Hockey analysis has really grown over the past few years on the Internet. Various writers, contributors, commenters, and others have developed and refined new concepts and statistics to measure how well a team or player is doing. If you've read hockey blogs in general, you've probably come across them and resources like Behind the Net, Hockey Prospectus, and Hockey Analysis. While the original posts and discussions are out there somewhere, it can be daunting to pick up on new discussions, conclusions, and findings with respect to advanced statistics (which is a misnomer since most of it is just counting different events). And that's even if you're familiar with the writers, blogs, sites, and so forth like I am. One of the areas lacking in the "hockey analysis community" is a comprehensive resource that really explains what's been done. While all kinds of posts and explanations have been given, it's really helpful to just have one that is both useful for the fan who's realizing there's more to it than just points and someone who has a good grasp on all this
Rob Vollman has taken on that challenge and has met it for the most part with Rob Vollman's Hockey Abstract. In the interest of disclosure, I must point out that Vollman gave me a PDF copy of the book for review purposes. I'm also working on the Hockey Prospectus 2013-14 annual (three guesses as to what team I'm working on), a project that also includes Vollman. Lastly, I've written one post at HP, a site that also features Vollman's writings. That all said, I have read the book and will judge it on it's own merits.
As Vollman writes right in the introduction, this is intended to be "a fun and informative book, not a text book." In that regard, he answers common questions fans have such as "Who is the best defensive player?", "Who will finish first next year?," and "What is the key to winning?" with several of these developed metrics. In this way, Vollman highlights both the purpose of the stat and how it can be used. After several of these that make up about half of the book, Vollman goes into explanation of other stats and further details on those brought up earlier. For someone unfamiliar with advanced stats, I think that is an especially good way Vollman is also ready to note the shortcomings in these arguments and metrics. He is also fairly careful to point out that he doesn't have any definitive answers. As he also stated right in the introduction, "I want to refuel arguments - not end them!" He adheres to that goal while still presenting stats and reasoning beyond the simple stats page at NHL.com.
Vollman's Hockey Abstract definitely has enough material in it to do just that. Some more than others make the book worth it all on it's own. In asking, "Who was the luckiest team?," he explained the multiple factors teams and players can't control - what we call luck as shorthand - is comprehensive and a must-read. Vollman's player usage charts are also a standout. Vollman also provides and explains updated league conversion rates, building off the original work provided by Gabe Desjardens. Anyone interested in determining how the production from a prospect or free agent outside of the NHL compares should find that to be vital. The only quibble I have with it is that the CCHA's rate won't be useful as that league won't exist in 2013-14. But there's nothing that could be done about that since there's no historical data for the Big Ten or the new NCHC. Again, a quibble; the rest of the section is very worthwhile. Vollman's section on passes, an estimated metric to highlight who's making plays beyond just assists, is also of keen interest for the future. Lastly, the explanations of shot-based metrics are clearly made and Vollman does well to note what they do and do not accomplish.
That all said, it's not a perfect compendium. Unless I missed it (and if I did, I'm sorry, Mr. Vollman), I didn't see anything specific about zone-start adjustments for Corsi and Fenwick. It's a strange omission considering it's not an unknown concept for shot-based metrics and the concept is pretty simple. It's touched on in the Player Usage Chart section, but I think the concept demands more than that. A chart of who's been best by adjusted on-ice Corsi rate over the past five seasons would have been useful to highlight it's effects. In addition, I definitely didn't see much in the way of equations for some of usages of metrics. I understand Hockey Abstract isn't meant to be a text book, but I'd rather have seen one or two to go with text explanations for how some of the metrics are made, such as the league conversion rates or zone-star adjustment for Corsi and Fenwick. That may be a personal perference, though. While most of Vollman's questions are pretty well reasoned, I felt the response to "Who is the best coach?" was fairly weak as it focused on expected results without including the results of the team in shot-based metrics, something a coach would have a hand in by way of using players and instructing tactics. Vollman did admit that it's only a first step, though. In Vollman's final section, where he identifies "Do it all" players, I'm not sure why he included grit by way of hits and blocks, especially after noting how they are subject to scorer bias. It's his definition, but I wouldn't have included it if it were up to me as I would interpret such a player as someone who plays well in all situations. I also felt there was a bit too much of an emphasis on GVT and GVS in spots, but at least all were explained as well as he could. Lastly, if you've been aware of all of this online, then you may not learn much that is new; but the book does work as a refresher.
If you do decide to get it, I want to recommend the PDF version over the book. Not only is it currently cheaper, but the hyperlink references in the book can actually be used. Vollman appropriately references and acknowleges where metrics, stats and concepts have come from by other writers and sources online. This is not only the right thing to do, but it's good for the reader should the reader want to read more about it. The URLs do work in the PDF for easy access. It won't in the book. I know Vollman tried to make an e-book version but the sheer number of charts make one untenable. I would suggest, as a compromise, that Vollman put up just an electronic list of the referenced sites for book-based readers to still easily access them somewhere. Is it a deal-breaker? Absolutely not, but it would throw them a bone.
In general, I certainly do recommend Rob Vollman's Hockey Abstract. I've been been dabbling in advanced stats as a user for years and I got plenty of out of it. If you've been reading sites like this one but aren't all that clear about them, then Vollman's book should be very useful. If you have even a passing interest in trying to understand the game beyond the scoreline and what you see, then you'll get a lot of out of it. If you just want to see what he thinks the Devils might do in the coming season, then you absolutely want his expected finish results to come true. If you want a book about hockey that goes beyond the usual fare, then you'll get what you want. It is by no means a perfect compendium, but it does a very good job at hitting many of the major points without being incredibly dry or uninteresting. I would say Vollman met his goal of writing a book on hockey stats that was informative without being a textbook or squash any further questions we continue to ask. If he didn't give me a review copy, then I would've purchased it myself. Based on the links at his own site, appropriately called Hockey Abstract, you can purchase Rob Vollman's Hockey Abstract as a book from Amazon and as a PDF from Payloadz, which I recommend.
I'd like to thank Rob Vollman for providing a review copy and I'd like to thank you for reading. Please let me know what you think of the book - whether you've read it, plan to read it, or be interested in it - in the comments.