Showing posts with label Bullpen. Show all posts
Showing posts with label Bullpen. Show all posts

Friday, December 10, 2010

Kevin Gregg

Jeff Zrebiec at The Sun has news today that the Orioles have a 2-year, $8-$10 million offer out to former Blue Jays closer Kevin Gregg.  Jeff is wondering whether Gregg would "unseat" Koji as the closer if he signs.  Leaving aside whether or not the O's should be throwing millions of dollars at free agent relievers (they shouldn't be), having Gregg close games might not actually be a bad idea.  He's a Proven Closer (TM), though not a great pitcher (career 8.3 K/9, 3.7 BB/9), meaning that the Orioles don't really want him pitching in the highest leverage situations.  Moving Koji--and his great strikeout numbers--into potentially higher leverage situations with runners on base in the 7th and 8th, could be a better use of bullpen assets.  I don't know if this is how MacPhail is approaching the situation, but if that's how it works out, we probably shouldn't complain too much.

Sunday, May 16, 2010

Relief Pitching

After last night's meltdown, I went looking at some Oriole relief pitching stats and came across this gem: during 22.2 IP in the 9th inning, Oriole pitchers have allowed a .340/.431/.505 line and a 7.15 ERA.  Yikes.

Monday, April 26, 2010

The Bullpen

With good reason, the Orioles bullpen has been the target of much wailing and gnashing of teeth in the early going.  So imagine my surprise today when I checked the pitching splits page at baseball-reference.com and saw that Orioles relievers (4.70) have a lower ERA than do Orioles starters (4.75)!  How is this possible?  There are two big reasons:

1) Brad Bergesen.  He's made three starts, combining for 10.1 IP and a 12.19 ERA.  Strip out those three starts and the rest of the O's starters have a 3.97 ERA over 99.2 IP, which is not too shabby at all.
2) Inherited runners.  As of April 26, O's starters have passed off 16 baserunners to the bullpen.  Nine of them have scored.  The league average rate for inherited runners scored is 34% (across all situations; it is possible that O's starters have biased their inherited runners toward high run expectancy situations like second and third with no outs).  If O's relievers could acheive that league-average percentage, the starters would have been charged with 4 or 5 fewer runs, bringing their collective ERA to 4.34 or 4.42.  Control for Bergesen (who, ironically, has only had 1 of his 3 bequeathed runners score) and the top four O's starters have an ERA in the 3.60 range.

As for the more skill-dependent numbers, Orioles starters are performing better than their relief counterparts there, too:

K/BB ratio: 2.40 for O's starters, 1.85 for O's relievers
HR/9: 0.90 for O's starters, 1.17 for O's relievers

So, yes, the Orioles rotation has been much better than the bullpen, especially when you look at the numbers without Brad Bergesen.  With the offense finally scoring a few runs this weekend, the bullpen needs to start pulling its weight to get the O's a few more wins.
_______________________________

Thursday, February 18, 2010

Koji Uehara


It's official: Koji Uehara is headed to the bullpen. I happen to think he will perform very well in this role, mainly because he pitched fairly well as a starter last season. Brittany Ghiroli also makes another point: he pitches much better the first time through the batting order. Then again, so do all starting pitchers. Koji, though, is extreme:



tOPS 1sttOPS 2ndtOPS 3rd
Uehara51108160
League94105111


tOPS is simply the OPS allowed relative to a player's (or the league's) normal. So if a player performs at 100, he is performing at his average level. Below 100 for pitcher, means he pitched better than his normal.

As you can see, all pitchers tend to perform more poorly as they advance through the game; this is part of the reason why pitchers post better numbers as relievers than as starters. You can also see that Uehara's splits were far greater than the rest of the league. If he pitches out of the bullpen like he did his first time through the order (.202/.224/.337 with a 9:1 strikeout to walk ratio), the O's will have a shut down reliever.

(I apologize for the looks of that table; I'm very, very rusty on my html.)
______________________________

Wednesday, January 20, 2010

Relief Pitching


Back in the dark ages of 2006, the Orioles had a really bad bullpen (I know, hard to believe). A relief corps anchored by LaTroy Hawkins, Bruce Chen, Russ Ortiz, John Halama, and Todd Williams somehow managed to punch up a 5.27 ERA and 1.55 WHIP in 512.1 IP. In response, the club decided to spend big money on middle relievers in the free agent market: Danys Baez (3 years, $19 million), Jamie Walker (3 years, $12.0 million), and Chad Bradford (3 years, $10.5 million) received a combined $41.5 million. The results? Disastrous.

Over the past three seasons (and including the season and a half Bradford spent with Tampa Bay), those three "pitchers" (and I use that term loosely) were worth a combined 0.8 wins above replacement, with Bradford the only one earning positive returns. That's right, the Orioles paid $51.9 million per marginal win! Granted, the rest of the league wasn't so good either that offseason. Seven relievers (Baez, Walker, Bradford, Keith Foulke, Guillermo Mota, Scott Schoeneweis, and Justin Speier) received $80.3 million in guaranteed money and returned -0.8 wins to their clubs! I think it's safe to say that the free agent market is not a good place to go looking for wins from your bullpen.
_________________________________

Wednesday, December 16, 2009

Mike Gonzalez?


Well, this lowers my faith in Andy MacPhail. I don't care what the terms are; signing a relief pitcher at the cost of a second round draft pick is not a move the Orioles should be making. If the baseball industry has learned anything in the last few years it is that relievers are largely replaceable. I don't know why the Orioles need to bring in a "closer" when that costs them both dollars and a draft pick. Is Gonzalez a good pitcher? Yes; he struck out 90 batters in 74.1 IP last season. But, he also made 80 appearances (previous career high: 54), and the year prior he pitched just 33.2 innings. And in 2007? 17.0 innings. Why the Orioles want an injury prone reliever that costs a draft pick, I have no idea. I blame Peter Schmuck and the other rabble-rousers at the Sun.
____________________________

Tuesday, December 8, 2009

Bullpen Psychology


Ivy League to MLB has an interesting post today on the psychology effects of bullpen role definitions. I agree with this concept, but not necessarily the conclusion. I wrote about this earlier in the season, and respond in the comments:
The problem isn't that bullpen guys have defined roles; it is how those roles
are defined. I think that a bullpen coach who is actively engaged and
effectively communicates with the pitching staff could easily get a 'pen
prepared for non-closer centric decision making.
_________________________

Friday, May 29, 2009

Bullpen


More ineffective bullpenning from Trembley. More later...

Follow Up: Sorry I couldn't post this last night, but it is much easier to type on a real keyboard than on my BlackBerry while trying to watch the game. Last night's mistake was a relatively minor one, but an obvious error nonetheless. After the 8th inning, Brad Bergesen, who had pitched very well, had thrown, I believe, 98 pitches. Aside from his start at New York two weeks ago when he threw 107 pitches over 6.1 innings, 98 was his highest pitch count of the season and 8.0 innings was easily his longest outing. Plus, the heart of the Detroit order was due up: Polanco, Ordonez, and Cabrera. Assuming that you, as manager, will allow Bergesen to begin the 9th inning--a decision with which I completely agree--it would be reasonable to anticipate that Bergesen may get into a bit of trouble. After all, he's reaching uncharted territory on his pitch count and the Tigers are sending their best hitters to the plate. Thus, any manager thinking ahead would have his next pitcher ready to enter the game on a moment's notice, correct? Well, this is where we discover Dave Trembley may not always be thinking ahead.

While the Orioles batted in the bottom of the 8th, I commented that the Orioles bullpen was very quiet. "I agree that Bergesen should start the 9th, but what if he gets in trouble? Two runners on and a quick 3-run homer makes this a completely different ballgame. Shouldn't the Orioles have someone hot at the start of the inning?" Once Bergesen went back out for the 9th, Trembley apparently realized the exact same thing and Jim Johnson began to throw in the bullpen. Predictably, Placido Polanco laced the first pitch of the inning for a double to left and it was obvious that Trembley needed to stall. Following some planned wandering around the mound by Bergesen, out comes the pitching coach. After the umpire breaks up the conference, and Bergesen--who is an exceptionally quick worker on the mound--begins to pitch to Magglio Ordonez. Of course, at this point his rhythm has been disrupted due to the forced stalling in order to get a reliever loose. Ordonez worked a 2-2 count and ground a single through the third base side and runners were on the corners. Unfortunately, Johnson still wasn't quite loose, so more stalling was needed. At last, Trembley emerges from the dugout to pull his starter; Bergesen gets a well-deserved standing ovation, Johnson retires the side and the Orioles win. We all go home happy, but a few minutes later than necessary.

This may seem minor, but what if Johnson had come in and walked his first batter or given up a quick double? Or had shown in the bullpen an inability to get the ball over the plate today? By not planning ahead--even by just two or three minutes--Trembley could have turned an easy win into a nailbiter. I think that Trembley is just the right guy to have in charge as the club gets younger; he stated in the paper just a few days ago that he enjoys the teaching aspects of his job most. Look at the progress made by Adam Jones and Nick Markakis and at the intensity with which the team plays on a daily basis with Trembley at the helm and it is difficult to argue that he isn't a good fit for a rebuilding team. Unfortunately, I do not think he possesses the skills that will be necessary to guide the Orioles to contention once all the young talent is established. I hope that Andy MacPhail--and Dave Trembley--can see his shortcomings and that there is a succession plan in place.
___________________________________________

Thursday, May 21, 2009

Chris Ray


Chris Ray ranks among my favorite Orioles; being a college teammate with someone tends to make your root for them, especially when they play on your favorite baseball team. Needless to say, his performance so far this year has been disappointing, if not wholly unexpected when considering his return from Tommy John surgery. Dave Trembley has been disappointed, too, and there is word today that Ray's role may be changing until he finds more consistency in his delivery. I have just one suggestion for our fearless manager: stop using him against the Yankees.

Chris Ray, career vs. Yankees: 1-4, 8.35 ERA, 5.89 BB/9, 2.95 HR/9 in 18.1 IP
Chris Ray, career vs. all other teams: 9-9, 3.01 ERA, 3.76 BB/9, 1.00 HR/9 in 143.2 IP

Chris Ray, 2009 vs. Yankees: 30.86 ERA, 7.71 HR/9 in 2.1 IP
Chris Ray, 2009 vs. all other teams: 3.48 ERA, 0.00 HR/9 in 10.1 IP

I'm not a big believer in micro-splits such as these, but they're interesting nonetheless. Am I cherry picking stats? Yes. But, when you look at Ray's performance against some individual batters, it's easy to see why he struggles against the Evil Empire:

Johnny Damon: .583/.583/.917, 1 homer in 12 plate appearances
Derek Jeter: .429/.636/.571, 6 RBI in 11 plate appearances
Alex Rodriguez: .500/.556/1.375, 2 homers in 9 plate appearances

Maybe Dave Trembley will keep those numbers in mind tonight if he needs someone to pitch in a high leverage 7th inning.
____________________________________

Thursday, May 7, 2009

It's Official


Dave Trembley is incapable of handling a bullpen.

Top of the 7th, 4-3 lead, top of the lineup due up for the Twins:

Span (L)
Tolbert (S)
Mauer (L)
Morneau (L)
Kubel (L)

Now is the perfect time to bring in George Sherrill. You've stated that he's lost the closer's job, and it's a late inning situation with key lefties coming up for the Twins. Why not ask Sherrill to go five batters, bring in Ray for the next part of the lineup and then have Johnson (who shows no major platoon split) close it out? Or Baez, who has been pitching well? I give up hope for Dave Trembley as an Orioles manager.

Update: Hey, look! George Sherrill can shut down the lefties! I stand by my earlier statements: you bring Sherrill in for the 7th and it may not have been necessary to score in the bottom of the 8th to take the lead.
___________________________

Tuesday, May 5, 2009

Closer by Committee


The Rays executed the Closer by Committee to perfection in the 9th inning of today's win over the Orioles, and Jim Palmer correctly pointed out that this is what we are (hopefully) likely to see from the Orioles. After Matt Garza, the Rays starter, gave up a lead-off double to Adam Jones, Rays Manager Joe Maddon went to lefty specialist Brian Shouse. He promptly induced dribblers from Nick Markakis and Aubrey Huff. With the righthanded Melvin Mora up next for the O's, Maddon went back to the pen. Troy Percival came on, and three pitches later, Mora was down on strikes, and Tampa Bay had a win. Percival may have gotten the save, but Shouse did just as much heavy lifting. Each should be quite pleased with his contribution to the victory, regardless of who records the (S) in the box score. Let's hope Dave Trembley was taking notes.
_________________________

Monday, May 4, 2009

More Bullpen


Does Dave Trembley read the Birds Nest???

I think we can rule out George Sherrill as a fan of mine: "Closer-by-committee doesn't work and won't work," Sherrill said. "We'll have to sit down and talk about it and figure out what to do." And this is why the manager assigns the roles.
__________________________

Sunday, April 26, 2009

The Bullpen


Even the name of this post is enough to give me nightmares....

I went to the game on Friday night, and was obviously disappointed that the Orioles, ahead 4-1 going to the 7th inning, were unable to come away with a victory. More disappointing, however, were Dave Trembley's comments after the game about bullpen usage, calling Chris Ray his 7th inning guy, Jim Johnson his 8th inning guy and George Sherrill his 9th inning guy. Managers love to give their bullpen guys set roles, and those bullpen guys love to have innings define those roles. As Chris Ray said, "You don't have to jump every time the phone rings when you know what your role is." That's all well and good, but shouldn't relievers who are paying to attention to the flow of the game be able to anticipate when their manager might need them? Why do "defined roles" have to be defined in terms of innings?

The Orioles bullpen is filled with good match-up relievers, but no real dominant pitchers. Here are the career splits of those three key relievers:

George Sherrill: .169/.232./.280 against lefties, .268/.384/.402 against righties
Chris Ray: .192/.256/.311 against righties, .244/.345/.437 against lefties
Jim Johnson: .260/.328/.331 against righties, .237/.346/.259 against lefties

Johnson actually shows a bit of a reverse platoon split, but both Sherrill and Ray exhibit the traditional tendency to outperform against batters of the same hand. Since both have experience, and some moderate success, in the closer's role, it seems to me that the Orioles bullpen is tailor made for roles defined by the situation, not by the inning. Let's examine Friday night's game without being constrained by the inning.

In the top of the 7th, Koji Uehara was beginning to tire. He'd thrown very well, allowing just one run over 6.0 innings. Hank Blalock (batting left) led off the inning with a homer, Marlon Byrd (batting right) popped foul to third and Nelson Cruz (right) flew out to deep center. At this point, Uehara had thrown 94 pitches and Chris Davis (left) was due up. Davis homered off Uehara back on April 13, has prodigious power, and Uehara had already allowed a homer to lefty Blalock in the inning. Trembley went to specialist Jamie Walker, a move with which I agree. Unfortunately, Davis dribbled a swinging bunt and reached base on an infield single.

Now, decision time looms for Trembley. Walker is pretty much a LOOGY (Left-handed One Out Guy), so it makes sense that he makes another call to the pen, right? Well, Jarrod Saltalamacchia (switch) is due up for the Rangers, followed by Elvis Andrus (right), Ian Kinsler (right), Michael Young (right) and Josh Hamilton (left). Trembley obviously won't let Walker face more than Saltalamacchia, but should he leave him in to face the Rangers catcher, or should he call for a right-hander to turn Saltalamacchia around? Let's look at Saltalamacchia's career splits: .214/.262/.325 batting righty and .289/.359/.451 as a lefty. In my opinion, Walker should face Saltalamacchia. Instead, since this is the 7th inning, Trembley goes to the pen for Chris Ray. Not surprisingly, Saltalamacchia (batting left) singles and Andrus (batting right) grounds out to end the inning. Ray needed just 10 pitches to get the final out of the 7th.

Instead of leaving a well-rested Chris Ray (he last pitched three nights ago on April 21) in to face Kinsler and Young, Trembley sticks to his formula and calls for Jim Johnson to start the 8th. After retiring Kinsler on a bunt, Young singles, Hamilton lines out, Blalock singles, Byrd reaches on an infield single (scoring a run to make it 4-3), and then Cruz reaches on an infield single. With two men on in the top of the 8th, Trembley rightly senses this is potentially the key at-bat and summons his nominal closer, George Sherrill, to face Chris Davis. Davis strikes out. Crisis averted, right? Well, maybe. Unfortunately, by rigidly sticking to his bullpen formula, Trembley has locked himself into a 9th inning with Sherrill on the mound. Which wouldn't necessarily be a problem, but there are no left-handed batters due up in the 9th. And, as we all know, Kinsler singled, Michael Young homered and Orioles fans went home disappointed.

What could Trembley have done differently? If he didn't want Walker to face Saltalamacchia, he could have still allowed Chris Ray to start the 8th. Kinsler and Young were both at a bigger disadvantage against Ray than against Johnson. Pending the results to those first two at-bats, you could either allow Ray to face Hamilton or summon Sherrill from the pen to face the lefty/lefty duo of Hamilton and Blalock. If Sherrill then runs into trouble, you still have Johnson available and there is just one more lefty in the Rangers lineup (Chris Davis). Unfortunately, that doesn't fit neatly with Trembley's innings constrained roles.

I understand that role definition is important. Players want to know what their manager expects of them, and they need to be mentally prepared so that they can physically succeed. Thus, Trembley should change his definitions, and I propose the following framework.

Jamie Walker: situational match-up lefty (LOOGY)
Chris Ray: late-inning, key situations where the hitters due up are predominantly right-handed
George Sherrill: late-inning, key situations where the hitters due up are predominantly left-handed
Jim Johnson: late-inning, key situations where there is no lineup bias

I know for sure that Chris Ray is a smart guy, and I'm assuming that George Sherrill and Jim Johnson can both read a lineup card. If not, let's hope Alan Dunn (bullpen coach) is able to do so. If Jamie Walker can be flexible as to the inning in which he pitches, why can't the rest of the pen? Because they lack a true stopper, Trembley needs to be flexible in his bullpen usage. Luckily for him, he has complementary weapons, Ray and Sherrill, that lend themselves very nicely to this flexibility. Let's hope that Trembley can understand his bullpen's very real strengths--and corresponding weaknesses--before the Orioles lose too many more games in the late innings.
____________________________