I recently had a chance to do a quick Q&A with Eric Seidman – author of the blog of Statistically Speaking – regarding his new book: “Bridging the Statistical Gap.”

Here’s what we covered:

WW: If you had to explain to someone in five sentences or less what your new book was about, what would you say?

ES: I would say that “Bridging the Statistical Gap” is capable of being a casual fan’s first foray into sabermetrics or statistical analysis. I do not cover every single statistic or method known to man, but a first foray doesn’t need to do that. This is a book that will pique the interest of someone consistently using wins and losses or batting average as strong evaluative barometers and teach them the history of these numbers, why they do or do not work, as well as how to find numbers out there capable of telling them the desired knowledge. It is a book designed to show fans intimidated by the numbers that they aren’t necessarily as scary and that they already understand most of this stuff whether or not it is realized.

WW: The first chapter of “Bridging The Statistical Gap” (entitled “The Great Batting Average Debate”) does give some history and definitions of sabermetric tools. How did you decide which to include (like Weighted On Base Average) and which ones to leave out (like Offensive Winning Percentage)?

ES: That first chapter is a great example of how I wrote the book, in that I take readers on a logical train of thought to arrive at a given destination. In that chapter we started by looking at batting average and determining that all hits are counted as singles, and walks are not included. Then we look at on base percentage since it included the walks but realize it’s still counting hits as singles. Next comes slugging percentage, which differentiates between the hits, fails to include walks, but actually overcompensates in its differentiation. There’s a discussion of stats like Isolated Power, Equivalent Average (or “EqA”), and others as well, that brings us to an endpoint that the best type of metric capable of being a barometer at measuring offense in a batting average type form would be one that properly weights all of the numbers and scales it similarly to batting average. This essentially exists in the form of Weighted On Base Average. If the book were a strict definitions book, Offensive Winning Percentage and others would be included, but a casual fan has no idea what Offensive Winning Percentage is, and it wouldn’t make much sense to include it given the context of the chapter.

Plus, even though I find it interesting, I can’t see myself engaging in a conversation involving the line, “Yeah, Todd Helton’s Offensive Winning Percentage is .750, meaning that if he took up every spot in the lineup, his team would win 75% of the time.”

WW: In the book, you introduce your “Starting Pitcher Effectiveness Model.” If you had to describe this model, at a very high level, what would you say?

ES: Well, I had actually introduced that a while ago at Statistically Speaking but I got such a favorable response from more casual fans (who were turned onto it by some fellow readers at MVN) that I felt it was important to include, much more in-depth in the book. Realistically, the method is not an end-all by any means, and there are plenty of other, better metrics out there, but it takes a bunch of stats that fans can find on the first-tier of numbers at baseball-reference.com and weights them to come to an overall total of points. And based on the points accrued it determines if you meet the benchmark for a #1, #2, #3 SP, etc.

WW: Are “Park Factors” or “Bequeathed Runners Allowed To Score” or “Strength of Opponent” included among those variables?

ES: These three factors are not included in the system. Bequeathed Runners is one I beat myself up over a lot, but Park Factors or Strength of Opponent were never under consideration because, again, the book is not designed to include every single aspect of statistical analysis, but rather show numerous examples of it through actual conducted analyses rather than just the “for dummies” definitions/teaching format. Strength of Opponent is taken into account somewhat in the tough losses section, because some pitchers will consistently pitch against the other team’s ace, which could hinder their offense’s ability to support them throughout the game. Ultimately, though, if Bequeathed Runners gets a strong enough fanbase I have no problem including it. The thing to remember is that this isn’t an official number but rather something for those not statistically inclined to get their feet wet with.

WW: Having developed the model, running the numbers, and seeing the results, what one finding did you derive in the end that surprised you the most?

ES: The finding that surprised me the most was the 2001 season for Mike Mussina and Roger Clemens. Clemens won the Cy Young Award that year, but, in the Starting Pitcher Effectiveness Model wasn’t even in the top five. Mussina led the AL with a +77, 32 higher than Clemens’ +45. On top of that, Mussina had a better ERA and Fielding Independent Pitching [mark], and his Adjusted W-L record, a metric that adjusts the decisions received to measure the number of good games versus bad games, turning tough losses into wins and cheap wins into losses, Mussina was 21-7 whereas Roger was 14-9. All told, Mussina had a better Adj W-L, Starting Pitcher Effectiveness Model total, ERA, and Fielding Independent Pitching, yet did not win the award.

WW: Speaking of Clemens, your book also shares a formula that you’ve come up with for determining the best post-season pitchers. How does that method work and were does Clemens rank in your findings?

ES: The playoff chapter offers two different scenarios: results with all starts included and results relative to games started. Clemens, and Pettitte for that matter, both fall into the first category because they have made so many playoff starts. Their overall playoff scores rank quite highly, but when we look at score relative to games started, which I termed relative success, they aren’t in the top ten, meaning they were of benefit from having so many opportunity. However, and it’s an important however, it doesn’t mean they weren’t “good” in the post-season, it just means they had many more opportunities. What’s often lost on fans is that more opportunity works both ways: it offers more room for success, yes, but equally more room for failure!

WW: There’s a section in your book where you discuss clutch and meaningless stat-padding performance. Could you share a high-level summary of what that section is all about?

ES: Clutch hitting seems to have been tackled by everyone out there in many different ways, shapes, or forms. This chapter essentially is a high-level summary of all of these studies, and the general idea or debate surrounding what clutch hitting is, if it’s quantifiable, if it’s important, and why it’s easy to fall victim to the availability heuristic in that what we see on SportsCenter is the complete truth; so, while ESPN has made David Ortiz out to be a clutch hero, the last couple years he ranked towards the bottom of the AL according to certain clutch metrics. The chapter discusses the ideas of transient phenomena and persistent phenomena, with the former serving as one that disappears after a bit and the latter holds strong(er) year to year. Meaningless stats seem to be something tackled less, and to me it’s even more important. For instance, people always knock A-Rod for a lack of “clutch” performance, but that implies that he pads his stats when the game isn’t on the line. Well, we can check this now! For all we know it might be that he performs extremely well in medium-pressure situations, and that his below performance in clutch situations, while worse for A-Rod is still better than many people.

WW: So, without giving too much away, how does A-Rod do medium-pressure spots? Does he pad his stats?

ES: Well, looking at his OPS, for his career, A-Rod has actually performed better in high leverage (important) situations than in all others, which seems a bit counterintuitive given his reputation and his rank towards the bottom of the ‘clutch’ stat kept on Fangraphs.com. His stats are ridiculous to begin with, but seems to be 50/50… he isn’t nearly as bad as people make him out to be in the clutch, at least looking at his career as a whole — I’m guessing his playoff performance in recent years have worsened this reputation—but he has hit a lot of meaningless home runs, which I’ve termed home runs when the win expectancy is 90% or higher for either team.

WW: Is the one thing included in your book that you’re most proud of – above everything else?

ES: To be honest I really like the fourth chapter of the book, which discusses scouting, splits, small sample sizes and standard deviations through the most in-depth analysis of Michael Jordan’s 1994 minor league season ever concocted. The Birmingham Barons literally sent me photocopied scorecards for every game that year and I was able to construct Jordan’s splits in a plethora of categories. While there are of course things you wish you could go back and redo, as is with any venture, I’m quite proud I was able to get this done and the feedback I’ve gotten from many readers has made it worthwhile. From here on out I plan to get more involved with historical books, as I’m currently writing a book on Bucky Walters (former MVP of the 1939 Reds) but I’ll always look back on another proud feat of mine, which is learning the process of writing a book. It’s not easy… at all. So many things to keep in mind and staying sane with those burdens is tough.

That’s it. My thanks to Eric for his time on this and providing some more information about his new book: “Bridging the Statistical Gap.” If you have any questions on the book, feel free to post them in the comments section below…and I’ll ask Eric to address them as soon as possible.