ). And I like that too, back in the day I would die a little bit each time Dave King would agree with idiocy from a reporter, even if it meant directly contradicting something he had said himself moments earlier, but I digress.

Anyhow, Ron Wilson is into some madass heavy analytical shit. Dude has realtime stats recorded during games and sent to him on the bench, over and above the stuff that the NHL records real time. Obviously the analysis it is only part of what he does as a coach, but it is clearly some pretty evolved stuff. And he's in cohorts with Doug Wilson and Joe Will, who started in the drafting side of the Sharks business I think (forget Pierre McGuire and his obsession with Parise ... Will and the Wilsons traded up to take Bernier one spot ahead of Parise, right now I'd say that's a better subject for the draft junkies at HF to be dissecting).

You all know how to Google, but here are a couple of good tidbits to get you started if you're into this sort of thing.

An article from the Washington Post with a funny little quote, this is about two years old.

Baseball these days, relies heavily on statistical analysis to mine previously unrecognized talent. The philosophy was made famous last year by the Michael Lewis bestseller "Moneyball." Wilson said he found Lewis's book "interesting, because I've been doing something similar for 10 years in hockey."

Hmm.

This business article interviewing Joe Will provides some fascinating insight as well. A shame that the writer didn't know more about hockey, because he got Joe Will talking and really could have gotten some relevant information out of him if he knew what to ask.

Best are the TV or radio interviews though, the curious stuff that he says rarely seems to get transfered to print.

Of course there is always a danger of outsmarting yourself as well. And one could argue that Wilson is one of the guys pushing it too far. You could also argue that a smart guy doesn't give away his edges by telling every other effer how he's doing it. The good thing about him though, is that he loves to talk, you can piece together what he's up to if you listen to him speak enough, at least somewhat.

I could ramble on this subject until absolutely nobody was reading any more, but I won't, I promise. :-) Just a couple of points.

I noticed in the Bernier interview between periods that this young player acknowledged that Thornton usually plays against the other team's top defensemen and best forwards. That his line's job was to score and they'd been doing that, something to that effect. That may seem minor, or even obvious to the type of person that reads this blog, but that's damn unusual for an NHLer to have any clue as to the 'hows' and 'whys'. A guy like Bowman would never even think of letting players in on that shit, clearly Wilson does.

I've got a good read on a lot of things they are up to, common sense really. And clearly a lot of this analysis is derivative of the time-on-ice data, from different leagues. This is the source material for the shift charts. But some things I don't get, the focus on 2nd period data, and on shifts length and pattern. Hell in an brief interview with Doug Wilson on CBC in either game 1 or 2 he brings up shift lengths as one of the keys in the series too, WTF?, I mean there are a million other things to talk about. And what happened the shift before the shift that something else happened ... I mean anyone who has kicked at Ruff's 'shots directed at net' metric and has a feel for the way the lines rolled can see some obvious patterns, but I have no idea how a guy would even begin to quantify that.

Interesting guy, this Ron Wilson.

20 Comments:

Anonymous lowetide said...

This is terrific stuff and I think someone is going to make a fine living from being the "Bill James" of hockey and identifying what these fellows are doing.

I wonder how many NHL teams and coaches aren't paying attention to this stuff?

5/12/2006 8:20 am  
Blogger Dennis said...

I suspect that just like in MLB some teams will never use that kind of data. Maybe I've been just reading less because of the Oilers recent run but it seems like the old "new school vs old school" debate as died down a little with both sides retreating to their respective corners and thus letting their team's performances cast the next stones.

Good point about Wilson outsmarting himself by letting everyone in on his MO. But I think the guy's such an egomaniac that he wants people to know why he's got an edge and if you happen to hone in on his methods and thus gain an edge yourself then he's just hoping you'll reference him and that will be his reward. Wilson sorta reminds me of a serial killer in that regard. It's not enough that he gets away with murder but he's also gotta tell you where the bodies are buried. Poor analogy I know but just something that popped into my head.

I wish I'd been watching that Bernier interview with my buddies. I'm driving them nuts lately with all the shift charts, line matching, etc etc and it's tough sledding bringing them over to the numbers side.

5/12/2006 10:35 am  
Blogger Vic Ferrari said...

dennis:

I agree with all of that, this is ego driven for sure.

Wilson has credited Roger Neilson with introducing him to the analytical side of the game. I imagine that Wilson played for him somewhere, I dunno. I suspect that Wilson imagines himself like this generation's Nielson, an innovator who gets props from the next generation of coaches. Sort of like the way Dylan and the Beatles often get listed as musical influences by the cool rock bands. :-)

You have to credit Wilson for usually playing an aggressive style though (and I know his Sharks don't chase much and played long stretches in a passive 2-2-1 last game, but still, on the whole it's assertive hockey). Most of the really technical coaches are control freaks by nature, and are loathe to hand that sort of decision making power to the players on the ice. Especially with a young squad like he has now.

As for the 'numbers thing' ... I think that the stuff you did here a few times, like the scoring chances and matchups and such for game 2. That's not really numbers, but anyone who reads that should have a much better idea of what happened in the game, and why. And most coaches should start seeming a lot brighter than fandom usually credits them. The numbers from the shift chart data could prove you are right if someone argued with you, but really that's just the less detailed shorthand of what you wrote. Amongst other things, a lot seems to point to Wilson having figured a way to turn the shitload of numbers that are the TOI lists (for the NHL, SEL, DEL, wherever) into something meaningful, something more like what you did with game 2. Which probably helps Wilson's bosses (GM and player personnel) more than it helps himself.

.

On the Bernier thing, that surprised me. I remember hearing a radio interview with Hull when he was a Wing, he wondered out loud why he had never played a shift with Fedorov. Talked about how he thought he and Fedorov's skills could mesh and they could score a whack of goals together, spoke like a teenager who'd spent the day playing NHL2K or something.

Fedorov was playing some of the toughest minutes in the league then, no way in hell was Bowman going to let Hull see that level of opposition.

Hull talked about how he'd barely had a conversation with Bowman since he'd been a Wing, but he was PKing quite a bit and took that as an indication that he was better defensively that people thought. (Really it was just Bowman's way of limiting his 5on5 minutes, and as per a Wilson ditty in the links above ... Kariya was/is a helluva PKer too)

Good Christ. Think Brett, THINK! :D

5/12/2006 11:36 am  
Blogger RiversQ said...

Funny that OF has a thread titled "Ron Wilson: Putz" right now. Man, that board is a gong show right now.

Anyway, it was interesting to see the trades that team has made in the last two years in that article. They really have done quite well on that front and they've picked on big money teams to do it. It's also funny to me how Sather deals cheap and useful Ekman to SJ a couple of years ago and then proceeds to make the same mistake with Niemenen this year.

Wilson's stuff is very interesting but he certainly does have a grating personality. This makes it easy to love the fact that he blew a lot of smoke about the goaltender interference penalty in this series only to watch his team pick up a six-pack of penalty minutes on that rare call in Game 3. That was beautiful. I swear you could see him wiping the egg off his face during the 5 min major when Marleau grabbed one.

5/12/2006 12:38 pm  
Blogger mudcrutch79 said...

I'm not sure what to make of Wilson's comment about Moneyball-given that he generally seems like such an egomaniac, I figure that he's saying "Look I was doing this a long time ago." I don't know if he knows his history of baseball that well but Billy Beane didn't invent that stuff.

Another interesting comment that I wonder about-it's a quote from the story, not a direct quote from Ron W.: "They determined that statistics such as penalty-killing and goals against average were critical." Not sure what to make of that-maybe we're talking about a player's goals against average there as opposed to a goalie's?

Another great quote:

"And since the difference between winning and losing is like one-tenth of a goal, you're dead. For us to get from 30th to the top 10, we had to give up one less power-play goal every three games."

There's something from the Joe Will article that screams "We pay attention to goal differential" too.

Great stuff Vic.

5/12/2006 1:24 pm  
Blogger Black Dog Hates Skunks said...

Great post Vic - really interesting to me because I'm definitely more of a "feel" type of a guy when it comes to watching a game and the stuff that you guys and Mc79 for that matter bring to the table in terms of analysis blows my mind sometimes - it quite often belies the accepted "wisdom" of the moment.

Wilson is a smart smart guy and he knows it too - interesting but totally rubs me the wrong way.

They have built a great team out there though - only misstep in my mind was the big contract for Nabokov and I think they'll unload that. And trading Brad Boyes - but what did they get for him? Maybe not bad.

5/12/2006 2:16 pm  
Blogger RiversQ said...

That's definitely a player's GAA, mudcrutch79.

It's also pretty clear that the Sharks are in the Lemaire school regarding taking penalties - it's a serious no-no. Obviously they're measuring this as well. Possibly some combination of ES GAA and non-offsetting penalties/hr. MC79, you could probably put together a decent list like that with the data you've already got, at least from strictly the penalties angle.

5/12/2006 2:43 pm  
Blogger James Mirtle said...

Neilson was Wilson's first NHL coach actually, back in 1977 with the Maple Leafs, of all teams. Both of them only lasted a year and a half or so with Toronto, and Wilson was a 20-year-old kid at the time, so it's interesting they stayed in touch.

Maybe this just shows some coaches have been playing the stats for a heckuva lot longer than any of us though.

5/12/2006 2:43 pm  
Blogger Dennis said...

I was thinking it was the Leaves because I'd see Wilson in a TO uni during a game on the NHL Network.

Maybe Wilson always had a head for the game or maybe he realized at a young age that his future wasn't really as a player.

5/12/2006 3:53 pm  
Blogger Black Dog Hates Skunks said...

meh - watching Wilson's postgame remarks

all "lucky bounces" and "they got the breaks"

if he's worried he sure doesn't show it - he is one arrogant prick

"we're unbeatable" at home he says

what I would give for a Oilers' win on Sunday

5/12/2006 9:25 pm  
Anonymous momentai said...

As I was driving home today, I listened to MacT's mid day press conference on the TEAM 1260 and I came across something somewhat relevant to this discussion on Wilson.

MacT mentioned about playing the percentages. He cited an example where if 1 line was out on the ice against a certain line that resulted in 8/10 chance of playing in the offensive zone... you have to play the percentages. So, we know that MacT is thinking about those things.

Wilson is a smart guy, no doubt. But I would worry about how long a lifespan he has in an organization. Everywhere he has been has had a limited tenure. You've got to wonder if the ego is a factor in that.

Oh yeah. Good win by the Oil tonight. ;)

5/12/2006 9:34 pm  
Anonymous lowetide said...

Wilson's press conferences are a blast. I love this stuff, not the "well we gotta get out there and try, try, try" crap.

MacT's was good too, but it's easier being the winner.

5/12/2006 9:36 pm  
Blogger Dennis said...

The Sharks had better be unbeatable at home because they were up by two goals with 27 minutes left in G4 and basically wound up shitting the bed. Now to be fair Tosokla did start the suckfest but his teammates collapsed in the 3rd period.

So I would't say Wilson would bet 10 bucks on the Sharks winning G6 in Edm.

5/13/2006 11:18 am  
Anonymous huffpuff said...

Thanks Vic. Very cool stuff.

Had a quick peek at RinkNet. This is their intro:


RinkNet Scouting Software began developing software in 1998 with the first client being the National Hockey League’s Nashville Predators. Before the end of the first season, the Atlanta Thrashers and Columbus Blue Jackets joined, creating a full-service product. As the program and its database grew, the RinkNet family has grown each year. Today, RinkNet services 26 NHL clubs as well as 13 Canadian Hockey League teams plus national team programs for Canada, Finland and most recently in Sweden. A total of over 700 scouts, managers, and hockey office personnel currently use the system.


Columbus? I guess Mr. MacLean didn't make very good use of his resources.

The 4 teams who are not their clients, I wonder if they totally ignore this stuff or perhaps use a different company. Maybe they even do it in-house?

.

That tidbit about Kariya being a PKer...do you mean this link?

http://sports.espn.go.com/espn/wire?section=nhl&id=1801596

About what you wrote earlier regarding shift lengths/patterns....there is an interesting quote from the article (which I'm sure you've already seen) which seems relevant. Though it's not directly from Wilson's mouth, I think the writer probably wrote it b/c Wilson mentioned it to him some time during the interview:

...the coaches compile statistics with a zeal that nearly matches baseball's sabermetricians. Want to cross-reference shift lengths against goal-scoring in every month of the season? Wilson can look it up on FileMaker Pro...

Sounds as though he may be trying to figure out so sort of optimal avg shift length that maximizes GD. I dunno, but interesting...

.

Here's another article (which I'm also sure you've seen):

http://www.wired.com/news/gizmos/0,1452,63105,00.html?tw=rss.GAD

I find this tidbit interesting:


The team burns DVDs of players' shifts on the ice and hands them out to individuals. The Sharks regularly watch the DVDs on their laptops while traveling from game to game.


And I thought they just play cards or listen to ipods. Wonder if Wilson gave them pop quizzes.

5/14/2006 2:01 am  
Blogger Vic Ferrari said...

RiversQ:

Ya, he's big on GAA by player. His metric for players, (5on5 points) / (5on5 icetime) - player's GAA ... that drives that point home. Though really the main difference between that and 5on5+ and 5on5- is that defencemen get devalued big time. I suspect that the main reason for that is that it will make the 'tough minutes' numbers a bit cleaner with a smaller sample.

Though San Jose clearly seems to like defencemen that can register points. And don't seem to have a helluva lot of their payroll tied up in defencemen.

5/14/2006 1:47 pm  
Blogger Vic Ferrari said...

James Mirtle:

Ya, I think you are right. Some of this stuff dates back a helluva long ways. I think that the NHL was way ahead of most sports in the use of meaningful stats. I'll post some early Roger Neilson stuff one day, since he is the inspiration for the majority of today's technical coaches.

I mean it is hard to imagine that somebody could go through the game tape, tearing apart every shift, without seeing some patterns and seeing ways to keep track of it all. In fact it is unimaginable.

Hell, we know that the NHL has been producing TOI sheets (every player's shift start and end time in every game) for at least 15 years. They weren't doing it for their health. And that Neilson was using off ice staff to record various data (mostly centred around scoring chances) in the 80s. So as to have real time stats on the bench.

It's not a stretch to think that he was pulling 2nd generation stats off of game tape in the 70s. In fact it would be surprising if he wasn't. Others, like Bob Johnson, too. But probably not so early.

5/14/2006 1:54 pm  
Blogger Vic Ferrari said...

momentai:

Terrific stuff, thanks. The radio interviews, except postgames, they're always the best.

I wondered why we didn't see power vs power in G3. But this does make sense. I would bet that Thornton started his shifts started with possession only about one time in four through that game.

With this in mind, I had a look at the shift chart (a bitch to read this one because it's so condensed, and my TOI chart reader didn't want to work on this chart for some reason). Anyhow, it looks like Wilson was continuing on with his "Marleau after Thornton" thing, and running a short bench, made it easy for MacTavish to get Horcoff out the shift before Thornton ... about 15 out of 19 shifts by my count before there were some penalties and it got muddy.

And the Horcoff line was out against the Sharks depth forwards (and Hannan) for the OT winner, and a helluva lot of chances before.

Going back to Horcoff vs Thornton in G4 was another odd one. I assume he was predicting Wilson's next move there. And if Dvorak was healthy then that's the obvious move in any case I think. Winchester isn't good enough though IMHO.

Methinks that this team is going to be a bitch to play against if Hemsky can ever become the kind of player that can go toe to toe with the likes of Thornton.

.

Another radio interview bit: On Gregor's 1260 postgame show they had Huddy speaking to the media after the long OT game, in lieu of MacTavish. He had no problem talking about the D, jumped on those questions, but was oblivious to what the forward matchups were or even how they had played. Division of labour I guess. :D Hell he didn't even know who the Oiler forwards were that got rested with the OT timeout.

Another Huddy tidbit: In a PPV game in San Jose earlier in the year, the surprisingly clever Karius asked him about how he was matching the D so well. He mentioned that the fact San Jose were a three line team made it easier, and that he looked over to get cues as to which forwards the Sharks would be sending out next shift. Lesson being: good questions beget good answers. :-)

I've watched him a few times on the bench at home games since ... and there is no way in hell he is looking over there. Off ice guys must be giving them the tips. I know the Oilers have video monitors linking them to office coaches. Obviously audio as well. And I'd assume that Billy Moores is in a video room somewhere grinding into the more specific details as the game is being played.

5/14/2006 2:15 pm  
Blogger Vic Ferrari said...

huffpuff:

Great post, I'm sure I've read much of that before.

On the shift thing, I'm sure he looks at more than just shift length, though he does break it down to that simple measure a lot in interviews I've heard and read. And Doug Wilson specifically mentioned shift lengths as a key to the series in a CBC interview sometime during or before game 1 or 2. Which is a effin strange topic to even bring up, so who knows?

I read a paper on database development for the New York Rangers (published in 1994 in Alabama I think, but clearly dating back to Neilson's days as head coach in the late 80s, early 90s) ... one of the many curious things they were tracking was "players who were on the ice two minutes before the goal was scored" ... I think we can all see what they are driving at, and I'm sure guys like Hitchcock and Ruff have their own ways of evaluating that which may or may not be more sophisticated. Still, interesting to see guys try and peg values on things like "possession and momentum". Anything for a small edge I guess.

On Rinknet: Props for that. Last I'd read they only had 16 teams subscribed (one of them was the Oilers). By my memory they also make scouting software, so whether teams are investing because of that or the data from other leagues ... hard to say. Hell, several are probably buying it just because everyone else is.

On the Kariya thing ... it's a favourite story of Wilson's. He tells it much better than it reads there (though I would have sworn the other player was Todd Harvey, who wasn't even in the league at the time ... so my memory should never be trusted fully :) ) I think that he says that was the first time he actually hit a player with numbers, which clearly set his path, otherwise there is no way in hell a player like Bernier should appear to know so much about the "whys" of the gameplan.

The fact that a depth player was ragging on him personally about PK time says something about his style with the players too. Try that with Crawford or Bowman and you'd be back on the 4th line or in the AHL in a heartbeat.

On the DVD burning thing ... damn obvious when you think of it. And though I've never heard of anyone doing it before ... I imagine that tonnes of coaches do the same. Or if they didn't before then they started when they read that ;)

I've read a bunch of times that Tim Hunter does a lot of the video editting and such. Good for big Tim, he was a terrible interview when he first broke into the league, came across as a bit slow. Clearly he isn't. BTW: I read a blurb once where a reporter said he was talking like "a geeky Radio Shack clerk". Good Christ, anyone with sense would refer to him as either "Tim Hunter", "Mr. Hunter", or "sir" :D

5/14/2006 2:39 pm  
Blogger Vic Ferrari said...

One more thing:

If anyone hears any rambling interviews from Mactavish or Wilson or the like, where they go a little deeper into the details than normal, much appreciated if you let us know here.

I love that stuff.

5/14/2006 2:41 pm  
Anonymous Julian said...

I was trying to get a friend of mine to apply for a job Rinknet had posted on their website, but he ended up going elsewhere. I was thinking I'd have my own personal mole in there.

5/15/2006 10:58 am  

Post a Comment

<< Home