As the Summer of Advanced Statistics winds into what will maybe be the Autumn of Fancy Stats before perhaps turning into the Winter of Analytics just before Nerd Spring, I recently got my hands on a great piece of reading on the topic, a book that Jaime Eisner of Five for Howling called "The Rosetta Stone of advanced hockey stats." Rob Vollman was nice enough to share this year's version of Hockey Abstract with me, even after I wasted two paragraphs in last year's review complaining about a throwaway quip.
Fortunately, I have no such complaints about this version.
New to this year's version are contributors Tom Awad and Iain Fyffe. While both were mentioned heavily in last year's version, this time around they're full contributors and each have written several sections of the new book. This creates several positives for the book. For one, it makes the project feel more open and approachable. While Vollman did a great job last year (and another great job in this version) of referencing existing expertise to bolster the analysis, having extra contributors takes it away from feeling like just one gatekeeper presenting the overall state of hockey analytics. Additionally, the differences in each author's voice adds a bit of refreshing variety.
The separation of the voices to give the book a bit more of a community feel also helps in how Hockey Abstract is laid out. Where last year's book had the feel of a "fun" part and a separate technically heavy part, the way this one is laid out mixes the two ideas extremely well.
Returning are attempts to find the best skaters and goalies in the league at various specialties and the team usage charts I've come to love. These already-strong staples have been bolstered by additional considerations which I feel have added value, especially the team-by-team section which factors in the usage charts, additional context, and even rates each team on 13 categories identified as important to success. Additionally, the new contributions include interesting discussion pieces such as Fyffe's study on what factors lead to a player's induction into the Hall of Fame from different eras. I was also extremely satisfied to see he took the time to show the scores for not-yet-eligible players because I found myself constantly wondering where Nicklas Lidstrom stands by his measures (because any measure that wouldn't have him as a stone-cold lock is one I'd easily dismiss).
Probably one of my favorite parts is the section dedicated to shot quality which takes a very serious look at trying to account for how much more dangerous one shot can be compared to another, which has long been a contentious issue in discussions which include possession metrics used in attempts to quantify player contribution.
The technical stuff has survived the new layout, but instead of being pushed down in the book where it was necessary to anchor the analysis while staying out of the way of the fun, it's now mixed into each section where the conversation would naturally call for such study. I found that the logical progression of the analyses are easy to follow while answering the questions which popped up over the normal course of each section.
From a general standpoint on criticisms against the more numbers-heavy analysis aren't directly covered, but you can read that the authors are aware of such criticisms and sensitive to how they come across. If you can read through Hockey Abstract and come away with the opinion that you're being told their answers are absolute, unquestionable, and make a complete replacement for actually watching hockey games, then you're clearly either too stupid or too stubborn to understand, in which case I suggest you use your money on a certain kind of prophylactic instead of hockey books which hurt your brain.
This isn't to necessarily say that Hockey Abstract isn't strongly written or willing to take any solid stances. There's an entire chapter dedicated to masterfully attacking the dying role of the enforcer in hockey while explaining in great detail why the traditional defenses for having such worthless players aren't compelling. It's just that there's a certain strength to a study on much more multi-faceted and subjective measures being willing to say that there's absolutely room for both disagreement and improvement. The authors responsibly point out what they feel the data can and cannot do without proselytizing around its shortfalls.
Being a Red Wings blogger, I'd be remiss if I didn't comment on the book's treatment of the team. Honestly, it's very respectful of Detroit and perhaps a bit more optimistic than I'd be. Detroit's advanced stats have remained good, despite their struggles in the standings caused by a number of factors (to include questionable roster/lineup decisions). I do feel that injuries probably don't get a fair shake here and that's likely the difference between their level of optimism and mine. Perhaps future iterations will attempt to find ways to build injury prediction factors into analysis.
Overall, I'd say Hockey Abstract is a very good purchase for the die-hard hockey fan who enjoys both the study of the game and its history. A casual fan looking to become a bit more hard core should find a great deal of value out of this book as well. A casual fan who isn't interested in learning probably isn't terribly interested in books and likely hasn't read this far, so I don't have a recommendation for that. You can purchase the entire edition here at HockeyAbstract.com. Both paperback and PDF versions are available for purchase. I had the PDF version to review, but I wouldn't hesitate to recommend the real thing to toss on your coffee table and show off how smart you are.