March 10 2013 08:43AM
Photo by Michael Miller, via Wikimedia Commons
Half-way through the season and not much is changing in the predictive power rankings even though we're still adding to the model. One item of note is the race for 8th in the Western Conference. If Kari Lehtonen is healthy, Dallas is clear of the bottom portion of the conference for the last playoff spot. Minnesota, Calgary and Phoenix are clumped together behind the Stars as the 9th, 10th and 11th best teams in the West. Even though the Wild aren't better than last year's team, and Calgary is a mess in goal, they have an advantage over the Stars and Coyotes - they play each other and the awful Oilers and Avalanche a combined 7 times before the season is over.
These rankings aren't presented as a look at the current standings or last couple of weeks' worth of performances. Our goal is to build a predictive model that gives us a glimpse into the season-ending standings and first-round playoff matchups. Our rankings aren't based solely on the current NHL standings, though points earned are a significant portion of the model, they are based on a number of underlying metrics so the rankings are constantly in flux. This week we've added remaining opponents in the hopes of building a better model.
We're still tweaking that model, so we're not quite ready to unveil it, but consider this NHLNumbers' current best effort at predicting the standings and playoff pairings at the end of the season. One item of note - 38% of the game is luck, and we don't attempt to predict or model that 38%, and we don't plan to.
March 08 2013 03:52PM
Yesterday I discussed some of the entrenched misperceptions about statistical analysis in hockey and why they are misguided. Today, in a post called "Hockey's Counting Problem", Cam Charron looked at some of the significant obstacles standing in the way of effective, evidence-based analysis truly getting a foothold in the NHL's upper offices.
Two challenges he touches on that I want to discuss in greater depth are top-down vs bottom-up processes and the clash between long-term decision making and short-term incentives.
March 07 2013 01:10PM
A couple of things running through my head between Flames games today. The first topic has to do with hitting and winning versus shooting and winning. The second topic is on how advanced analysis in hockey is gaining prominence (but is still obviously misunderstood).
March 07 2013 07:31AM
Photo by leech44, via Wikimedia Commons
Seth Jones is likely to be the first overall pick in the 2013 NHL Draft. Jones has pulled away from Nathan MacKinnon in the NHLNumbers Conensus Top 100. The Scouting Report is included in those rankings and has Jones at #1 overall:
An excellent skater especially for his size, Jones is a rare combination of athleticism and skill. More than capable of playing in the NHL next year, Jones will be a quality transition defenseman with top 4 capabilities from day 1. A strong slap shot and good hockey sense will allow him to also see powerplay minutes early in his career.
Though he'll be the #1 overall pick, he'll be #2 in the NHL. After Jones is drafted by the Blue Jackets, Oilers or whatever team wins the new lottery, he will suit up for his first NHL game sometime in October. When he does, Seth Jones will become only the second player named Seth to ever play in the NHL. Seth Martin, a goaltender for the St. Louis Blues, played 30 games in 1967-68. Jones will surpass that games played mark some time before Christmas.
March 06 2013 12:31PM
There's been some talk about the importance of hitting in the hockey blogosphere lately. Most of it has come on Twitter in the wake of the Edmonton Oilers trading for Mike Brown and to whether or not Brown's hitting will make a lick of difference to the Oilers lineup.
My gut reaction is "no, no it won't". I did a big of leg-work in this post at the Backhand Shelf that showed "hits" "giveaways" and "takeaways" were subjective statistics that didn't really mean anything. Every scorer in every building has a different definition of a hit. What I wanted to see was whether actual hits lead to turnovers in any given game.