Innovator of Quality Starts, cap-measurement tool GVS (Goals Versus Salary), OZQoC charts to graphically represent player roles, the Snepsts System of finding similar historical players and making projections, and advances in the field of league translations and equivalencies (e.g. AHL, KHL), Robert Vollman happily makes all his data available for public use.
KHL goaltending statistics directly from the KHL's official website. Includes all goalies from 2008-09 until December 11, 2013 (midway through 2013-14 season). There is a second tab with the combined totals for each goalie.
250+ stats on all 839 players who were in even a single game in the 2012-13 season.
An annual tradition for almost a decade, this spreadsheet contains a collection of statistics, both basic and advanced, from a wide variety of analysts and web sites including Hockey Reference, Gabriel Desjardins' Behind the Net, Tom Awad, Cap Geek, NHL's official feed, and much more.
As always it includes biographical information, the basics, useful metrics (like zone starts, penalties drawn and percentages), advanced Corsi-based statistics, and established stats like GVT and Point Shares - all broken down by situation.
Check back often as additional information is always added as they become available.
(Aug 11, 2013): Updated with new statistic (Net Penalties), and corrections to GVT and offensive zone start percentage.
(Aug 15, 2013): Added TOI QoC (Quality of Competition based on average ice-time of one's opponents) and SPD (shooting percentage differential, a re-presenting of PDO).
This is your annual super-stats spreadsheet for goalies. It features 71 pieces of data for all 82 goalies who saw even a single minute of action in 2012-13.
As usual, it includes biographical information and basic statistics, broken down by manpower situation and also by whether the goalie was starting or coming in relief. It includes shoot-out data, penalty shot data, and even contract-related data like cap hit.
It also includes advanced statitics like Tom Awad's GVT, Justin Kubalko's Point Shares and Robert Vollman's Quality Starts. Check back occasionally for any updates.
When Hockey Prospectus first launched
in 2009 (as Puck Prospectus) the very first article was about
translating scoring totals from the Russian league to the NHL, and was
quickly followed up with a piece that looked at the Swedish Elite
League (elitserien). Each piece used a system pioneered by Gabriel
Desjardins in 2004, borrowed from baseball’s Bill James, to
calculate the average change in scoring totals for all the players
who came to the NHL after playing in other leagues.
This type of work depends on having a set of data that includes the raw data for everyone who played in another league, like the European, US College or Canadian junior leagues, the year before playing in the NHL. Using the attached spreadsheet allows everyone to conduct their own analysis with NHL translations.
A good coach will make a bad team competitive, and a good team great. While it is difficult to rate coaches, a good start is to measure how well their team did relative to how well they did the previous season, once regressed to the norm by 35%. With sufficient data we can start to see which of history's coaches might have been the most effective.
It probably comes as no surprise to see Scotty Bowman far, far atop the list, with Ken Hitchcock right behind as the league's most accomplished active coach. It's also interesting to see how much better non-NHLers did as coaches relative to those who had played in the NHL, or how among those who played in the NHL it doesn't seem to make a difference if they made the Hall of Fame, but does seem to matter what position they played (defensemen make better coaches).
Download the attached data and make your own calculations, perhaps restricting the study only to the post-expansion era, or changing the degree of regression (possibly by era).
Rob Vollman brought in the new year with a new statistic: passes. Passes are defined as those passes which result in shots on goal, and can determine which players are most adept at setting up shots in a way that's independent of shooting percentages. Unfortunately this statistic is available only as an estimate based on the player's assists and the average shooting percentage of his linemates.
One spreadsheet has the original estimate for the past four seasons, while the one stretched all the way back to 2007-08 is refined to look only at primary assists, and breaks them out by manpower situation (even-strength, power play or shorthanded).
NHL players went to a variety of different leagues during the 2004-05 lockout. For every NHL player that played somewhere during the lockout, and immediately came back to the NHL - even if just for a single game in either case, attached is a spreadsheet of their basic statistics (GP, G, A, PTS) for each of those seasons.
If the NHL is once again unfortunately locked out in 2012-13, ideally the attached information can be used to help predict how well these players will score in those other leagues, and how they might fare upon their return.
The system of searching NHL's history to find statistically similar players for a variety of purposes, most notably future scoring projections, began in 2009. In the grand tradition of naming such systems after rugged former NHLers, it is informally known as the Snepsts system, but is unaffiliated with Harold Snepsts (who nonetheless finds it amusing that the system is named after someone whose scoring was so modest).
Throughout the 2011-12 season the system was further refined with the assistance of several people, most notably Neil Greenberg, and among other improvements, it most notably includes only players from the post-expansion era (1967), and looks at shots, power play scoring and GVT in addition to the goals and assists it did previously. It is therefore informally known as Snepsts67.
The method is reasonably simple: it searches post-expansion NHL history for players of a similar position and age whose era-adjusted statistics most closely match the target player, both for the current season, preceding season and previous career totals, in a 4:2:1 ratio.
The ten most similar players are used to set low-water, high-water and average scoring expectations for today's players, and also to initiate interesting conversations about a player's role and who he should be used. Attached is a simple spreadsheet with this data for every NHL player who played at least a single game in each of the preceding two seasons.
NOTE: Viewing this file in your browser will limit you to just the first 1000 rows, so please download the file to access all the data.
Your one-stop spread sheet for all the data you'll need July 1st to evaluate the available unrestricted free agents. It includes a targeted version of all the data available in the annual super-spreadsheet for every single available unrestricted free agent, divided by defensemen, forwards and goalies.
This edition also includes two extra tabs that put all the available forwards and defensemen onto their own Player Usage Charts so you can see which players can fill which roles.
Please check back regularly as it is updated with every signing - the first column should show you their latest status, and only the available free agents should appear on the usage charts.
Player Usage Charts were first introduced in the 2011 off-season as OZQoC Charts after their two primary components: Offensive Zone Starts (OZ) on the horizontal axis and Quality of Competition (QoC) on their vertical axis.
Player Usage Charts were quickly adopted by dozens of statistical analysts (here's a list
of the earliest adopters), and even some front offices, most of whom also made various improvements. Naturally the biggest improvement of all is the new name!
The largest contribution to Player Usage Charts was undoubtedly made by Eric T of Broad Street Hockey, who not only suggested the Corsi bubbles, but also created a new way of measuring the quality of one's competition: their average ice-time.
Attached are the Player Usage Charts for every NHL team in 2011-12, including expert analysis in the PDF, and the raw data so they can be sculpted as everyone desires (unfortunately the graphs in the spreadsheet will ONLY format correctly in Excel 2010, and will otherwise look like a mess).
Nov 2012: Also attached are the Player Usage Charts using Eric T's new way of measuring quality of competition (TOI). No analysis included in this version.