Win Shares

By Bill James & Jim Henzler
Reviewed by Paul Wysard

Before the Red Sox hired 28-year-old Theo Epstein as General Manager, before the ensuing sports page cartoon pictured the young exec proposing trades in a nursery school sandbox, the ancient franchise made an even more unorthodox appointment. It employed analyst, writer, and sabrmetrics specialist Bill James as a consultant. Maybe this is inevitable in our maturing "cyber-age," using computer-generated statistics to evaluate players and to quantify roster needs. Epstein himself is a devotee of the approach, and why not? To get a possible edge in personnel, it is thinking "out of the box" --- in this case, that sandbox.

One of the tools James will surely bring to the workbench is his concept introduced in his 2002 Baseball Abstract and published in more detail several months later with the assistance of Jim Henzler: Win Shares.

There is no way this reviewer can adequately or comfortably explain all of the aspects of this new methodology. Formulas abound, many of which I don't understand. There are some seemingly arbitrary numerical standards and ratios, from which much else flows, but which are not fully explained. A lot of the related calculation reminds me of algebra, where I obviously did not excel; to wit, Student: "Why do we do that? How was that figured?" Teacher: "Never mind. It's a rule."

Having said all of the above, the whole idea, it's myriad of supporting data, and its recollections of good and familiar players over the decades --- "old friends," as it were-- are both fascinating and entertaining. James does not look upon this new work as a final answer, but as another way to measure player performance, a method which took years to develop and about which he feels justifiably confident. Other evaluation systems, mostly linear, place players along a spectrum, from high-plus, to plus, to zero (about average), to minus, to low-minus(doesn't belong). James contends that no player who is in the game, say, 100 times per season, is a "zero." Such players are clearly not Ruth or Mays or Maddux but each contributes and therefore has some value. That value is determined by his contributions ("shares") in the primary goal ("wins") of his team.

Following is a very elementary primer on the system, with apologies to James and Henzler if I miss any critically important elements. Wins are the reason the game is played and runs are the building blocks of wins. So step number one is to identify and isolate marginal runs. These are all runs scored by a team in a season minus 1/2 the league average AND all runs allowed by the team in a season less than 1 and 1/2 times the league average. The more marginal runs earned, and the fewer allowed, the more wins there are. Then runs created and prevented are calculated through several other moves and formulas, including outs made by each hitter. Wins are determined, and 1/3 of those are batting win shares. Players who create the most runs receive most of the batting shares. The opposite is in force for pitching and defense. Finally, ball park effects have to be considered; Coors is hitter-friendly, Jacobs is somewhat hitter-friendly, Turner Field and Dodger Stadium favor pitchers. The authors have appropriate calculations for these factors, spanning five-year periods.

The best way, I discovered, to understand all this is to take all the steps through a so-called "monster year" by a player, in this case Barry Bonds in 2001. All of those walks and homeruns catapulted him away from all the others, and the system is clarified. Bonds amassed 54 win shares --- simply off the charts --- as players with 40 are either strong MVP contenders or have won the award. He followed that up with 49 this past season, joining some other huge totals over the past half-century or so: Musial with 46 in 1948, Williams with 49 in 1946, Mantle with 51 in 1957.

Career Win Shares leaders will not surprise anybody. Ruth had a great pitching record added to the hitting, so he leads with 756. Cobb's superb fielding and base-running complemented his hitting at 722. Walter Johnson was a pretty good hitter and that helped him supplement the many wins and low ERA to reach 560. Bonds has now barged into the top ten, all-time, with 572. Rickey Henderson looks absolutely excellent at 534. Recent Hall of Fame inductee Eddie Murray is in the top 35 with 437.

There are many more ingredients in this vast recipe, including some that will seem esoteric to the average fan, i.e., Component ERA, Claim Percentages, Excess Batters Faced by Lefties. Why does the latter make a difference? Because right-handed hitters will hit more balls to third and short off of left-handers, thereby increasing the chances for assists and put-outs in those locations.

Interesting analyses and comparisons appear throughout the book. Of the many, here are just two: (1) Buckner vs. Garvey. The latter has many more unassisted put-outs because he was hesitant to throw, even at shorter distances. The former was a stickler for his pitchers covering first to receive his tosses to register outs, and (2) Ashburn and Hamner. In their time, and shortly thereafter, both were thought to be great fielders. Ashburn often led the league in put-outs in center, but won no Gold Gloves. Hamner later was shrugged off because his fielding numbers, especially double-plays, were far behind other middle infielders. Is there an explanation for the situation? James and Henzler are certain there is: The Phillies staff in those '50s years, led by Robin Roberts, were fly ball pitchers, hence lots of easier outs for Ashburn, far fewer ground balls for Hamner.

The reader is presented with much discussion and many charts regarding Win Share totals in MVP and Cy Young awards. Many times the two do not match. One of the largest gaps occurred in the case of Terry Pendleton with the Braves in 1991. He had 29 Win Shares that year, as much as 10 or 12 ( a lot) behind usual or normal MVP awardees. Yet the voters saw some things that numbers can't show us --- leadership and the mentoring of inexperienced players, to name two. The authors are well aware of these intangibles, but nevertheless believe their system helps to solidify opinion about who is or was the best and to give, or reduce, credit, where it is or is not due. Nice reading and a compelling argument to spice up these winter months.