Book Excerpt: Management By Baseball
The following is an excerpt from Management by Baseball: The Official Rules for Winning Management in Any Field by Jeff Angus. For more information on the book, which was released this week by Harper Collins, click here.
Unquestioned Assumptions: Return on Equity is the RBI of Business
Some business executives manage their operation for ROE (return on equity). The experience of these folks offers an excellent example of missing the point even when they have accurate data (Chapter 6), because they can achieve a target that doesn’t advance their operation. They have an analog in baseball: teams that value players based on RBI (runs batted in).
ROE is a perfectly logical-sounding statistic that encourages its devotees to distort their investments in ways that undermine the business’ vitality and long-term survival. RBI is a perfectly logical-sounding stat that encourages many general managers to pay extra millions for ordinary players based on those G.M.s’ inability to step back and examine their implicit faith in “the way it’s always been.”
ROE is one small measure among many, it’s highly contextual, and it may mask more important factors. An exec can increase a company’s ROE in a dozen ways that will undermine a company. If I cloned myself and managed two companies with everything identical except “Company A” had less equity as a result of some poorly chosen spending and “Company B” was storing cash for some prudent investments, the clone who managed “A” would have better ROE than the “B” clone. Not because of skill or ability to advance the company, but simply because of the “benefit” of having less equity.
The RBI is parallel, a highly contextual measure, and it may mask more important factors. The most prolific homer hitter in the majors the last five years (2001 – 2005) Alex Rodriguez is a great RBI man. The Yankee clean-up hitter has 100 or more RBIs in eight consecutive seasons (1998-2005). Awesome. But he’s a clean-up hitter, so he usually has the three other Yankees best at getting on base batting in front of him. When he comes up, there’s a better chance of him hitting with men on base. He has many runners who represent RBI chances.
Let’s test opportunity. Bat Alex clean-up and clone him thrice. Bat his clones leadoff, fifth and eighth. In the first inning of every game, Leadoff A-Rod is the first one to the plate, so once each game he’s guaranteed an at-bat with no one on base to knock in. His next time up he’s likely to bat behind the #9 batter (usually the least-skilled in the line-up), incrementally reducing the probability of runners on base.
Batting-Fifth A-Rod isn’t quite as hosed But the guy who bats in front of him (the Original A-Rod) is the most productive home run hitter in the majors over the last four years, and when he’s whacked one over the fence, the fifth hitter is coming up with the bases empty. In fact, Batting-Fifth A-Rod, by being a very scary dude to pitch to, is going to see fewer baserunners during the season because pitchers aren’t likely to intentionally walk the clean-up hitter to face Batting-Fifth A-Rod.
Batting-Eighth A-Rod is dinged, too. The batters in front of him, hitting 6th and 7th, are among the weakest in the lineup, and their strength is likely power hitting, not high on-base average, so they may not be on base when our last A-Rod gets up. To undermine our clone further, he’s going to have the #9 batter, the least-skilled on the lineup card, next up. Of course, this increases somewhat his chance of being walked intentionally, and with men on base, increases greatly his not getting good pitches to swing at, because an opponent is more afraid of being hurt by a scary hitter like Batting-Eighth A-Rod than by walking him and facing #9’s dubious batting skill.
Like a sales team that has regions or account portfolios with highly variable potential, RBI opportunities are not evenly distributed throughout the lineup. Therefore, it’s not a solid measure of helping the team, or “how good” a player is. A batter can do a great job but have fewer opportunities, while another has more opportunities and delivers less, and RBI will reward the less-effective one. Sales managers too often use “gross sales” as a metric the way team executives generally use the RBI: a survival that’s become an article of faith.
You can make useful measures for employees, sales or otherwise, the same way you can make RBI a useful stat. Adjust it for opportunity as researcher Tom Ruane did. One, take the average probability of delivering an RBI in every combination of baserunners-on and number of outs for a batter at the plate. Two, find out how many times that batter has been up in each situation. Three, project how many RBIs the league average should be, and compare his actual RBI count to what the average would achieve in that composite set of situations.
Look at Ruane’s table of the best and worst performers in opportunity-adjusted RBI since 1960. RBI is what you think it is, and ERBI are the number of RBI a league average batter would get in the runs-on-base and outs situations the player faced. Over is the number of extra (or when negative, fewer) RBIs the batter had than that league average. RPRW is the ultimate value, RBI adjusted to the context of the batter’s home park and then converted into the number of extra wins a batter’s actual RBI production would add for his team. So Barry Bonds in 2001 added about nine wins to the Giants’ season over the league average with his RBI production, and Neifi Perez in 2000 cost the Rockies between four and five wins.
BEST TEAM(s) RBI ERBI Over RPRW Barry Bonds 2001 SF N 137 52.3 84.7 9.4 Mark McGwire 1998 STL N 147 65.6 81.4 8.9 Harmon Killebrew 1969 MIN A 140 67.7 72.3 8.5 Dick Allen 1972 CHI A 113 52.0 61.0 8.5 Roger Maris 1961 NY A 141 67.2 73.8 8.4 Sammy Sosa 1998 CHI N 158 78.1 79.9 8.4 Willie Stargell 1971 PIT N 125 59.0 66.0 8.3 WORST TEAM(s) RBI ERBI Over RPRW Neifi Perez 2000 COL N 71 90.6 -19.6 -4.6 Neifi Perez 1998 COL N 59 79.6 -20.6 -4.4 Walt Weiss 1995 COL N 25 48.5 -23.5 -4.3 Ivan DeJesus 1978 CHI N 35 61.5 -26.5 -4.3 Walt Weiss 1996 COL N 48 66.2 -18.2 -4.0 Larry Bowa 1974 PHI N 36 65.8 -29.8 -4.0 Felix Fermin 1989 CLE A 21 53.7 -32.7 -3.9
The uncloned, real-life Alex Rodriguez is in between these legendary successes and failures as an RBI man. Here are his numbers adjusted for opportunity during the period 2001-2004 (Ruane hasn’t yet published 2005 data).
NAME YEAR TEAM RBI ERBI Over RPRW Alex Rodriguez 2001 TEX A 135 84.4 50.6 4.9 Alex Rodriguez 2002 TEX A 142 80.3 61.7 5.6 Alex Rodriguez 2003 TEX A 118 73.9 44.1 3.7 Alex Rodriguez 2004 NY A 106 85.0 21.0 2.2
In A-Rod’s best year, 2002, he led his league in adjusted RBI. He’s been a good performer, adding to his team’s ability to win by knocking in runs.
In baseball, the momentum of the implicit has been hard to overcome. Branch Rickey pointed out ages ago three reasons to ignore un-adjusted RBI, but even a recognized genius struggles to overcome the faith of the lazy-minded. In your organization, there are a breathtaking number of implicit, unquestioned assumptions. A key element of intellectual self-awareness is to question one’s own and one’s employer’s implicit assumptions. Otherwise, one invests in illusions like ROE or whatever distortions hold sway in a shop.
Only when you overcome the limitations of your own and others’ implicit assumptions can you consistently meet and manage change, our inevitable battery-mate.
References & Resources
Copyright 2006 Jeff Angus. All rights reserved. No part of this may be used or reproduced without written permission of the author. Management by Baseball: The Official Rules for Winning Management in Any Field is available online or from bookstores in the U.S. & Canada.