Taylor Hall: Advanced Statistics

EDMONTON, CANADA - OCTOBER 16: Edmonton Oilers left wing Taylor Hall  warms up before a game against the Calgary Flames at the Scotiabank Saddledome on October 16, 2010 in Calgary, Alberta, Canada. The Flames beat the Oilers 5-3. (Photo by Dylan Lynch/Getty Images)

By eye, Taylor Hall is an NHL player. This is what I think, and it’s certainly the consensus opinion among editorialists in the mainstream media. Despite that opinion, Hall’s stats line remains underwhelming: seven games played, one point (an assist), and a team-worst minus-4. What do Hall’s underlying numbers suggest?

There are a few different angles we can look at Hall’s season through, but I thought we’d start with even-strength scoring chances, since even those skeptical of advanced statistics seem to find some value in those numbers (correlation to those fancy Fenwick/Corsi calculations notwithstanding). Courtesy of Dennis over at mc79hockey.com:

  • Game One: +8/-3 = +5
  • Game Two: +2/-3 = -1
  • Game Three: +1/-3 = -2
  • Game Four: +3/-5 = -2
  • Game Five: +6/-4 = +2
  • Game Six: +1/-3 = -2
  • Game Seven: +4/-4 = even
  • Totals: +25/-25 = even

“Even” may not sound impressive, but it is; not only is Hall coming out even against NHL caliber opposition, but he’s easily outpacing the team as a whole (they’re a cumulative minus-18 at even-strength through seven games).

Corsi numbers tell a similar story; Hall’s a hair below even by that count but easily leads the team overall, and he’s done it without the advantage of a bunch of offensive zone starts. He’s also been drawing more penalties than he takes in the early going.

Despite all those numbers, the offence hasn’t come yet, either on the power play (where Hall’s getting regular minutes) or at even-strength (where Hall’s 14:05 per night puts him third among forwards). I’m confident that it will, at least at even-strength, and the underlying numbers support that confidence.

Does that mean the Oilers will be right when they choose to keep him up past the nine game mark? Likely it depends on individual perspective – those who value the greatest possible competitive integrity will say yes, while those looking at the Oilers long-term cap situation and prospects for 2010-11 will say no, both with some justification. The former can point to the fact that Hall is ready, while the latter will be disappointed that the Oilers didn’t use Hall’s superficially slow start as an excuse to save some money down the road.

Both should be happy that the first overall pick looks like such a good player so early in his NHL career.

  • OB1 Team Yakopov - F.S.T.N.F

    Willis’ stats are not bunk. The are stats which are used by NHL executives to determine which players are playing outside of the simple points mark.

  • VMR

    I guess it would be like someone teaching a grade 6 math class about integration. The grade 6 student would obviously find it “bunk” (which is also a grade 6 word) becaus they don’t understand it’s applicability.

    Great article, Willis.

  • Quicksilver ballet

    If there was ever a thread that required a bailiff, this is it.

    Taylor pots his first as an Oiler this evening, but not before getting into his first fight.

  • ogorr

    Thanks JW for the article,

    I have a question I’ve been wanting to ask, maybe you’ll be able to eloborate on it for me.

    First off I do beleive that advanced statistics have a place in evaluating players performance, obviously there are certain factors that can not be quantified, “compete level” being one of them and I think that there are certainly cases where these non-quantifiable variables directly relate to actual numbers but there is probably cases where they don’t. Maybe its specific to a players situation, maybe something else? I don’t know

    In the case of Hall I think that his intangible tendencies will help influence his numbers, if not now then definately in the future. I just see a guy who doesn’t give up.

    Anyways, my question is how much creedence do advanced stats factor into management and coaching staff decisions on the team, be it line combos, trades, drafting (if available and if relevant at lower levels)? Does it have a presence in the day to day operations of a club? Specifically I am curious about the oilers but also interested in other teams.

    To me, my view of coaches doesn’t include the advanced statistical analysis of a player. I kind of think that alot of these coaches and management types are going to trust their gut and their eyes despite what stats may say.

  • Delightful.

    Given that DSF can’t even correctly identify who is tracking the scoring chances (despite the fact that it’s in the article above) it’s pretty hard to take his thoughts on them seriously.

    Zamboni Driver hates advanced stats. So he clicks on an article with “advanced stats” in the title and tosses out a comment indicating his distaste for them. That’s been his M.O. for a while; once again, it’s difficult to care.

    Traktor takes an article from two years ago, then presents his wildly inaccurate remembrance of a position I took at that time, hoping to get under my skin. This has been par for the course for him since Lowetide was a fresh face on the internet. It remains, as always, difficult to muster up much enthusiasm to waste time debating him.

    Thanks for coming out, guys. Your tired and generally irrelevant comments generate pseudo-debate, which bring more comments and traffic, which helps the site. It’s appreciated.

  • ogorr

    @ ogorr:

    The general tack I take is to evaluate players on result more than skillset. It’s not a perfect tack – for instance, prior to this year I’d grossly underrated Martin Marincin – but intangible things that impact performance are accounted for.

  • ogorr

    How would you feel about Taylor Hall if he ended up with this stat line:

    Games=75 Goals=12 Assists=21 Points=33 PPG Average=.44

    Right now he is at:

    Games=8 Goals=1 Assists=2 Points=3 PPG Average=.375

    That first stat line was the an 18 year old Moose in his rookie year. So far Hall is on pace to be within a couple of points of Mark Messier’s rookie season. Hall’s play to date gives every indication he can put up even better numbers moving forward.

    Pass on the do-over, thanks.