This new, regular feature on NHLNumbers will share interesting stats-related posts from around the web every week.
- First up is a non-hockey article from Deadspin. Confessions of an NBA scorekeeper looks at how certain results in pro basketball are “fudgible” and how even earnest or well-meaning stats counters can be influenced to favorably adjust things like assists or blocks for a certain team or player on any given night.
The NHL has similar problems when it comes to accounting for shots on or at the net (although a dubious second assist sometimes slips through the net as well). Back in 2009, Sunny Mehta showed how the Devils shot counter totals only seemed to have a passing resemblence to reality. Vic Ferrari and Tom Awad also detailed the difference in shot counting between teams across the league that same year.
Gabriel Desjardins did a systematic take-down of Madison Square Garden’s stats counter three years ago as well. The problem with the Rangers home counter is actually shot location – with shots being counted sometimes 15-20 feet closer to the net than they actually were. Gabe followed up a year later to see if the issue was persistent. Alas, pucks were still magically being shot from a lot closer to the net than anywhere else in the league.
Aside from not being able to fully trust shot totals from certain rinks, these biases affect measures like corsi, fenwick and any study seeking to determine shot quality or shot independent goaltender ratings. As a result, it’s usually a good idea to restrict large scale investigations that including shot/block totals and locations to away locations because the assumption is the various biases and errors will more or less cancel each other out. It cuts the sample in half, but until the league can improve the quality of its data there isn’t much of an alternative.
- Next up is an interesting new stats developed by SnarkSd of Fear the Fin: P+Corp (points + corsi over replacement level). Snark describes his measure as
At it’s heart it’s a simple stat that serves as a good substitute or comparison piece to the current gold standard of player evaluation, GVT. The beauty is in it’s ease of calculation, which requires looking up only 3 (or 4 for Defensemen) stats from the main Behind The Net page. My underlying assumption is that ice time increases with line status, ie. top lines are given the most ice time, but everything else falls out from detailed data collection. I didn’t fine tune these numbers so the best players are at the top of the list, this is evidenced-based hockey analysis.
… In addition, it is well integrated with replacement level. A score of 0 will always be our baseline replacement level talent. As P+CORP increases (and theoretically decreases) talent level increases as compared to (ie. over) replacement level.
The explanation and discussion is quite detailed, so make sure to take look. It is a great attempt at creating a new, integrated stat that can measure players versus replacement level.
- Hockey Prospectus and FlamesNation contributor Rob Vollman recently released leaguewide player usage charts. Also called Ozqoc charts, these chart a players usage by looking at his individual quality of competition versus zone start ratio. At a glance they tell you how easy or difficult each players assignment was last year.
Related: Thomas Drance recently looked at Alain Vigneault’s deployment of the Canucks roster this past season through an Ozqoc chart and compared his tactics to the rest of the league. The extreme difference in circumstances between the bottom and top of Vancouver’s roster becomes readily obvious in this article.
- Finally, Gabriel Desjadrins re-vists the Winnipeg Jets decision to take Mark Scheifele over Sean Couturier in the entry draft last June. The decision was a head scratcher at the time and seems even less defensible one year out. Bonus helping of regret and self-doubt for the Jets scouting staff: the massive season by Sven Baertschi, who was taken a few picks alter by the Calgary Flames.
If you have any stats-related links of interest, fele free to share in the comments or send them to Kent.Wilson@gmail.com for future inclusion.