Fire Gardy

A lot of people have spent a lot of time thinking about how to measure the performance of a player on the field, and they’ve largely done the same for the performance of a GM in his office. But when it comes to the manager, there’s the hand-wavy “Well it’s the players’ talent that’s important, the manager doesn’t do anything” dismissal.* Some go so far as to say a good manager can be worth up to a couple extra wins over the course of the season; conversely, perhaps, a bad manager may be worth a couple of extra losses.

* This dismissal seems strange to me, given that there are new stats like WXRL and WPA/LI, etc, which take into account success — and opportunities — in tough, meaningful situations. IE, exactly the situations in which a good manager is supposed to make the moves that put the right players in the best position to succeed. If there’s something to what a manager does, and it’s not just random dice rolling that determines whether that reliever records a strikeout or that pinch hitter hits a home run, then there must be some way to measure it.

But it seems that nobody’s tried to actually measure a manager’s impact. The “best manager” discussion is always couched in terms of total wins, or total championships, or total-number-of-nostalgic-stories-told-by-old-sportswriters, rather than attempting to objectively discover who have been the best (and worst) managers in the history of baseball.

Thus, I have set out to try and figure it out.

I start with the notion that a team’s Pythagorean record “predicts” how many games a team should win based on its totals of runs scored versus runs against. I would say that one (significant) goal of every manager is to attempt to squeeze more actual wins out of the team’s runs for/against than the formula would predict. (Though I doubt many managers would couch their goals in terms like that.)

So I went through the entire Retrosheet database from 1871-2008, counting the actual win-loss record and summing the runs scored and given up for all the managers, ever. It doesn’t go by team, but by manager; for example, if Ron Gardenhire were suspended for arguing with an umpire, and Scott Ullger replaced him for a couple of games, then those games and runs would count toward Ullger rather than Gardenhire. Similarly, if Ned Yost is fired in the middle of the season and Dale Sveum takes over, any games Sveum managed count towards Sveum, not Yost. Should be obvious enough.

Once I’ve summed up the win-loss record and the runs for/against, I can calculate each manager’s expected win total based on their Pythagorean record. If this method makes sense, then good managers will have more actual wins than expected wins, and bad managers will have fewer than expected.

Retrosheet’s gamelog files for many seasons in the 1970s-1990s have been down for weeks. Thus, my data are incomplete, which is too bad. Hopefully at some point I’ll be able to pull down the rest of Retrosheet’s data to complete my analysis. In the meantime, I’ve limited my analysis to 2002-2008, which is the only modern stretch for which I have data. I also decided to limit it to managers with at least 200 total wins, to keep any small-sample-size bias out.

In any case, looking at the limited data, here are the 11 managers who were able to average one win more than expected, per season, over the course of the last seven years:

+----------------+------------+------------------+-----------+------------+-----------------+
| managername   | totalwins | totaldiff       | max(diff) | min(diff)  | averagediff    |
+----------------+------------+------------------+-----------+------------+-----------------+
| Felipe Alou    |        341 | 15.1456443071365 |   6.07211 |    1.95034 | 3.7864110767841 |
| Ozzie Guillen  |        432 | 13.0560693144798 |    6.8328 |   -1.13229 |  2.611213862896 |
| Jack McKeon    |        241 | 7.37834513559937 |   3.75759 | -0.0560862 | 2.4594483785331 |
| Frank Robinson |        385 | 10.7293409258127 |   4.19535 |   0.104871 | 2.1458681851625 |
| Joe Torre      |        673 | 14.6219349503517 |   11.7966 |   -4.97134 | 2.0888478500502 |
| Ron Gardenhire |        615 | 14.1632239818573 |   7.41682 |   -1.74507 | 2.0233177116939 |
| Mike Scioscia  |        645 | 12.9273184239864 |   11.4814 |   -4.00882 | 1.8467597748552 |
| Ken Macha      |        368 | 6.79638695716858 |   7.24576 |   -5.83313 | 1.6990967392921 |
| Bob Melvin     |        481 | 9.29994755983353 |   12.5254 |   -5.65932 | 1.5499912599723 |
| Ned Yost       |        457 | 6.62773448228836 |   4.62946 |   -3.30011 | 1.1046224137147 |
| Tony LaRussa   |        634 | 7.61316386237741 |   7.79336 |   -3.73349 | 1.0875948374825 |
+----------------+------------+------------------+-----------+------------+-----------------+

Felipe Alou’s second career as a manager — with the Giants — was shortlived, but he did a hell of a job. He averaged +3.79 wins per season above expected, and was never worse than +1.9 wins. (I don’t have any data from his time with the Expos, unfortunately, so this is just when he was with the Giants.)

It’s absolutely not surprising that Scioscia is high on the list, given that he was +11 Wins Above Expected in 2008 alone. (Indeed, if you removed 2008 from the sample, his average would drop from +1.8 to +0.2.)

Ozzie Guillen, Frank Robinson, Joe Torre, and Ron Gardenhire are generally regarded as good managers, and they appear high on the list. Tony LaRussa is well regarded, but for all his late inning machinations, he’s only +1 WAE/year recently. (Perhaps he was better when he was younger. Further investigation is needed.)

Also, how would you like to be Ned Yost, fired at the end of a season in which you guided your team to the playoffs, after having been a top ten manager over the course of the decade? Probably wouldn’t like it much.

If this is a viable investigation, looking at the bottom of the list should show bad managers. So let’s look for everyone who won at least 100 games in the last seven years, and averaged worse than -1 WAE (ie, they were one game worse than expected per season).

+----------------+------------+-------------------+-----------+-----------+------------------+
| managername   | totalwins | totaldiff        | max(diff) | min(diff) | averagediff     |
+----------------+------------+-------------------+-----------+-----------+------------------+
| Alan Trammell  |        186 | -14.2434198856354 |  -3.14609 |  -7.35206 | -4.7478066285451 |
| Jimy Williams  |        214 | -13.0165817737579 |  -1.92939 |   -7.8883 | -4.3388605912526 |
| Eric Wedge     |        495 | -22.6846791505814 |   3.61521 |  -11.6051 | -3.7807798584302 |
| John Gibbons   |        308 |  -16.938233718276 |  0.306474 |  -8.64506 |  -2.823038953046 |
| Jerry Manuel   |        222 | -8.36836788058281 |  0.459961 |   -5.6738 | -2.7894559601943 |
| Buddy Black    |        152 | -4.59854626655579 |  -1.16408 |  -3.43447 | -2.2992731332779 |
| Lee Mazzilli   |        129 | -4.35584957897663 |  -0.19323 |  -4.16262 | -2.1779247894883 |
| Jim Leyland    |        257 | -6.27698922157288 |   -1.6333 |   -2.8878 | -2.0923297405243 |
| Grady Little   |        358 |  -8.0911720469594 |  0.189979 |  -8.29321 | -2.0227930117399 |
| Bob Geren      |        151 | -3.36636139452457 |  -0.20335 |  -3.16301 | -1.6831806972623 |
| Bobby Cox      |        624 | -11.7078827321529 |   3.74093 |  -6.35538 | -1.6725546760218 |
| Clint Hurdle   |        516 |  -11.283863902092 |    4.6587 |  -5.09969 | -1.6119805574417 |
| Larry Bowa     |        251 | -4.51537179946899 |   1.07168 |  -5.19319 | -1.5051239331563 |
| Dusty Baker    |        491 | -8.68119883537292 |   3.29847 |  -5.71592 | -1.4468664725622 |
| Jerry Narron   |        230 | -5.34940385818481 |   4.42873 |  -5.33926 | -1.3373509645462 |
| Carlos Tosca   |        188 | -3.84612214565277 | -0.553967 |  -1.98905 | -1.2820407152176 |
| Buck Showalter |        318 | -4.37028300762177 |      2.84 |  -6.09809 | -1.0925707519054 |
+----------------+------------+-------------------+-----------+-----------+------------------+

It’s not exactly surprising that these guys all managed teams that underperformed, given that that’s basically what we’re measuring. How much of these underperformances were random ups-and-downs, and how much were they influenced by a poor job by the manager?

For the most part, the guys on this list are bad managers of bad teams. It may be that Wedge, Leyland, Bowa and Cox are actually better managers than appears on this list and that because of our subjective opinions of them, we want to discount this analysis. But it’s probably worth some harder thought into it. Cox was a great manager, and he’s had a great career; however, he’s getting old and may not be able to connect with his players or think as quickly as he used to. I’ve always thought Wedge was a good manager, but how many times does he have to take a roster that’s absolutely loaded with talent and guide them to a terribly disappointing season before maybe it’s not random, and is in fact his fault?

I’d like to fill this out with more years worth of data from Retrosheet, to get a better look at the careers of Cox, LaRussa, and Torre, along with some historical managers. I think this is a good start, but would be better if I could look at a manager’s year to year performance and see if there are any managers who are consistently above their expected wins (like Guillen and Gardenhire appear to be) rather than relying on one amazing season (like Scioscia).

Thus far it looks like it should be possible to measure the performance of a manager, in such a way that one can be compared to another. I’m not there yet, but this appears to be a (small) step in the right direction.

5 comments

Twins 3, Blue Jays 2- 11 innings.

We had a good old fashioned pitchers duel last night.  The second time Perkins has been involved in one this season.  However, the offense didn’t  let him down as badly this time.  Both lineups were held in check throughout most of the game.  It was very frsutrating to see the Twins load the bases with no outs in the sixth only to have Cuddyer strike out, and Crede ground into a double play.  Perkins didn’t let this get him down, he continued to mow through the Jays lineup until the 8th when he gave up a double to Rios, who eventually scored on a single.  His line was still spectacular.  8 innings, 2 ER, 4 Ks, 1 BB.  Can’t ask for much more out of your pitcher, especially when facing the highest scoring team in the AL.

Crede more than made up for the inning ending DP, with a walk off double off of Jesse Carlson.  Our very own Jesse pitched two scoreless to pick up the win. 

This was a great game, however I would like to see the offense stop wasting opportunities like they have all season.  I understand we don’t have Joe Mauer, but that doesn’t mean scoring no runs after loading the bases with no outs is OK. 

Baker gets the start tonight, with the team moving Duensing to Rochester to make room.  This means we are hanging on to RA Dickey for a while.  I’m OK with this move, Duensing will be inserted into the rotation in AAA which is probably better for him. 

Before the season did anyone think Glen Perkins would be our best starting pitcher after two times through the rotation?  I certainly didn’t.

10 comments

With exactly one good start from our group of young pitchers (minus Baker) is it time to hit the panic button in Twins land?  It is more than possible that all of our starting pitchers experience some sort of  sophomore slump.  Hopefully this isn’t the case.  I like to think that the Twins were careful enough with this group last year. 

Kevin Slowey gave up about 100 hits to the Blue Jays last night.  On Sunday Blackburn was cruising until the wheels fell of.  Same goes for Liriano on Saturday.  RA Dickey doesn’t really count, and the offense couldn’t do anything for Perkins on Thursday.  What does this all mean? Probably nothing, since its been 8 games.  However, if we can’t beat teams like Seattle and Toronto, we might be in trouble.  It helps that our divisional rivals are struggling as well.

So what do you guys think?  Is there some silver lining I’m missing to these first 8 games?  At what point to we have to get worried about our pitchers? 3 starts, 4 starts, July?

5 comments

Today Baseball Prospectus managed to get a good quote from Gardenhire regarding the playing time situation for our set of outfielders.

Gardenhire had been asked about the configuration throughout the winter and spring training, and it’s clearly begun wearing on him, as he cut off reporters on opening night after being asked just one question about the situation. “It’s going to be that [way] all year off and on,” Gardenhire said. “We could sit here and do this every game, because we’re going to mix them up. These guys are preparing themselves to play. The only people who are going to bring it up every day are [the media]. I don’t want to sit here and discuss all the reasons. Read the stats. Check them out. That probably will tell you a little more about who’s in, who’s out, and we’ll go from there. I’m going to play them all year long. I don’t have any other way I can do it. I have four outfielders and three places for them to play. Something’s got to give. There’s no easy solution here. You could sit and say I’m going to platoon here or there, that’s not right. That’s not the way I want to go about it. You know what? They are good players, and no matter who we put out there, we’re going to have an opportunity to win games. I need to mix and match the best I can, and keep them all busy and all playing, which is almost impossible to do.

My only question is … exactly what stats is Gardy looking at? Will he admit to saying that if you asked him about it?

Oh, and here’s an interesting tidbit from the same article:

When Cardinals left-hander Dennys Reyes pitched a perfect ninth inning to close out the Pirates on Thursday, he recorded his first save since September 27, 1999, when he was with the Reds and pitching against St. Louis. The gap of nine years and 192 days between saves was the longest since Curt Schilling went 13 years and 77 days before saving the Red Sox‘ win over the Devil Rays on July 19, 2005.

Good job, Reyes. (Though the failings of Motte probably have more to do with it.)

1 comment

Last night we could all enjoy a comfortable game where the Twins coasted to victory over the White Sox. The game was actually close coming into the seventh inning, when the Twins batted around and plated 7 runs.

In my ongoing effort to introduce everyone to the WPA graphs from FanGraphs, here’s what it looks like when a team scores a bunch of runs and turns a tight matchup into a comfortable win:

As you can see, the Twins and Sox were trading right around 50% through the first half of the game, until Morneau started the 7th with a home run and after that the graph just went to zero for the rest of the game.

Last night I received a question about the second graph below the WPA chart. My first thought was that it represented the WPA change for each play, but it didn’t take long to realize why that doesn’t make sense, given that it’s basically a repeat of the data in the top graph. But, obviously, the label underneath it calls it the “Leverage Index,” which is a term I’m familiar with; basically, it represents how important a situation is to a game, before the outcome (WPA represents what happened after the outcome).

The tallest bar isn’t a perfect representation of how the WPA graph isn’t necessarily tied to the LI graph below it. That bar represents the situation when Josh Fields came to the plate in the bottom of the second, with the bases loaded and two out. The LI bar represents that that’s a critical situation. The spike in the WPA graph indicates that he came through. If he hadn’t come through, the WPA graph would have gone slightly down rather than way up (because it represents post-outcome probability), but the LI graph would still have a tall bar (because it represents pre-outcome leverage).

So today we got to look at another shape of a WPA graph, and learned a little more about the information the graph can tell us. I’ll try to keep posting WPA graphs on my game recaps, perhaps only when they’re interesting or novel.

No comments

In case anyone wants to feel hopeful about Dickey in Chicago tonight, here’s an interesting take on knuckleballers in cold weather, from intelligent reader thrylos98:

There is another physical factor that affects a knuckleball more than the wind: Air density. The colder the temperature, the denser the air, the less the spin. And throwing a knuckleball with no spin, is like throwing a brick to the batters.

The game is just starting, and Dickey hasn’t taken the mound yet. Hopefully that’s exactly how it works tonight.

No comments

Yesterday my entire day was stopped cold upon learning about the Adenhart tragedy. I’m not one for heartfelt words that make one feel better or put their lives in perspective, or anything like that, so I’m just going to try to move on and attempt to use baseball as an escape from life’s harsh realities. And I’m not even going to pull a Joe Sheehan and say something like that and then say something heartfelt that helps put things in perspective. I’m really not.

So, let’s just try to get things going today.

The Twins are coming to Chicago to face the White Sox, which means three things:

1. I will not be able to watch the game on MLB.TV, since it’s be on Comcast Sports Net here in Chicago
2. Thus, I will not be able to choose the broadcast, and will be stuck with the infinitely annoying Hawk Harrelson*
3. RA Dickey will be on the mound tonight, which should be extremely interesting in the windy weather we’ve got going down here today

If you think the Twins’ offense looked miserable this week against what wasn’t supposed to be a great Mariners pitching staff, the White Sox looked even worse against an even worse Royals staff. Both teams will be looking to get their offenses back on track, as well as hoping for some better pitching (though the Twins are more in need of that than are the Sox at the moment).

* My only hope in the event of a (probably inevitable) home run is that the fireworks that’ll launch right down the street from me manage to drown out Hawk’s unbearable home run call. It’s happened before … though sometimes the TV feed is delayed long enough that I hear the fireworks before I see the home run. Believe me, that’s disheartening.

This isn’t a newspaper, so I don’t need to attempt to invent storylines like “Game 163 Revisited” or anything lame like that. The important thing is that this is a big rivalry series and we need to win these games. We need to take two of the games this weekend, and not allow the White Sox to bounce back from their opening week woes (which appears to be the wont of the Twins over the last several years).

Go Twins.

2 comments

Remember when Gardy refused to DH Jason Kubel because he didn’t think a young player should be a DH, and needs a position in the field until he’s too old and decrepit to play in the field? Well, he finally got over that with Kubel, who has become our standard DH.

Apparently, though, he’s REALLY over it, because it appears that he’s decided the proper use for Delmon Young, all of 23 years old, is as a platoon DH against left handed hitters. He’ll start in place of Kubel at DH this afternoon, the second time thus far in the young season this has happened (although last time, Cuddyer was the DH and Young started in LF, but the concept remains the same).

Like Mauer and Morneau, Kubel could be good enough to face left handed pitching. With enough right handed bats off the bench, though, a platoon isn’t a terrible thing for him, even if it lowers his value. (He’s no longer an every day player.)

But it lowers Delmon’s possible value even further; in Spring Training it was apparent that Young had worked hard on his defense, and sitting on the bench most nights is not a good way to develop his bat or keep him happy. In typical Gardenhire fashion, it’s much more important to keep Cuddyer happy, given that he’s six years older and over the course of his career has proven that the performance you can expect from him is just about the same as what you could pessimistically expect from Delmon Young.

While having too many viable starters is a much better problem to have than having too few, before the season started there was at least some hope that Gardy would manage to shuttle the guys in and out effectively. Thus far, the only players who have sat on the bench are Jason Kubel and Delmon Young, who also happen to be the two guys with their names carved in stone above the door on Gardy’s doghouse.

There’s not an easy solution to this problem, unless someone goes down with an injury. But given that Gardy’s strategy seems unsatisfactory, I’m going to put forward a strategy that could be better for the team now and down the road.

Gardy’s strategy: Gomez and Span play every night. Kubel is the DH against RHP, bench against LHP. Cuddyer plays every night, either in RF or DH against LHP. Young sits most nights, can play LF or DF against lefties.

My strategy: Gomez and Span play every night. Kubel is the DH against RHP, Young is the DH against LHP. Against RHP, split the time between Cuddyer and Young in RF; don’t alternate games, but attempt to go with the hot bat. Additionally, Delmon Young is a right fielder; he should only play LF if Cuddyer is also on the field, which should never happen.

It’s possible that Gardy’s plan is, in fact, to alternate Young and Cuddyer in the outfield against RHP, and not to do it every game but to attempt to ride a hot bat. If it is, it only appears that he plans to play Cuddyer every day because it just so happens that he got the first turn at it. (Though I’d prefer it if when Cuddyer and Young both play, that Young is the DH; also that Young never plays LF, basically setting Span as the every day LF.)

Is it too early to worry about Young’s playing time this season? Probably. But given that Gardy doesn’t like to have his team figured out until June, and in a tight race like the AL Central is bound to be in 2009, it’s almost certainly better that we pay attention to this kind of thing in the hopes that Gardy notices that Cuddyer isn’t sitting down often enough.

9 comments

Look, I know this happened two days ago and is therefore old news, but I just can’t let it slide. Remember when we came back in the bottom of the ninth against the Mariners? (In quite thrilling fashion, I might add.) Well, what if we hadn’t managed to take the lead, and the game had been tied going into the 10th? Recall that Kubel had pinch-hit for Morales and Buscher had pinch-hit for Punto, meaning our catcher and shortstop were out of the game. And the backup catcher, Redmond, is injured.

What’s a manager to do?

Mike Redmond and his pulled groin muscle was going to hobble out to the field to catch the tenth if the game had been tied. And get this: Twins manager Ron Gardenhire had told R.A. Dickey to get ready to play in the field if an emergency situation.

“Gardy told me to get my spikes on.” Dickey said. “I was going to go out there and Babe Ruth it. Yesterday, I was the bullpen coach (in place of the ill Rick Stelmaszek). Today I was going to be the right fielder (if needed). This is a heck of an organization.”

Well, Redmond was going to play. Because he’s tough like that.

But RA Dickey, the knuckleballer who made the team because he struck out everyone he faced this spring and Scott Baker got injured, was going to play in the outfield.

I really enjoyed seeing the comeback and I couldn’t be more pleased with the win. But is anyone else imagining the difference between Cuddyer firing a missle from RF and Dickey fluttering in a knuckleball from the warning track? Doesn’t anyone else want to see that?

Also, when he calls it “a heck of an organization,” I don’t know if he means it in a good way, or if he meant to say “wreck.” I’m going to assume the former, because I don’t want to start wishing ill on an entertaining knuckleballer who happens to be on our team.

3 comments

Homers by Morneau and Span were the most obvious evidence of the offense awakening in game 3, showing off the first power the Twins have found so far this year, lifting the team to a 2-1 record despite a rough night from Slowey — the third consecutive sub-par start we’ve had to try to overcome. Given that a big strength of the team is supposed to be the starting pitching, this is a little worrying. But again, it’s still early; I think they’ll hit their stride.

This WPA graph from FanGraphs is evidence of a significantly more comfortable game than we’ve played so far this year, and hopefully we can enjoy plenty more games with graphs like this. Especially when the game trades back and forth in such exciting fashion. Absolutely nothing wrong with entertainment like this.

I want to point out Morneau’s “double” in the 5th inning, when he drove in a run to tie the game. It was a hard grounder up the middle, that got through and went to center. It scored a run, sure, but the impressive part of the play was the Morneau managed to take second before the ball came back in from the outfield. Blyleven apparently* pointed out that Morneau saw that the CF was running to his left and would thus have to stop his momentum before throwing the ball back into the infield.

* I say “apparently” because I had the game on mute, and my dad reported Bert’s comments to me via Skype. (Which, by the way, is a great way to watch a ballgame when there’s nobody you know within miles of you.

However it happened, it was good. Because Morneau scored what turned out to be the winning run on the next play when Kubel doubled him home — Morneau and his “speed” could not have scored from first. It was simply a great play by the Twins’ MVP.

Despite my concerns about Slowey and all our starting pitching at this point, this was a good win and we’ve pushed ourselves into the above-.500-column, which is exactly where we want to be. Game 4 is at noon on Thursday, and hopefully we can push some runs across again and win the series.

4 comments