Advanced analytics are the Blackhawks’ secret formula for success

[EldrickOnIce]

Originally Posted by Street Pharmacist

Not true at all. Data could reveal anything, just depends on what you're looking for. For example, what is the corsi for/against when McGrattan is dressed? Gut feel is somewhat helpful buy a wide angle lens will naturally see more than only looking at it one way

Which is one of the problems with stats. The data may not reveal anything. But it could perhaps be carefully manipulated to show almost anything you want.

[Major Major]

Wait, so I'm to understand that the way Bowman got his foot in the door was by introducing a kindergarten style 1-5 grading system, and it was later that he discovered that this was stupid?

Was 2001 such a far off and distant time that I could have walked into an NHL head office and said, 'What if the coach rates the player subjectively on a scale that is so vague that it couldn't possibly mean anything at all? I'll tell you what, a spreadsheet. That's right, it's this knew thing developed by a fringe software company called makerosoft or something. Anyway, this is going to be neat!' This gets your foot in the door?

The more Bowman talks, the more I suspect that he had nothing to do with his stanley cup winning team. If it weren't for a goofed fax, that I'm now convinced was his doing, we would only hear his name in reference to his pops.

[handgroen]

weird how unanimously panned the idea of his methodology has been so far in this thread.

nobody else thinks its a good idea to give the devil his due?
in his presser treliving hinted that he has his own system. and here's a wild idea, it helps him do his job.
So what does it matter if you think it's all toews and kane? i'm just glad to peek behind the curtain and see a little bit about how decisions are made in one of the most successful hockey franchises of the last decade.

[Iceman90]

Lets rip off this system and sell it to Burkie for cash. We need to plant a mole in the Blackhawks organization. Any volunteers?

[bluejays]

Originally Posted by neo45

I've heard rumors that Burke is actually pretty big into "advanced stats" but plays it down to the media. I feel like they can definitely contribute to a player's profile but are not the be-all end-all of scouting.

A friend of a friend I know who had an MBA and specialized in statistics was apparently in an stats presentation to Burke and other brass and the guy didn't just didn't seem interested at all. He was obligated to be there, but apparently just wasn't into it, so I do believe him when he downplays it.

[Major Major]

Originally Posted by handgroen

weird how unanimously panned the idea of his methodology has been so far in this thread.

nobody else thinks its a good idea to give the devil his due?
in his presser treliving hinted that he has his own system. and here's a wild idea, it helps him do his job.
So what does it matter if you think it's all toews and kane? i'm just glad to peek behind the curtain and see a little bit about how decisions are made in one of the most successful hockey franchises of the last decade.

But he didn't tell us anything... No real peak at all. Mostly just jerking himself off about how smart he is.

[united]

The value of "advanced" stats isn't so much sucking and drafting superstars - anyone can do that - it's in maximizing the value of your 2nd to 4th liners, and 2nd and 3rd pairing. Both performance wise and money wise.

If a team were to assemble a roster of guys that "advanced" stats rate highly - Moneypuck guys - like Mikael Backlund, Justin Williams, Clarke MacArthur, Brandon Dubisnky, Cody Franson, Justin Braun, Andy Greene, for example, it probably wouldn't make the playoffs. However maximizing the 2nd to 4th lines and 2nd and 3rd pairings with those guys allows a team to pay for the truly elite 1st line and 1st pairing talent to push them over the edge, while having competitive depth players.

Bowman doesn't mean they drafted or employ Toews, Kane, Hossa, Seabrook, and Keith solely because "advanced" stats say to - anyone who thinks that is stupid - he uses it to maximize the value of the rest of the roster to support those players.

The Penguins' issue (aside from inconsistent goaltending) is they have the elite talent (obviously) but their depth players are not only poor quality relative to where they stand on the depth chart but also overpaid. If they replaced overpaid/underperforming players like Orpik, Scuderi, Glass, Pyatt with equally priced or perhaps cheaper Moneypuck guys like above, they would be a lot better off and maybe even have some extra cap space. Unless of course you need to pay Orpik a few million dollars for leadership the Captain of Team Canada apparently can't bring on his own.


tl;dr - nobody says "advanced" stats are needed to see Toews, Kane, Keith are elite. They are used to get effective depth players on undervalued contracts, allowing you to pay elite players.

[Enoch Root]

The term 'advanced' is just so annoying - there is nothing advanced about counting shots for and against. Or comparing them against the rest of the team. Etc. It's just a few more ways to count opportunities for and against. It's more data.

More importantly though, what really annoys me about 'advanced' stats is that most fans don't understand statistics and the fact that it is mostly noise due to the small actual amount of data. In order to have really meaningful numbers, you would need years of data on a player and they would be retired before you acquired enough to be really confident about its value. So you end up having to use data from smaller sample sizes that is largely noise. In and of itself, that is fine. But fans put way too much weight in the results because most don't understand this.

But that's all fine. Every team uses the data because why wouldn't you? Every bit of information you can acquire helps. Having said that, there are no secret recipes that will give one team an advantage over another team - they are all looking at the data.

***

As for Chicago, I would expect that many - if not most, if not all - teams have some proprietary stats. That is not what is giving Chicago an advantage. What is giving them an advantage is that their team is extremely talented. And because they are so talented and deep, every decision that they make tends to work out well.

A good example from the past is the 80s Oilers. They were so talented that every single player they acquired worked out well and became a better player. Whether it was Tikannen, MacTavish, Krushelniski or whoever (there were dozens over the years), they all turned out to be good additions.

Conversely, look at any lousy team - like the current Oilers. On a bad team, every decision doesn't work. No matter who they acquire, that player almost invariably becomes a worse player. Is it because every decision they make is stupid? Or is it because the team sucks and no matter who you add to it, that player becomes worse?

Sure, there is probably some of both going on. But for the most part, good talented teams will make their management look smarter. And bad teams will make their management look dumber.

(my apologies for the long post and all that dirty Oiler talk)

[getbak]

“You can get bogged down in stats and analytics,” Bowman said. “That’s the concern, it’s still a dynamic game, there’s still an awful lot happening. It’s a quick, reactive game. It’s different than some other sports that are a little more static. So there might be a limit to how much you can use analytics. They don’t replace anything for us. But they do complement and supplement.”

A person with a more colourful way of speaking might even say, "Statistics are like a lamppost to a drunk, useful for support but not for illumination."

[The Fonz]

Agreeing with United.

The Hawks didn't use advanced stats to draft Kane and Toews, that's obvious. They got them from sucking really bad. But take a look at their transactions over the last decade:

Draft (outside 1st round):
Saad - 43rd in 2011
Shaw - 139th in 2011
Pirri - 59th in 2009
Kruger - 149th in 2009
Smith - 169th in 2008
Hjalmarsson - 108th in 2005
Bickell - 41st in 2004
Brouwer - 214th in 2004
Crawford - 52nd in 2003
Byfuglien - 245th in 2003
Keith - 54th in 2002
Wisniewski - 156th in 2002
Burish - 282nd in 2002

Trades/UFAs:
Sharp for Ellison + 3rd rounder
Hossa, UFA
Versteeg for Bochenski
Leddy for Barker + Johnsson
Handzus for a 4th rounder


I challenge anyone to find a team in the league that's done that well outside of the 1st round, other than Detroit. As for the trades/Hossa, they're all a landslide in favor of Chicago.

That management group can very obviously assess talent better than the norm.