As a recap, here’s my original piece on PER (or Payroll Efficiency Rating), followed up by some adjustments, and then the first part of this series two weeks ago.

Now, before we get to the rankings, let’s run down some issues that seem to have gotten lost in translation since this system was first introduced in August:

  • How is this not just dollars/win?
    There are two major adjustments factored in here that make it “better” than a simple dollars/win calculation. The first is each team’s unique revenue-earning potential. Since the Yankees make far more money per win than the Marlins do, it makes perfect economic sense that the Yankees should also spend more per win.
    The second major adjustment involves the marginal revenue curve, introduced by Nate Silver in Baseball Between the Numbers. Since all wins are not created equal (going from 88 to 92 is far more valuable than going from 64 to 68), we factor this into the revenue estimates. Combined with each team’s earning potential, and adjusting for “baseball inflation,” we can guess how much marginal revenue each team will likely bring in. We then weigh that against how much they shouldhave brought in, had they won as many games as their payroll alone would have predicted. (Check out my article from two weeks ago for a fuller example.)

  • Are large-market GMs being penalized?
    Since GMs are being judged only within the context of their own market, the system puts large-market and small-market GMs on a pretty even playing field. If there appears to be a bias (i.e., if it seems like there are more large-market GMs on the bottom), it’s probably due to real-world trends; out of necessity, small-market teams have been more likely to hire stats-savvy GMs, who have tended to be toward the top of these lists. Large-market teams, meanwhile, are responsible for some of the biggest flameouts in the database, such as the A-Rod-era Rangers, the 2001-03 Mets, and the 2008 Tigers.

  • Why does it only use third-order wins and not include the postseason?
    This was a personal choice. Unless we can find a really significant way to predict W3W or post-season success (Nate’s Secret Sauce is a nice tool, but it’s not quite there), W3 gives us the best way to evaluate a team’s performance.

Now, moving to the rankings. You can find the entire data set here. Only GMs with three or more years on the job in the ’90s are considered. Starting with the bottom:

GM Teams Years PER
10. Sandy Alderson Athletics 1990-97 0.905
9. Ed Lynch Cubs 1995-99 0.904
8. Fred Claire Dodgers 1990-98 0.903
7. Bob Gebhard Rockies 1993-99 0.889
6. Herk Robinson Royals 1991-99 0.879

Alderson is probably a surprise here, since he is known as the godfather of the modern sabermetric-savvy GMs, but the record is what it is, the A’s had a lot of bad years in the ’90s, and they spent far more on payroll than they really needed to.

GM Teams Years PER
5. Walt Jocketty Cardinals 1995-99 0.868
4. Gord Ash Blue Jays 1995-99 0.858
3. Gene Michael Yankees 1991-95 0.822
2. Al Rosen Giants 1990-92 0.713
1. Lou Gorman Red Sox 1990-93 0.680

Jocketty is the one that stands out in this group, since he was fifth-best of the ’00s. There’s an amazingly clear divide between his first five years with the Cardinals (’95-’99) and his last eight, which I guess makes sense, when you consider that Jim Edmonds arrived in 2000, Albert Pujols in 2001, and Scott Rolen in 2002.

As for the others, Ash never really had any success with the Blue Jays, Rosen quit just in time for the Giants to get Barry Bonds, and Gorman did a pretty terrible job with the Red Sox in his last few years. Michael probably wouldn’t be on the list if he had held on a couple more years becuase he legitimately rebuilt the Yankees in the early ’90s, and his last year as GM (1995) was the first of 13 straight that the team would make the postseason. Of course, they really didn’t need a top-tier payroll before then, which is why he’s here.

And now for the best:

GM Teams Years PER
10. Jim Bowden Reds 1993-99 1.120
9. Dan Duquette Expos, Red Sox 1992-99 1.149
8. Andy MacPhail Twins 1990-94 1.152
7. John Hart Indians 1991-99 1.152
6. Bob Watson Astros, Yankees 1994-97 1.163

Hart is an obvious one here, as he took a small-market team without much of a history of success and turned it into a powerhouse. Watson and MacPhail also make plenty of sense since both led World Series-winning teams, MacPhail with the Twins and Watson with the Yankees. Also, I guess there was a reason that Bowden and Duquette kept their jobs for so long, before ultimately flaming out in the ’00s.

GM Teams Years PER
5. Kevin Towers Padres 1996-99 1.188
4. John Schuerholz Royals, Braves 1990-99 1.214
3. Brian Sabean Giants 1997-99 1.233
2. Gerry Hunsicker Mets, Astros 1991-92, 1996-99 1.238
1. Kevin Malone Expos, Dodgers 1994-95, 1999 1.243

Malone is on top almost by coincidence since he happened to be the guy that took over the Expos in 1994, which turned out to be the best individual season of the entire decade (see below). He then executed the team’s fire sale before resigning the following year, and he went on to have a miserable run with the Dodgers from 1999-2001.

The others make more sense, although I’m sure it pains a good amount of you to see Sabean once again near the top. (Blame Bonds … again.) Schuerholz certainly needs no explanation, nor does Towers, who took the Padres to the World Series in 1998. Hunsicker was in the top five of the ’00s as well, although he had a slightly more interesting go of it in the ’90s. His run with the Astros was terrific, just as it was this past decade, but he was also in charge of the Bobby Bonilla-era Mets.

Now for the best and worst individual seasons:

GM Team Year PER
278. Fred Claire Dodgers 1992 0.53
277. Sandy Alderson Athletics 1991 0.54
276. Gerry Hunsicker Mets 1992 0.55
275. Gord Ash Blue Jays 1995 0.55
274. Lou Gorman Red Sox 1992 0.56
10. John Schuerholz Braves 1991 1.61
9. Bob Quinn Giants 1993 1.65
8. Gerry Hunsicker Astros 1999 1.65
7. Gerry Hunsicker Astros 1997 1.69
6. John Hart Indians 1995 1.73
5. Lee Thomas Phillies 1993 1.73
4. John Hart Indians 1994 1.75
3. Gerry Hunsicker Astros 1998 1.98
2. Jim Bowden Reds 1999 2.07
1. Kevin Malone Expos 1994 2.07

And the best- and worst-run teams overall:



30. Royals John Scheurholz (1990), Herk Robinson (1991-99) 0.850
29. Dodgers Fred Claire (1990-98), Kevin Malone (1999) 0.874
28. Cubs Jim Frey (1990-91), Larry Himes (1992-94), Ed Lynch (1995-99) 0.876
27. Yankees Harding Peterson (1990), Gene Michael (1991-95), Bob Watson (1996-97), Brian Cashman (1998-99) 0.883
26. Rockies Bob Gebhard (1993-99) 0.889
5. Pirates Larry Doughty (1990-91), Ted Simmons (1992-93), Cam Bonifay (1994-99) 1.145
4. D’backs Joe Garagiola (1998-99) 1.200
3. Expos Dave Dombrowski (1990-91), Dan Duquette (1992-93), Kevin Malone (1994-95), Jim Beattie (1996-99) 1.223
2. Braves Bobby Cox (1990), John Schuerholz (1991-99) 1.244

1. Astros

Bill Wood (1990-93), Bob Watson (1994-95), Gerry Hunsicker (1996-99) 1.331

Once again, here’s the full data.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
Yet another reason for Astros fan depression.
As a Dodger fan, I have a really hard time with a system that puts Kevin Malone as the best of anything.
Same here (Dodgers fan). When your system shows Kevin "The Sheriff" Malone as GM of the decade, it's time to go back to the drawing board.
Another Dodger fan chiming in here. Kevin Malone at #1 immediately invalidates this whole line of thinking.
Sabean's high ranking here makes a lot of sense. The man is never questioned by the San Francisco mainstream media at all. Back in the 1990s he established his press persona as an arrogant yet brilliant GM who "doesn't suffer fools" (I've seen that exact phrase used about him soooo often.) To continue to hoodwink even a fawning media, you need at least a little substance behind the reputation.

The important thing to remember is that this is evaluating Sabean '97-'99.

He traded for Jeff Kent before the 97 season, and Kent was good in '97 and great in '98/'99. Sabean also didn't do anything monumentally stupid during that time.

Hell, even trading for J.T. Snow before the 97 season turned out well in 97 and 99.
Agreed. Sabean made good moves to complement Bonds during that period. Really if you just look at Sabean's moves prior to the Pierzynksi debacle, his only bad move was signing Marvin Benard to a 3-year deal.

Other positive moves: Getting Nen during the Marlins fire sale, getting Ellis Burks, the white flag trade with the White Sox.

The biggest criticism of him during this period would be the poor quality of the draft picks. Of course, it turned out to be a decent strategy because they were always flipping those poor picks for decent veterans.
Sabean's short-term moves that helped the Giants are pretty obvious. But, let's be clear here, his "ranking" is largely a reflection of the absolute greatness of one Baseball Anti-Christ Bonds. Rather than seeing Sabean as ranking well as a GM (especially now), I see Sabean covered by Bonds.
Funny that Cox was at the head of both the worst-run and 2nd-best-run organizations in the 90s.
You mean John Scheurholz, I'm assuming.
You mean Gerry Hunsicker, I'm assuming.
Apparently no one here can ctr-c.
No, I very clearly meant Schuerholz.
One thing about the '90s Braves: a lot of the foundation of that team was done during the '80s, under Bobby Cox. Javy Lopez, Ryan Klesko, David Justice, John Smoltz (trade), Jeff Blauser (defensive wiz), Mark Lemke (another defensive wiz), Tom Glavine, Steve Avery, and Mark Wholers were all '80s draft picks/international signings.

Of the '90s draft picks, Chipper was the #1 overall pick in '90, and Andruw Jones was an awesome international signing in '93. Schuerholz did a good job of plugging in holes where he had to (especially in the bullpen and 1B), but he should still share the '90s success with Cox.
> 5. Pirates Larry Doughty (1990-91), Ted Simmons (1992-93), Cam Bonifay (1994-99) 1.145

Ubba grubba grooby dooby dip dip fa-tang sum-bubba doogie doogie doogie. Ju hubba hubba nooby faloopa. Buggy buggy buggy buggy. Hip hip hip hip mugga mugga mugga mugga.

I'm sorry, my brain exploded. I could not grok this.
Sabean wouldn't come out so well if you consider 2005 to 2010. Part of that is the hangover from gutting the farm system in order to surround Bonds with veterans to 'win now'. But a huge chunk of it is overpaying for free agents that weren't going to help much, and neglecting the draft and Latin America. It's hard to justify the contracts given to Zito, Renteria, Rowand, and Freddy Sanchez.

The system needs further tinkering. The Pirates coming out as the 5th-best run franchise of the 90s just looks, smells, sounds and feels SO wrong;

or the system needs further explanation as to what it's measuring and what it's NOT measuring. Seems to me all it's doing is estimating how well teams managed payroll without affecting the won-loss record too adversely. In the very short run. Only one aspect of general managing a baseball team.
Let me second buffum's take above. I almost vomited. But most of it is from 90-92 -- they had some really good years early on, and some over-performing teams in '97 and '99. Taken altogether, it actually makes sense.

Remember that the Littlefield Pirates were fifth worst of the 2000s, if I'm remembering right.
I agree. Those early 90s Pirates teams lead by Barry Bonds, Bobby Bonilla, Andy Van Slyke, Doug Drabek, Zane Smith, etc. were very good.

Those late 90s teams also had some upside, but they had the propensity of extending the contracts on the wrong players (i.e. Kevin Young) or signing the wrong free agents (i.e. Pat Meares). Those bad moves spiraled out of control once they hit the 2000s.
With all respect I don't think this sort of methodology works when evaluating the efficacy of a general manager. For starters, I just have trouble buying a system that purports Brian Sabean to be damn near the best GM in baseball. I think numbers can certainly evaluate the utility of a player or the effectiveness of a team but a general manager's job goes beyond that. Certainly Gene Michael's early 90s Yankees lost a lot of games, but if Gene Michael stayed on as the general manager into the late 90s I have to imagine we'd see him on the other side of the list. His reconstruction and development of the farm system along with the choice of keeping home grown talent (Jeter, Williams, Pettitte, Rivera, Posada, etc) and intelligent trades (O'Neill, Cone) are the reason why the Yankees were able to win in the late 90s and this system completely fails to take that into account.

Basically the point I'm getting at is that I think any system to evaluate a GM that doesn't properly account for minor league success and development fails to grasp an enormous part of a GM's role.
Bozarowski is exactly right. This analysis is a nice start, but Gene Michael took over a team that was a disaster and left with a team that was on the verge of arguably the best run in the last 50 years. Sure he doesn't get 100% of the credit for 1998, but he has to get some. You could account for that either in a blunt way (e.g., accounting for the 3 seasons before and after) or in a fine way (e.g., tracking individual players/contracts acquired on the GM's watch), but either is probably a tricky exercise....
I tend to agree. One of the ways in which a GM gets ahead in these rankings, if I understand them properly, is by having a good-to-great team that is reasonably cheap. The best way to do that is to have a good number of young (that is, inexpensive) players on hand. But we know that it usually takes at least two-three years for draft picks/international signings to make it to the big leagues, and sometimes then another couple of years before those players do well. As was shown above for the Braves, it was Bobby Cox who drafted most of the guys that became the Braves' core in the 90s. It doesn't seem quite fair to give Schuerholz all the credit for that - or to ding Michaels for not being GM when all the young Yankees really told hold in the late 90s. All that said, I also agree that I'm not sure what the right solution is.
Agreed. The results of this analysis miserably fail the smell test. Any system that declares the Yankees to be among the worst-run teams and the Pirates among the best in the 90's is flawed to the point of uselessness.

This is a great idea and I hope you continue to develop your model but there needs to be a much more complete picture of both the inputs (retention and development of homegrown players) and the outcomes (pennants and World Series) of franchise success before any non-esoteric claims about GM performance can be made.
well said, completely agreed, this system isn't measuring
the 'inputs' for the players into the teams' systems unless they happened during the same decade or under the same GM
Payroll Efficiency is a terrible way to rate General Managers. The Pirates are terrible, have been since 1993. Sure the first three years of the decade were awesome and 97 and 99 they did better than they should have. But the rest of that decade was a disaster. Furthermore, look at their draft picks from 1990-1999. If BP did an article on teams effectiveness on draft day (if that article exists, can you link to it?), the Pirates of the 90's would probably come in dead last. And besides the 00 and 01 drafts, the Pirates pretty much continued draft day irrelevancy until 2008 when they took Pedro Alvarez with the 2nd overall pick.
Doughty (2 winning seasons), Simmons (1 of each), and Bonifay (8 straight losing seasons and his worst record yet in the final year of his last 5-year plan- 2001, 62-100, which opened PNC Park). Which one of these GMs was not like the other 2?
I don't understand why, if a win in New York generates more revenue, that this means that a NY GM can without penalty spend more money to get that win. It seems to me that if the marginal value of a $2M player in wins is the same as the value of a $10M player, then the NY GM should be properly penalized for going with the $2M player.
You're thinking about it backwards. Given two equal 85 win teams in competition for the same 2-win player, it makes more sense for the Yankee GM to win the auction. The extra 2 wins could be worth $8M in revenue to the small-market GM, but $14M to the Yankee GM.

So the Yankee GM pays $5M per marginal win to outbid the small-market GM (who can only pay a max of $4M without actually losing money per marginal win). If you do strict "expenditure per marginal win", then the Yankee GM comes out weaker every time even though he's succeeding in making money for his team.
I think the idea of payroll efficiency is important and interesting.

I don't think that any study that shows the Yankees as one of the 5 worst run franchises of the 90's, with the Pirates as one of the best, is showing what it thinks it is showing. More precisely, the conclusion that payroll efficiency is the sole way to evaluate how good a team is run seems to be missing something.

I generally agree, but then I think showing that there's a whole lot more to a GMs job than winning a faux payroll efficiency championship is a good thing to note.
To further bolster Hunsicker's credentials, he's been integrally involved in the Rays front office since 2007 as a VP, and look where they've gone since.
Thought of course guys like Crawford, Upton, and Longoria were drafted prior to that.
I'm sorry, but an evaluative system that puts Jocketty in the bottom 5, an Bowden and Sabean in the top 10, is useless.
I have to agree. I think the trouble with this article is not in the content, but in the title. This isn't the "Best and Worst" GM's. This is merely a ranking of major league teams under a GM's watch on a dollars per win basis, with some extra math thrown in.

Just like the last article, this seems to ignore the long-term effects a GM can have on the front office mindset, farm system, draft, resource allocation. It also ignores the ages of the players involved on those teams and their production along their career path.

Why on earth should Kevin Malone get credit for inheriting a team with Pedro Martinez, Larry Walker, Marquis Grissom, and Moises Alou, just to name a few?

It'd be like if somebody took over for Brian Cashman in 2010 then everybody crediting him with a 97-win season for the Yankees.
Or people blaming Neal Huntington for the Pirates having two 90+ loss seasons. Like it's his fault he inherited the worst run team in baseball with an empty minor league system with one good player (McCutchen) and a bad major league team.
I think it's interesting that the 5 best-run teams of the '90s won a total of one World Series between them.
Well, when the 27th best-run team wins 3 championships and a strike cancels a 4th, there just aren't that many left for the so-called "best-run" teams.
You really need to look at what happens after a GM has been in place for 5 years before you can really judge their performance. That's when the prospects that they drafted or traded for will be reaching the majors, and the contracts that were signed before their tenure will be expiring.
I like the idea behind the methodology, but when I see the list, it seems people are rated up or down because of things like having Bonds, or being in New York, or being lucky enough to have a good year or two.

It just seems the list has more to do with how lucky a GM was than how well they performed. Did they happen to be GMing a team when it happened to win, etc.
To be fair to Shawn, I think he recognizes that. All his comments indicate that the GMs are at their position in the rankings largely due to external factors or timing issues. That does seem to call into question the purpose of the system, though, if you can write off each ranking as being due to this or that factor that the GM was not necessarily responsible for.
While I feel there is merit to Shawn's system, and that it does offer some glimpses of promise, a truly innovative system would be one that takes into account -- on an objective, numbers based level -- the situation the GM inherited as well as the situation he left with. That would be pretty tough, but much more informative.
There's a major timing issue even in the short run. E.g., Kevin Malone took over the Expos on January 27, 1994 (according to Baseball America's splendid exec database). So while he sat in the GM's chair for the strike-shortened '94 season, wasn't that roster/payroll largely determined by his predecessor? The winter meetings and major FA signing periods were both done deals by then.
Perhaps in cases of "split off-seasons" like this one (and there are no doubt many others), you can split the credit or blame for the subsequent season's results?
As an Expos fan, I looked at this back in the day. I can't find the data anymore, but as I recall almost any way you split up the credit for the 1994 Expos (playing time, runs above replacement, win shares, whatever) Dan Duquette gets the lions share of the credit for the 1994 Expos, with Dave Dombrowski second, and Malone third. Malone didn't really contribute much to the 1994 Expos, other than stepping back and staying out of the way.
Thanks, Scot. So, if Shawn is willing to adjust his averages for split off-seasons, the flukey result that Malone is "tops in the '90s" could be corrected. And since '94 was such a high-PER year, I suspect Duquette would move up the ladder considerably.
Perhaps a better scheme would be to judge GMs by the FAs signed, trades, and draft picks they made using something like WAR to sum up the players. Then rank by WAR/salary or something.
the clear flaw to the method is that it assumes all teams have only one goal in mind, to achieve the highest efficient return on their investment. however, clearly teams in different markets and with different operational goals have different normative aims. the marlins etc might just want to make a profit with minimal investment, while the yankees invest in stars, badly or wisely, to build their brand and thus business. at least recognize the gross simplification
I agree with the concept that GM's work from years before influences the coming years. Also, I think the statement that the large market teams are not penalized is flat wrong. In 1998, NYY had 114 wins and they end up with a score of 1.22...good for 52nd best year of the decade. What is the ceiling for a team with a payroll of $200 million? How many wins would it take for NYY to get 2.07? Is it even possible? I think this is another example of BP developing a metric to support their this case the small market/young cheap kids is the way to go.
Tempted to pick up Jay Jaffe's recent book on managers despite the hefty price tag. Anyone that has read through that, feel free to chime in on whether I should drop $35 or so there.
Haven't read the book, but it isn't Jay Jaffe's. It was written by *Chris* Jaffe, he of The Hardball Times.