D3

[Jayhawk Bill]

D3 is probably the best single, quickly available metric regarding the caliber of an MLB manager and his team.

 

***

 

What is D3?

 

To understand that, a few other definitions:

 

W, L : Actual team wins and losses.

 

First-order wins and losses: Expected wins and losses, based on a team's actual runs scored and runs allowed.

 

Equivalent average (EQA): A measure of total offensive value per out, with corrections for league offensive level, home park, and team pitching. EQA considers batting as well as baserunning, but not the value of a position player's defense. The EqA adjusted for all-time also has a correction for league difficulty. The scale is deliberately set to approximate that of batting average. League average EqA is always equal to .260. EqA is derived from Raw EqA, which is (H + TB + 1.5*(BB + HBP + SB) + SH + SF) divided by (AB + BB + HBP + SH + SF + CS + SB). REqA is then normalized to account for league difficulty and scale to create EqA.

 

Equivalent runs (EQR): EQR = 5 * OUT * EQA^2.5. In the fielding charts, the estimated number of EqR he had at the plate while playing this position in the field. In Adjusted Standings, EqR refers to the total number of equivalent runs scored by the team.

 

Second-order wins and losses: Expected wins and losses, based on a team's equivalent runs scored and equivalent runs allowed.

 

Adjusted equivalent runs (AEQR): EQR, adjusted for the exact strength of opposing teams' players.

 

Third-order wins and losses: Expected wins and losses, based on AEQR and AEQRA.

 

***

 

What is D3? D3 is the difference between actual wins and third-order wins. That's the difference between how a team actually did and how it should have done, given

 

1) Its exact hits, walks and other contributions on offense;

 

2) Its exact hits and walks allowed, and other contributions on defense;

 

3) The strength of its opposition.

 

A good (or lucky) manager has a positive D3, getting more wins than expected, given how his players did.

 

The mean of D3 across MLB is, of course, zero. The standard deviation is 3.0. Let's look at the worst MLB managers by D3:

 

John Gibbons, Blue Jays (-4.1 games): John Gibbons' own grandmother reportedly asked him if he was going to be fired, given that he was doing so badly.

 

Clint Hurdle, Rockies (-4.0 games): Clint Hurdle has been known by stats guys, especially Chris Jaffe (who has studied this carefully), as one of the worst managers in MLB history, and probably the worst of this century.

 

Jim Leyland, Tigers (-3.7 games): Jim Leyland's team was supposed to be a contender, and they're ten games back in a weak division. Part of it is the players, but part is that they're winning less than they should, too.

 

OK, these make sense. Who are the best?

 

Mike Scioscia, Angels (+8.2 games): Yes, that's right. The Angels have won 38 games, and eight of those games can be credited either to good luck or to the managerial skill of Mike Scioscia. Given that Scioscia has a strong reputation, and that he's a perennial leader in this metric, and that 8.2 is almost three standard deviations above the mean, it's probably not just luck. Mike Scioscia is probably the best manager in MLB right now.

 

Ron Gardenhire, Twins (+5.0 games): Another perennial strong manager, taking a "rebuilding team" into contention in the AL Central, Gardenhire is doing a fine job this year.

 

Manny Acta, Nationals (+3.5 games): For some reason, Acta is highly regarded, even though his team stinks. This metric supports that.

 

Tony LaRussa, Cardinals (+3.5 games): A perennial contender for Manager of the Year, and certainly one of the great MLB managers of all time, LaRussa is, again, exceeding expectations.

 

No surprises here, either.

 

***

 

Terry Francona is currently at -1.5 wins by D3. This isn't unusual; Terry Francona comes up short in most attempts at quantitative analysis. Chris Jaffe's work found him to be, along with Clint Hurdle, probably one of the two worst managers of his generation.

 

Let's see how things go for Terry Francona over the next few weeks with respect to D3. With the minor and major injuries on the ballclub, plus the recent suspensions, Terry Francona has a chance to make a difference. We'll see which way he makes that difference.

[SCM33]

Let's see how things go for Terry Francona over the next few weeks with respect to D3. With the minor and major injuries on the ballclub' date=' plus the recent suspensions, Terry Francona has a chance to make a difference. We'll see which way he makes that difference.[/quote']

 

 

In fairness, he had some pretty bad Phillies teams. Any way you oculd figure the D3 based on only time with the Sox?

[ORS]

The D3 he posted is for this season only. It's what our record is compared against what it should be based on the sum total of all individual performances.

[Jayhawk Bill]

In fairness' date=' he had some pretty bad Phillies teams. Any way you oculd figure the D3 based on only time with the Sox?[/quote']

 

First, ORS is right: the -1.5 games counts only 2008 results.

 

That said, having a bad team doesn't make one a bad manager by available metrics. It's the number of wins relative to expected wins that I'm citing here. Last year the big winners were Bob Melvin of the D-Backs in the NL and Mike Hargrove in the AL.*

 

Chris Jaffe's work (no longer available online) showed that Terry Francona had had a terrible year with the 2000 Phillies, but that he'd only had one year where his team didn't underperform, 2005. Also, Jaffe's work looked at not only wins relative to run creation but also player performance relative to their performance before and after working for a given manager, and Terry Francona really, really sucked regarding pitchers' performance. Jaffe had five separate metrics--Terry Francona was below average in all five.

 

Tito's D3 is probably a reasonable metric of where he stacks up right now. Managers have a career curve vs. age, same as players, and Tito is reaching the age where he's roughly stable in game management but growing in developing and maximizing players' contributions.

 

I'd agree that Terry Francona was worse with Philadelphia than he's been with Boston--remember, the Phillies got 21 games better the year after Terry Francona's last season managing them. I'd submit that a lot of the regular-season drama in 2004 was due to poor management by Terry Francona, and I'd remind folks that a whole lot of players have had their careers either ruined by Tito or resurrected after leaving his team these past five seasons.

 

But Terry Francona isn't as bad as he was, and he's an established part of both the clubhouse and the media-clubhouse interface. A -1.5 isn't statistically significant from a normal manager's 0.0: let's see how that metric shifts as Boston goes through a period where good or bad management is liable to make a big difference.

 

 

 

 

* Mike Hargrove got my vote for AL Manager of the Year in the Internet Baseball Awards last year. The Mariners are 22-40 under McLaren this year, and they were 43-41 under McLaren last year. Hargrove posted a 45-33 record with a Mariners team that played less well for him than it did later for McLaren.

[YAZMAN]

D3...

 

you sank my battleship.

[TheKilo]

I always liked the Mighty Ducks.

[Jayhawk Bill]

After two wins this weekend, including the tight win yesterday, Tito is up to -1.1 D3.

 

Edit: Well, ten days and several wins later, Tito is down to -2.2 D3.

[Jayhawk Bill]

Tito is down to -2.8 D3 as of this morning. Each AL manager with a lower D3 for the season has already been fired. Tito is now worst in the AL.

[jacksonianmarch]

how is girardi?

[Jayhawk Bill]

how is girardi?

 

-1.2 today, the worst I've seen him all year...he's been around -1.0 when I've looked.

[Jayhawk Bill]

Let's see how things go for Terry Francona over the next few weeks with respect to D3. With the minor and major injuries on the ballclub' date=' plus the recent suspensions, Terry Francona has a chance to make a difference. We'll see which way he makes that difference.[/quote']

 

Terry Francona was at -1.5 D3 when I posted this 20 days ago. He's now at -3.0 D3. The team has gone 11-6, but, astoundingly, the team played well enough to have done a win-and-a-half better, at least in theory. Looking back over the games, one could question a few decisions on June 13, June 20, and June 23--I guess that a good MLB manager would've won two of those three games, and even an average manager would've won at least one of the three.

[BoSox21]

I always liked the Mighty Ducks.

 

bah, beat me to it

[Jayhawk Bill]

Four days later, Tito is at -4.6 D3.

 

One day later, Tito is at -5.1 D3. Boston's pitchers and hitters did as well as the Rays' last night, but Boston took a full loss, not a half loss. Boston as a team is doing over five wins less well than one would expect given its players' actual performances--and that doesn't count the missing contributions of Big Papi and Schill, among others.

 

Here are the current worst teams in MLB by D3:

 

[table]Team | W | L | D3

Blue Jays | 41 | 44 | -6.9

Rockies | 33 | 51 | -5.8

Red Sox | 50 | 36 | -5.1

Braves | 40 | 44 | -4.9

Mariners | 32 | 51 | -3.7[/table]

 

The Blue Jays already fired their manager, as did the Mariners, and the new Jays manager, Cito Gaston, is only responsible for a small part of the MLB-leading D3. Jim Riggleman is posting an above-average D3 in his few games thus far. Bobby Cox of the Braves, once a great manager, is now 67 years old, and he may be fading. The other two lagging teams are managed by Clint Hurdle and Terry Francona, mentioned earlier in this thread as two of the worst managers in the history of MLB per Jaffe's work. Hurdle, Francona, and Cox stand out as the three worst 2008 MLB managers by the D3 metric.

 

The current best teams by D3:

 

[table]Team | W | L | D3

Angels | 50 | 34 | 8.0

Twins | 46 | 38 | 6.8

Pirates | 39 | 44 | 6.4

Cardinals | 48 | 37 | 5.1

Marlins | 43 | 40 | 4.9[/table]

[CrespoBlows]

How about after last night's debacle?

[Jayhawk Bill]

How about after last night's debacle?

 

Not yet compiled and posted by BP, but around -5.6. Boston had one more hit, eight more total bases, and only two fewer walks than the Rays did, and Boston lost the game anyway.

 

Even with the bullpen meltdown, Boston should've won.

[ORS]

I didn't finish the game, so I'm curious what moves Francona made that cost them this game. Going to MDC in the 7th with a 3 run lead seems like the right move. When he can get an out after 3 batters, going to Hansen is what I would have done. When he too puts his suck on display, well then you go down the pecking order, and Aardsma's as good a choice as any. This is on our gutless bullpen duo. They dissappoint more than they impress, at least to me.

[Jayhawk Bill]

I didn't finish the game' date=' so I'm curious what moves Francona made that cost them this game. Going to MDC in the 7th with a 3 run lead seems like the right move. When he can get an out after 3 batters, going to Hansen is what I would have done. When he too puts his suck on display, well then you go down the pecking order, and Aardsma's as good a choice as any. This is on our gutless bullpen duo. They dissappoint more than they impress, at least to me.[/quote']

 

1) Terry Francona needed to have Papelbon, who's certainly well-rested, ready to face the heart of the order in the 7th inning. I don't fault using either MDC or Hansen for three batters. I fault Tito for not being ready for the contingency that he'd have to use his closer--his highest-leverage pitcher--in the 7th, not the 9th. Boston's win probability went down 31% while MDC pitched and 45% while Hansen pitched. Those were the crucial moments, and, looking at the batting order, it was foreseeable. By the time Aardsma came in, the game was almost over--his WPA added was 4%, and Lopez only lost 5% despite allowing the runners to score.

 

2) There were two points at which Terry Francona should have pinch-hit for Tek. The first was in the top of the 7th, two on and two out, when the Rays had just walked Youk to get to Tek. Predictably, Tek struck out. Sean Casey would've had an excellent chance of plating one more run. But the other time, with Tek up, tying run on first, and just one out, was just pitiful. Boston still had an 18% chance of salvaging a win in that situation. Not pinch-hitting, and, especially, calling for a hit and run, was terrible management.

 

For WPA percentages, play-by-play and summed by pitcher, check BR box scores:

 

http://www.baseball-reference.com/boxes/TBA/TBA200807020.shtml

[ORS]

While I agree that using Papelbon earlier in the game is something that needs to be addressed, I'm not quite sure I'm ready to fault Francona there relative to the rest of the managers in baseball. None of them would have done that, as all of them have adopted the LaRussian bullpen management strategy to some degree. In other words, a qualified replacement does the same thing in that situation on the road 100 times out of 100. There's nothing to gain by firing and hiring a contemporary replacement. That change in philosophy needs to come from the front office. If they have communicated to Francona that they wish for him to use Papelbon as a true "relief ace", and he has subsequently ignored their directive, then he needs to be canned and a replacement found. Short of that happening, I fault the FO (and the sackless wonder twins for not performing).

 

I'm 100% in agreement on the PH'ing for Tek front. He's a liability at the plate. Captain or not, you PH for weak hitters when you have a chance to either break the game open or tie it up late.

[TheKilo]

The Red Sox have lost their last 5 games by a total of six runs. Guess how many innings their best reliever has pitched.

 

Hint - it's zero.

[Jayhawk Bill]

While I agree that using Papelbon earlier in the game is something that needs to be addressed' date=' I'm not quite sure I'm ready to fault Francona there relative to the rest of the managers in baseball. None of them would have done that, as all of them have adopted the LaRussian bullpen management strategy to some degree. In other words, a qualified replacement does the same thing in that situation on the road 100 times out of 100. There's nothing to gain by firing and hiring a contemporary replacement. That change in philosophy needs to come from the front office. If they have communicated to Francona that they wish for him to use Papelbon as a true "relief ace", and he has subsequently ignored their directive, then he needs to be canned and a replacement found. Short of that happening, I fault the FO (and the sackless wonder twins for not performing).[/quote']

 

1) Papelbon has already been used six times this season in non-save situations, a sixth of his appearances. What hasn't happened is Papelbon's appearing before the eighth inning. It's not that the FO is demanding that Papelbon only appear in the ninth inning for saves, it's that Tito lacks the imagination to use Papelbon at the critical moment.

 

2) Let's rerun a stat from yesterday.

 

Boston relief ERA in save situations: 4.63

Boston relief ERA in non-save situations: 3.82

 

Tito is misusing his pitching staff. I can't make it clearer. :dunno:

 

***

 

BTW, the new D3 is -5.7, worse than Hurdle and the Rockies' -5.6. Tito and the Red Sox have the lowest D3 in MLB of any team that has not fired its manager.

[ORS]

I realize has been used in non-save situations. These are generally when "he needs some work" coming on the heels of 1-run loss on the road where the other team scored late off a less talented reliever. I don't like it, either. Find a qualified replacement who would do it different. There are none. Those that would do it, lack the qualification. Those that are qualified, ie current or former MLB managers, play by the same book.

 

Judged against his peers, I see little room for improvement. This does not say that I disagree with the notion that he lacks imagination. I do, but that isn't what I'm talking about.

 

Those stats are damning, but who has he been using? Here's the breakdown of the team's 83.2 IP that were a save situation...

 

[table]Player|IP|%|SV ERA|NonSV ERA|Synopsis

Papelbon|28|33.5|1.93|2.25|not the problem

Okajima|21.2|25.9|4.15|1.32|meh, but hordes of inherited runners too

MDC|12|14.3|9.00|2.42|teh suck

Hansen|8.2|10.4|7.27|5.06|s*****

Lopez|5.1|6.4|6.75|1.32|bad[/table]

 

That's 75.3 of the IP. He's giving the ball to the right people, but they are folding under the pressure. The only legit case to see more opportunity there is Aardsma, but his walks have to be concerning. Everyone else, ie Timlin, sucks regardless of situation.

[Jayhawk Bill]

I realize has been used in non-save situations. These are generally when "he needs some work" coming on the heels of 1-run loss on the road where the other team scored late off a less talented reliever. I don't like it, either. Find a qualified replacement who would do it different. There are none. Those that would do it, lack the qualification. Those that are qualified, ie current or former MLB managers, play by the same book.

 

Judged against his peers, I see little room for improvement. This does not say that I disagree with the notion that he lacks imagination. I do, but that isn't what I'm talking about.

 

Those stats are damning, but who has he been using? Here's the breakdown of the team's 83.2 IP that were a save situation...

 

[table]Player|SV ERA|NonSV ERA|Synopsis

Papelbon|1.93|2.25|not the problem

Okajima|4.15|1.32|meh, but hordes of inherited runners too

MDC|9.00|2.42|teh suck

Hansen|7.27|5.06|s*****

Lopez|6.75|1.32|bad[/table]

 

That's 75.3 of the IP. He's giving the ball to the right people, but they are folding under the pressure. The only legit case to see more opportunity there is Aardsma, but his walks have to be concerning. Everyone else, ie Timlin, sucks regardless of situation.

 

OK...first a question...

 

Why do Tito's "right people" suck when he goes to them in save situations (usually hold situations, BTW)?

 

Let's get back to that later.

 

***

 

First, a couple of other tables:

 

[table]NAME | Fair RA

Javier Lopez | 2.15

David Aardsma | 2.56

Jonathan Papelbon | 2.66

Chris Smith | 4.94

Hideki Okajima | 5.07

Manny Delcarmen | 5.25

Craig Hansen | 5.88

Mike Timlin | 8.36[/table]

 

Fair RA considers inherited and bequeathed runners. If the guy behind you screws you, you only get charged with the runs an average MLB pitcher would've let score; if you screw the guy ahead of you, or if you save him, the credit for the difference from norms is yours.

 

By this metric, Boston has three obvious choices for late-inning high-leverage use: Papelbon, Lopez, and Aardsma. Are they being used this way?

 

[table]NAME | LEV

Hideki Okajima | 1.66

Jonathan Papelbon | 1.58

Craig Hansen | 1.08

Javier Lopez | 1.06

Mike Timlin | 1.04

Manny Delcarmen | 1.03

David Aardsma | 0.68

Chris Smith | 0.43[/table]

 

No, Tito seems to be using Okajima, repeatedly, in his most critical middle-late inning situations, giving Aardsma (and Smith) the least critical situations, and lumping his four other relief pitchers together into an amorphous state of "sometimes I trust you."

 

Have they equally earned that trust?

 

[table]NAME | WXRL

Jonathan Papelbon | 1.27

Javier Lopez | 0.98

David Aardsma | 0.76

Hideki Okajima | 0.05

Craig Hansen | 0.04

Chris Smith | 0.00

Manny Delcarmen | -0.47

Mike Timlin | -0.75[/table]

 

Well, we all knew that Mike Timlin wasn't having his best year.

 

More importantly, Papelbon, Lopez and Aardsma again rise to the top of the list. They are the three relief pitchers to trust.

 

MDC sinks lower than he should by Fair RA. That is the definition of choking. MDC is our low-leverage situation guy, or trade bait.

 

And, one more note: Hansen is doing far better at home, and far better vs. RHH (.203/.292/.281 allowed). Hansen should be a ROOGY on the road.

 

***

 

Back to that question...Why do Tito's "right people" suck when he goes to them in save situations?

 

1) He's not using Aardsma and Lopez enough.

 

2) He's using Hansen and, especially, Okajima and Delcarmen in critical situations too much.

 

I'd add, with less evidence, that,

 

3) On occasion, Papelbon needs to come in earlier than the ninth, and

 

4) Chris Smith may not suck.

 

But, in summary, ORS, I disagree with your position, "Judged against his peers, I see little room for improvement." I do, just as I see tangible evidence of bad decision-making, evidenced in the micro level by play-by-play decision game management frustrations and at the macro level by D3.

[ORS]

This, of course, presupposes that most MLB managers will supplant "their guy" for a couple of guys that have both failed miserably in the past, and failed for the most part in limited chances during the current season.

 

I agree, Lopez and Aardsma are the hot hands to ride right now. This has only been apparent for about the last month. How often are fans railing for the guy with the hot hand in favor of the guy with the manager's favor? In my viewing experience, it's quite often. Again, this is not a defense of Francona relative to "what should happen", but I can't reasonably think any of his peers would have made a switch yet.

 

Look at the AL guys doing well in D3.

 

Gardenhire: Despite the loss of Neshek, Crain came back this year and is kicking tail. And it doesn't hurt when burrito hound Dennys Reyes is having a career year.

 

Sciosscia: Shields sucked early this year, and the Angels suffered for it, but he was patient to "his guy". Shields has rebounded and all is well.

 

I find it totally unsurprising that the guys who look good via D3, are the same guys who have bullpens that don't have a proclivity to take large steaming turds on the mound when it matters most. This coversation isn't happening if MDC pitches even league average when his team needed him most.

[Jayhawk Bill]

This, of course, presupposes that most MLB managers will supplant "their guy" for a couple of guys that have both failed miserably in the past, and failed for the most part in limited chances during the current season.

 

I agree, Lopez and Aardsma are the hot hands to ride right now. This has only been apparent for about the last month. How often are fans railing for the guy with the hot hand in favor of the guy with the manager's favor? In my viewing experience, it's quite often. Again, this is not a defense of Francona relative to "what should happen", but I can't reasonably think any of his peers would have made a switch yet.

 

Perhaps, but an objective measure of game management, D3, has Tito's team at -5.6 again this morning, second-worst in MLB to the Blue Jays, who have already fired their manager.

 

Look at the AL guys doing well in D3.

 

Gardenhire: Despite the loss of Neshek, Crain came back this year and is kicking tail. And it doesn't hurt when burrito hound Dennys Reyes is having a career year.

 

Sciosscia: Shields sucked early this year, and the Angels suffered for it, but he was patient to "his guy". Shields has rebounded and all is well.

 

I find it totally unsurprising that the guys who look good via D3, are the same guys who have bullpens that don't have a proclivity to take large steaming turds on the mound when it matters most.

 

I did check those teams. Certainly Mike Scioscia has enjoyed good relief pitching when he's needed it. Surprisingly, though, Gardenhire has the same problem as Terry Francona: his relief pitchers have posted a 4.34 ERA in save situations and a 3.31 in non-save situations. What Gardenhire has done, though, is use his best pitchers in the highest-leverage situations: the OPS allowed by his pitchers goes from .715 to .729 to .815 as one goes from high to medium to low-leverage situations. Furthermore, Gardenhire has used his bench players at the moments best-suited to their skills and talents: his substitutes have posted an OPS of .820, contrasted to his starters' OPS of .732.

 

Boston subs have hit a miserable .203/.289/.297, a .586 OPS, despite the team's having had four starting-quality outfielders most of 2008, Sean Casey (.358/.409/.453) available all season to pinch-hit, and cups of coffee from Jed Lowrie, Jonathan Van Every, and Chris Carter, the three of whom combined to hit .333 in roughly a month-and-a-half of combined availability. Terry Francona has only pinch-hit 29 times in 88 games despite having had Cash, Varitek, and Lugo on his team, all of whom have struggled in certain times and situations.

 

This coversation isn't happening if MDC pitches even league average when his team needed him most.

 

Actually, I started the thread a month ago when I saw that injuries would force Terry Francona to have to make managerial decisions. MDC's weird recent appearances may have contributed to the current discussion, but Terry Francona has a history of bad game management that's been concealed by the excellence of the talent made available by the FO and by memories of winning seasons. I pointed out the metric D3 while Tito was still near the MLB mean. Things could've gone either way--but as it worked out, Boston struggled more than it should have and D3 captures that struggle.

 

Edit: You know, MDC came into yesterday's game to get A-Rod to ground out to Lowell with the bases loaded in the seventh. Kudos to Tito--his trusting MDC worked yesterday evening.

 

After two wins, though, the D3 is still -5.6. Tito's decisions aren't generating wins--the Red Sox are outplaying the Yankees in the first two games.

 

Edited again: OK, after Saturday's loss to the Yankees, Tito is at -6.0.

[ORS]

The tried and true, "I'm only going to use my closer in the 9th, and only with a lead of 1-3 runs (on the third Saturday when the moon is less than 1/2 full)" rears its head again yesterday.

 

When Kotchman was up, Morgan chimes in [paraphrased] "He's not going with Okajima for the LHP vs. LHH matchup because Delcarmen throws harder and he needs a strikeout". Then Miller chips in, "Kotchman has the fewest K/AB of AL regulars". So, of course, you leave in the RHP with the dead straight FB in a bad platoon spot. It isn't like there's a guy who has 11.29 K/9 out there in the BP who hasn't pitched since Tuesday, Tito. Oh, that's right, the cosmic alignment isn't right.

 

Until the 7th and 8th innings gain some stability, this will continue to be an issue for a manager who refuses to buck contemporary closer usage. Jacko was right in the Yankee game thread. He had to try and get a few more outs from Beckett and Wake late in the game because these guys have been so unreliable, and that failed. So the starters are failing to give that extra inning, the setup crew is failing to come in and shut it down with runners on, and that leaves only one other option. Use the closer to put out the fire, give the offense a chance to tack on insurance against weaker relievers (ie, not closers), and try the setup crew in a fresh inning with the lead still in hand. If that too fails, what's the big deal? You are exchanging failure for failure, but you are at least trying something different instead of pushing the same button while expecting different results.

[Jayhawk Bill]

Good points, ORS.

 

FWIW, Terry Francona is now at -6.3 D3. That's now the worst in MLB: even the Jays and the Mariners are better, including the stretches so bad that they fired their managers.