Established in 2006 as a Community of Reality

Welcome to the Neno's Place!

Neno's Place Established in 2006 as a Community of Reality


Neno

I can be reached by phone or text 8am-7pm cst 972-768-9772 or, once joining the board I can be reached by a (PM) Private Message.

Join the forum, it's quick and easy

Established in 2006 as a Community of Reality

Welcome to the Neno's Place!

Neno's Place Established in 2006 as a Community of Reality


Neno

I can be reached by phone or text 8am-7pm cst 972-768-9772 or, once joining the board I can be reached by a (PM) Private Message.

Established in 2006 as a Community of Reality

Would you like to react to this message? Create an account in a few clicks or log in to continue.
Established in 2006 as a Community of Reality

Many Topics Including The Oldest Dinar Community. Copyright © 2006-2020


    Los Angeles Kings

    jedi17
    jedi17
    jedi17
    Moderator
    Moderator


    Posts : 10738
    Join date : 2013-02-20
    • Post n°1

    Los Angeles Kings Empty Los Angeles Kings

    Post by jedi17 Thu 09 Feb 2017, 7:34 pm

    How Are Kings Like Bruins?; On Patrick O'Sullivan & Shot-Based Analytics
    February 9, 2017, 2:41 PM ET [7 Comments]
    Sheng Peng
    Los Angeles Kings La Los Angeles Kings Blogger • RSS • Archive • CONTACT



    It might be hard to stay optimistic about this Kings season after back-to-back 5-0 beatings.

    If you're still optimistic, as I am, this is essentially what you're hanging your hat on: 

    1) That the strong Drew Doughty/Jake Muzzin/Alec Martinez defensive foundation from last season has been improved by youngsters Derek Forbort and Kevin Gravel. Somewhat related -- Los Angeles is showing well in Scoring Chances Against/60, their best in this area since 2011-12.

    (This is using Corsica's definition for the stat, which roughly emphasizes opportunities from high-danger areas, such as the middle and low slots.)

    Los Angeles Kings Capture_zpsn6qt2zqw

    2) That Peter Budaj and the hopefully returning Jonathan Quick can continue to provide average to above-average netminding. Even after his recent shellackings, Budaj's +0.91 5v5 Adj.FSv% stands 11th out of 36 qualified goalies (1000+ 5v5 minutes).

    (Adj.FSv% stands for Adjusted Fenwick Save Percentage, which “shows if a goalie is saving more or less shots than an average NHL goalie would be expected to save if they faced the same quality of unblocked shots.”)

    3) That the recent scoring form shown from Marian Gaborik (five goals in his last 15, team-leading 11.17 5v5 Individual Shots/60 in that stretch) and Anze Kopitar (3-13-16 in his last 16) keeps up, and the returning Tyler Toffoli gets hot.

    My belief that Los Angeles makes the postseason remains unchanged. But I said the same thing about the 2014-15 Kings, who I still believe to be the best non-playoff team in recent years. Indeed, a lot can happen on the way to a Cup -- or to the putting cup.

    ***




    Patrick O'Sullivan caused a stir in the analytics community a couple days ago with his comments on the Claude Julien firing. 

    According to Natural Stat Trick, the Bruins are indeed third in the league with 28.8 5v5 Scoring Chances For (LA is fifth). However, while O'Sullivan's assertion that "Boston can't make plays in dangerous areas" might go too far, the same site has another stat which actually supports his argument.

    Despite their lofty Scoring Chances figure, Boston ranks just 15th with a 10.21 5v5 High-Danger Corsi For/60 rate (LA is 14th). 

    How can this be?

    Essentially, Natural Stat Trick has two stats (from War on Ice) for scoring chances: Scoring Chances encompass a wider area -- roughly "home plate" in -- while High-Danger Corsi For is middle slot in.

    For what it's worth, Corsica's definition for Scoring Chances, which like Natural Stat Trick's High-Danger Corsi For is based on high-danger area opportunities, is tougher on the Bruins -- their 7.02 Scoring Chances For/60 is 22nd, while the Kings are 12th. 

    In short, both play-driving giants Boston and Los Angeles -- ranked one-two respectively in 5v5 Corsi For % -- can stand to get more work in the dirty areas. 

    ***

    Speaking of O'Sullivan, reader @siptico asked what I thought about the ex-King's comments on analytics during TSN 1050's Leafs Lunch on February 6th:
    There's private companies that do analytics.

    The analytics you see on the internet? That's shot-based.

    You talk about possession time? Well, they do actual possession time. How long is the puck actually on your stick in the zone? They do it individually, they do it for teams. They look at pass completions, cleanly exiting your zone.
    [size]


    He double-downed on these statements later:




    There's a lot to talk about here. 

    First, it's no secret that there are "private companies that do analytics." Prominently, SPORTLOGiQ claimed to be working with 10 NHL teams as of last season. There's also Stathletes, co-founded bycurrent Arizona Coyotes General Manager John Chayka. And indeed, both's stats go beyond shot-based.

    [/size]
    While the NHL regularly collects data on shot attempts during games, Sportlogiq does far, far more, says Christopher Boucher, the manager of its hockey analytics department.

    "Our system... tracks every pass, deke, dump-in, dump-out, shot, blocked shot, blocked pass, stick-check, body-check, deflection, and includes a time stamp, player ID and XY coordinate for each of these events," says Boucher.

    "This allows us to quantify each player's impact on possession. In other words, while traditional hockey analytics focus on the result — shots — our system breaks down the process that leads to all shots."
    [size]


    None of this, however, makes shot-based stats "garbage." Beyond their accessibility, another beauty of shot-based stats are sheer volume -- small sample sizes aren't as much of an issue with these figures. That, and its degree of correlation with winning -- beyond other publicly-available stats like hits or faceoffs -- make shot-based stats at the very worst -- at the very worst -- debatably valuable. They're in the conversation -- and deservedly so.

    Totally disregarding them is folly. 

    That said, there's nothing wrong with admitting shot-based stats have their limitations. What they are, we're still in the process of figuring out. They're a piece of the puzzle -- the question is how big?

    They aren't be-all, end-all, and I don't think they were ever meant to be.

    The same could be said about the micro-stats that companies like SPORTLOGiQ are tracking -- all pieces of the puzzle -- but how big?

    No one knows yet.[/size]
    • Back to top
      Current date/time is Tue 26 Nov 2024, 3:44 pm