Thursday, April 22, 2010

Hillman and the Bullpen.

Jason Kendall is hitting .352, Yuniesky Betancourt is hitting .309…and the Royals are still 6-9. Apparently not even Lucifer can guarantee the Royals victory. If Joe Hardy played for Kansas City, “Damn Yankees” would end with the Royals losing their 100th game of the season.

The Royals incessant failures have allowed another meme to crop up, a meme that is explained well by Craig Brown here:

‘Asked by the traveling reporters if he considered bringing closer Joakim Soria into Tuesday’s seventh inning, Hillman offered this:

“There’s a thought there but, No. 1, it’s a very unusual time for Joakim Soria to pitch in a ballgame. No. 2, you’ve still got those same bats coming up in the ninth in a
higher-leverage situation — because it is the ninth, even if there are no runners on base.”

I added the emphasis because Hillman’s use of term “high leverage” is impressive. It would be more impressive if he knew what the hell he was talking about.

Following his logic, the higher the inning number, the higher the leverage. Sometimes, it actually works that way. Other times, like Tuesday, not so much.’

I’m going to do something a little unexpected here: I’m going to defend Trey Hillman – to a point. I don’t mean to pick on Craig specifically – he’s a great guy and one of the best Royals bloggers in the business. He wasn’t the first person to criticize Hillman for this quote; he just had the misfortune of being the last, or at least the one whose article I was reading (after Rob Neyer linked to it) when I decided to respond.

I’ve written many times (it’s not original; Bill James wrote this a quarter-century ago) that bad teams inevitably focus on their failures on their best players. (Just witness the manner in which the Royals have treated Alex Gordon this season; as Bob Dutton wrote last night, Gordon’s time in the big leagues might hinge on his production while Chris Getz recovers from a strained right oblique”. I like Chris Getz, but the notion that Gordon might lose his job to Getz is nutz.)

But I think that we can add a Fan’s Corollary to The Rule of Bad Teams, that goes something like this:

A fan base that has endured too much losing for too long inevitably focuses on minor or secondary causes for their team’s failures, because the primary cause of those failures are too basic to warrant mentioning.

In the Royals’ case, the root cause of the team’s failure so far this season is pretty simple: the bullpen can’t pitch worth a damn. But that’s too easy and definitive an explanation. It’s hard to admit that the problem is that we have exactly one good reliever, because that’s a problem that no amount of massaging can fix. So instead it’s tempting to say that the real problem is that Trey Hillman is doing a lousy job of using his one good reliever. It’s tempting to say that the problem would be solved by simply using Joakim Soria in the seventh inning.

It was Voltaire who originally said, “The perfect is the enemy of the good.” We could use a little Voltaire right now, and not just because he probably had a better slider than Roman Colon.

Look, I would love to see Soria come in when the Royals are in a jam in the 7th inning, work his way out of it, then close out the game in the 8th and 9th, just like my grandpappy told me Dan Quisenberry did it in the good ol’ days. I fervently believe that the reliever usage patterns of the 1970s and early 1980s were significantly better than those of today. But to think that the Royals, or any team, is going to turn the clock back 30 years in one jump is ridiculous, and holding on to that kind of hope will only lead to frustration.

I’m going to define a “Long Save” as a save in which the pitcher recorded 7 or more outs, i.e. he entered the game in the 7th inning or earlier, but stayed in to close out the game. By that definition, here’s a brief history of the Long Save:

Prior to 1960, the most Long Saves in a season was 9, by Hoyt Wilhelm in 1953.

From 1960 to 1969, a pitcher recorded 9 or more Long Saves in a season 11 times (three times by Wilhelm again); the record was 13, by Bob Lee in 1964.

In the 1970s, eighteen pitchers recorded 9 or more Long Saves in a season, and a 5 Long Save season was accomplished no fewer than 119 times. In 1979, Jim Kern broke Lee’s record with 16 Long Saves.

The Long Save was still prominent in the 1980s; 16 pitchers reached the 9-Long Save mark, and 117 pitchers had at least 5 Long Saves in a season. In 1982, Dan Quisenberry recorded 17 Long Saves, still the all-time record. He had 13 in 1983, and 15 (third-most all-time) in 1984.

But during this decade the counter-revolution began. In the first half of the 80s (1980-1984), there were eleven 9 Long Save seasons, and 68 seasons with 5 Long Saves. From 1985 to 1989, there were just 5 and 49, respectively.

In 1989 and 1990, the Oakland A’s had probably the best bullpen in major league history of a 2-year span. This was partly due to the way Tony La Russa handled Dennis Eckersley, protecting the broken-down starter by using him in one-inning roles almost exclusively. Imitation being the sincerest form of flattery, the Long Save disappeared from the game almost overnight.

In 1990, Duane Ward had 7 Long Saves for the Blue Jays. The following year six different pitchers had 5 Long Saves. Since 1992, only two pitchers – Derek Lowe and Scott Williamson, both in 1999 – have had 5 Long Saves in a season. Since 2000, the only pitcher with more than 5 Long Saves FOR THE DECADE is Tim Wakefield, with 6. That should be a clue that, in virtually every case, the rare Long Save that occurs in today’s game goes to pitchers who are long relievers, not closers, who happened to come into the game in the 6th inning of a 10-2 game, and went ahead and completed the game, thus earning a save by the rule that any pitcher who throws 3 innings to close out a victory gets one.

While it’s possible I’ve missed someone, as best as I can tell there has not been a single instance of a closer pitching more than 2 innings for a save this century. Williamson had a 7-out save for the Reds on May 5th, 2000, but he was sort of the co-closer with Danny Graves at the time. The closest I can find is Salomon Torres, who had a 3-inning save on April 4, 2008, and saved 28 games that year – but his next save wasn’t until May 24th, so it’s pretty clear he wasn’t in the closer’s role yet.

So yes, it would be perfect if Soria became the first closer in over two decades to enter a game in the 7th inning. But it’s not going to happen, and it won’t happen until we first see closers being regularly used in the 8th inning.

And in that category, Hillman is actually ahead of the curve. Let’s call a save of at least 4 outs a Medium Save. Like the Long Save, the Medium Save has become more and more rare over the years, but is still occasionally seen in the wild. In 1983, Quisenberry had 35 (!) Medium Saves – meaning only 10 of his then-record 45 saves were of the 9th-inning variety. In 1993, Jeff Montgomery had 18 Medium Saves – John Wetteland had 19 – and no one has had that many in a season since. But as recently as 2004, John Smoltz had 16 Medium Saves, the most of any pitcher in the last 17 seasons.

Here’s a list of the pitchers with the most Medium Saves in 2009:

8: Brian Wilson

7: Joakim Soria

7: Mariano Rivera

7: Ryan Franklin

7: Andrew Bailey

Now yes, it’s true that Hillman only started stretching Soria out after it was clear he had no choice – he recorded his first one on July 7th, and the second one on July 25th, after the memorable meltdown against Tampa Bay when the Royals lost three straight games in the 8th inning. But at least he did adapt. That’s more than I can say for other members of the organization.

If you want to argue that Hillman is not using Soria correctly, you’ll have to talk to someone else – I agree with you. But I have to point out that Hillman’s usage of his closer is no worse, and in many cases is better, than the way that 29 other managers use theirs. You could respond, as Rob Neyer frequently does, that when you’re the Royals, being “no worse” than other teams isn’t enough – given the inherent disadvantages to your market, you have to do things better than the competition.

I would agree to that as well. But Hillman doesn’t have to use Soria in the 7th inning to accomplish that. He simply needs to use him in the 8th more regularly.

Hillman has shown a willingness to do that. Twice this season, Soria has come in for the save with two outs in the 8th. Not impressed? Consider that he is the only pitcher in baseball with 2 Medium Saves this year.

Really, my only beef with Hillman’s usage of Soria is his bizarre delineation between the four-out save and the six-out save. Hillman has said on multiple occasions that he feels there’s a big difference between the two, which I find curious. To me, the bigger difference would be between the three-out and the four-out save – in the latter case, you come in to pitch, then sit down for 15 minutes, then get up again. Being forced to warm up twice instead of once would seem to me to be a bigger difference than throwing an extra 10 pitches in the 8th.

The refusal to use Soria for more than 4 outs nearly cost them the game yesterday, and did cost Greinke the victory. Yesterday was the perfect day to stretch Soria out a little – he hadn’t pitched since the 18th, and today is an off-day. With a one-run lead on the road, it was madness to start the inning with anyone else – and yet Hillman did just that, bringing in the guy who was in Triple-A ten days ago to start the inning. The Blue Jays predictably tied the game up.

Our story has a happy ending, because in the 9th Robinson Tejeda threw one of the most overpowering innings I have ever witnessed*, and then Alex Gordon had one of the most important hits of his career, a two-out, extra-inning home run off a left-handed pitcher after he had gone 0-for-3 with 2 Ks and a throwing error in the game. Soria got his 3-out save opportunity, and everyone went home happy. But they almost didn’t.

*: Tejeda threw 8 pitches in the inning, all strikes, and all (I believe) fastballs. He struck out Vernon Wells on 3 pitches, got Lyle Overbay to hit a lazy flyball on an 0-1 count, then struck out Jose Bautista on 3 pitches. It was a ridiculous display of power pitching, and proof that if he can just throw strikes, Tejeda can be that bridge to Soria the Royals so desperately need.

What makes Hillman’s refusal to use Soria to start the 8th so maddening is that he finally broke down and gave in to the impulse last season, and it worked beautifully. After the bullpen collapse against Tampa Bay, Soria came in to start the 8th six times the rest of the season. In one of those games, he came in with a 5-4 lead, gave up three runs in the 8th and was pulled after getting just two outs. In the other five games, he pitched two scoreless innings for the save. In another game Hillman brought him in to pitch the bottom of the ninth in a tie game at Detroit; the Royals scored in the 10th and Soria finished out the game to earn the win.

Soria earned 5 saves of the 6-out variety in 2009. No other pitcher in baseball had more than 2. The last pitcher with 5 such saves was Aquilino Lopez in 2003.

The Twins are frequently held up as the smartest team in the division; Joe Posnanski thinks that Ron Gardenhire is the best manager in the game. You know how many times Joe Nathan has gone 6 outs for the save in his career? Once – on June 15th, 2006.

Let’s go around the division. How many 6-out saves does Bobby Jenks have in his career? One – against the Royals, in September of 2005. The last Indians closer to go 6 outs for a save was Steve Karsay, in 2000. The last Tigers closer to go 6 outs was Matt Anderson, in 2001.

So is the difference between the Royals and the other teams in the division really the way that Trey Hillman uses his closer? Or is the difference that the other teams in the division actually have someone to pitch the 8th inning? Which is a bigger mistake – that Hillman didn’t go to Soria in the 8th yesterday, or that he did go to Josh Rupe – and that aside from Soria, Josh Rupe was probably the best option?

If Tejeda had pitched the 8th, and pitched it as well as he would end up pitching the 9th, this is all an academic exercise. And going forward, on a team that’s so starved for middle relief that literally one good outing can move you right behind Soria in the pecking order, we might well see Tejeda in that spot the next time the Royals need a bridge to their closer.

If the Royals continue to struggle this season, Hillman is probably going to get the axe, and I can’t say that I’ll be particularly sad to see him go. But his usage of Soria wouldn’t make my top ten list of reasons why he should be fired. Yes, I would argue that Hillman should use Soria for two innings instead of one, that he should use him in tie games rather than to protect three-run leads, and all that.

But look around baseball. On Saturday night, Jerry Manuel used every other reliever in his bullpen to pitch in a scoreless game in extra innings, before he finally got around to using his closer, Francisco Rodriguez, in the 19th inning. And Manuel out-managed Tony La Russa, who double-switched Matt Holliday out of the game, putting the pitcher in the lineup right behind Albert Pujols; inexplicably hit-and-ran with Pujols at the plate in the 19th inning, costing the Cardinals the game; and made a half-dozen other ridiculous decisions. Managers – even good managers – make sub-optimal decisions all the time. Hillman’s track record in this regard is hardly unusual.

Two years ago, Soria saved 42 games, yet only twice did he pitch more than 3 outs for the save. Yet no one was complaining too loudly about Hillman’s use of his bullpen, because the Royals had Ramon Ramirez and Leo Nunez to pitch the 8th. That the Royals no longer have Ramirez and Nunez to pitch the 8th is regrettable, but don’t pin the blame for that on Hillman.

The Royals have a bullpen of horrors for the second straight season, but the root cause for that is not a failure to manage their personnel. It’s a failure to select their personnel. And the fault for that does not reside primarily with Hillman. If you’re looking to blame someone for that, you have to aim higher.

Having Soria pitch in a tight situation in the 7th or even 6th inning would be perfect, but simply having Soria start the 8th inning to protect a one-run lead would be good. The perfect is the enemy of the good. Let’s concentrate our efforts on convincing Hillman to send Soria out there for six outs instead of four. I still hold out hope that one day baseball teams will go back to the way bullpens were run in 1977. But we have to get back to 1993 first.


Will said...

I think what many of us good statheads want is not merely longer saves, but Soria appearances that don't necessarily mean he finishes the game. True high-leverage uses against the heart of some other team's order in the 7th or 8th. Of course, it is unworkable when the other 50 guys in the bullpen are all absolutely terrible. Its just not realistic to expect that with the current roster, and maybe not with any roster at this time. Hillman's approach to using Soria is still entirely outs-based. He's getting better, but I think that that is still a flawed model.

Unknown said...

I can't say that I've been upset with Hillman this year. His bullpen management seems better this year over last year, but his personnel hasn't performed. Probably due to an overly high babip for the bullpen. And, they aren't good.

adoyleBU said...

I, for one, am glad we sacrificed what could be a decent team right now (you know, as long as they keep hitting .310 or so) for not even half a season of Coco Crisp and that full train wreck of a season of Mike Jacobs last year. That was totally worth giving up what was a pretty good bullpen for this garbage we get to watch today. Even if the bullpen situation does somehow get figured out in the next few weeks (Josh Rupe is the answer!), what are the chances we will still get 15 hits a game? Slim? None? Is a negative probability possible?

Anonymous said...

The save stat has ruined most manager's ability to make use of the bullpen in the most effective manner.

Nathan said...

I pretty much agree with this post. The one caveat I have is this: If we see a repeat of the debacle from last spring, when Soria wasn't used for ten days in a row because the team didn't happen to generate any save situations, I will lose faith in him as a manager. I realize there's a baseball orthodoxy in play here that won't be overthrown in a day. But failing to use your best reliever at all, while the bullpen collapses around him, manifestly defies common sense.

I do thank you for providing some context here, though. We should remember that Hillman has actually done a better job of bullpen management (especially since the 2nd half of last year) than most managers in baseball. He's not perfect, but he's not the source of our problems.

gbewing said...

Hillman is not managing his bullpen better than last year using his ace closer for 4 out saves is hardly evidence of competancy. He still doesn't utilize platoon splits corectly and has turned over his bullpen set up 3 or 4 times in first 2 weeks (before the meltdowns and after) He promotes a AAAA lifer and immediately makes him the primary set up man as the previous primary set up man was relaeased after a 2 batter trial opening day.

This is a lot of justification over nothing. look at the whole pen not just Soria

Anonymous said...

I don't think people are arguing than Hillman use Soria in the 7th and then have him finish the game. Rather, I think the argument is that Hillman should use Soria (or whoever a team's best reliever is) in a situation which has a higher leverage index, thus hoping to get to a point later in the game, with the lead, that has a smaller leverage index (in the 8th and or 9th) for relievers who are less effective than Soria.

rljaco said...

I agree with the Anon post above. Assume you're ahead 2-1, one out in the seventh, team behind has runners at second and first and your starter has had it. Why would you not use Soria THEN, hopefully get through the inning, and give your lesser talents a clean start in the last two innings. Sure, great if Soria can go an extra inning -- but aren't you just better off having him pitch the seventh in that example?

John said...

I've always found it interesting that LaRussa's A's of the 1988-92 period were able to so definitively change the way relievers were handled, and in all the years since, no one's thought to try anything else, even when the results cry out for it.

LaRussa set up the bullpen the way he did because that fit the talent he had at the time. If you look at his bullpen usage patterns for his first championship team, the '83 White Sox, it was very, very different. He often used both Dennis Lamp and Salome Barojas for long stints late in games.

One thing I remember vividly about LaRussa's heyday in Oakland was that everyone went along except Sparky Anderson. He kept using Mike Henneman in the old-school manner for years, sometimes using him exactly like fans want Soria used now, even if it means someone else pitches the 9th. Like the Royals of today, those Tigers didn't have anyone else decent.

I think part of the reticence on the part of managers to use closers this way is because of their agents. All you have to do is look at how K-Rod got a ton of money because he got 62 saves, even though he wasn't all that great. Henneman pitched for Sparky the way he did because the Tigers paid him commensurate with a quality relief-ace. But does anyone think Soria, used in the same pattern, would today be paid as much for a 25-save season as a 45-save season? The minute Hillman started using him that way, Soria's agent would be calling Dayton Moore.

Clint said...

he doesn't have to have a "long save" to just get out of trouble.. who says he needs to pitch the 8th and 9th.. i'd rather have our best pitcher pitch in the highest leverage situation and the rest try and get through an inning when there is NO ONE ON BASE.

Anonymous said...

>>I, for one, am glad we sacrificed what could be a decent team right now (you know, as long as they keep hitting .310 or so) for not even half a season of Coco Crisp>>

It's not really fair to judge the Crisp trade that way. Did you know that his season would be cut short? If not, how can you complain because Moore couldn't see into the future either? Crisp wasn't especially injury prone before they signed him. In the five previous seasons he averaged 130 games played.

Anonymous said...

>problem is that we have exactly one good reliever>

How much better does John Parrish have to pitch before you consider him "good"?

His ERA and inning pitched match Soria's exactly.

In the long run he's never going to be as good as Soria,of course. But, he's been great so far and doesn't deserve to be lumped in with the rest of them

Anonymous said...

The Cubs have moved a struggling starter (Zambrano) to the bullpen. Maybe the Royals should do the same with Meche.

Mark said...

I don't follow the Royals as closely or with as much discernment as you but I have a separate complaint about the bullpen. During spring traning it was the area of the team that seemed to be the most concerned about options. Who do we keep, who do we send down, I don't know lets look at their options. It plays to a kind of negative environment that seems to be circulating there. I don't know if it is possible to learn anything in Arizona but when Maier has good(great?) spring and can't get off the bench and the bullpen implodes almost before first contact it does make you wonder.

Wabbitkiller said...

Here's an interesting question to ponder: Which is more incompetent: Hillman, or Dayton Moore?

Matt Davis said...

This was a side point in your column, but I'm still curious: When you say "I like Chris Getz, but the notion that Gordon might lose his job to Getz is nutz." are you saying it's nutz because there's no way in hell the Royals are going to do it or because it doesn't make baseball sense? I think a very good case could be made for having Getz start at second because his defense there is superior to Callaspo's. But you have to have Callaspo in the lineup, especially seeing as he strikes out three times less than Gordon. I think Callaspo plays third and Gordon is the odd man out unless he can show he's not going to give away a free out with the strikeout every 4 at bats.

Chuck said...

Honestly, people, think about this. Soria enters in the seventh to put out a fire, and let's say he goes on to pitch the eighth as well, holding the lead. Then Tejeda, or Rupe or Parrish or whoever, enters in the ninth and blows the lead and game. Hillman might not survive the night, and would certainly be fired soon threafter if he did. You're not asking the man to be unconventional, you're asking him to be a veritable profile in courage. It's Not. Going. To Happen.

Anonymous said...

Crisp wasn't especially injury prone before they signed him. In the five previous seasons he averaged 130 games played.

I read this comment, and several blood vessels in my brain exploded. Averaging 130 games is pretty much precisely the definition of "injury-prone"; it means you missed a MONTH every season!

When Crisp was signed, his injury risk was one of the things we kept fretting about right from the START. I'm just absolutely gobsmacked that anyone could even make the above comment.

As for the bullpen, I can't blame Hillman much. I can't even blame him for bringing in Rupe -- not because he was the "best available option", but because he'd actually been GOOD so far. (He'd had some problems a couple of days previous, but it must be noted in all fairness that he had induced groundballs which SHOULD have resulted in three recorded outs but was rewarded by the incompetent Royal defense with only ONE.)

Anonymous said...

The Royals get very litle ink outside KC. If he did, no one would notice, or at the least he might get called out as being "gutsy". However, it's kind of like the football team that goes for it on 4th and 2. It just shows a lack of confidence in the Defense...

Going to Soria 1st out of the pen sends a really bad signal to the pen, and the writers, on what Hillman really thinks of his bullpen... Is he really gutsy enough to draw that much attention to the pen that DM handed him?

Jim M said...

Cruz and Mendoza out... Chen and Thompson in... not sure how this trade-off will pan out, but never understood how they struggled to make a decision on the bullpen in spring training, and then brought in Mendoza, so this has got to be a big plus on the Mendoza line.

Anonymous said...

Crudz and the Mario Mendoza of pitchers shouldn't have been released or DFA'd. They should have been asked to commit seppuku or drink hemlock!

Carl Willingham said...

To John who posted earlier, that was a great post my man. Some old school referrances mixed with a great point about agents and pitcher usage. Thanks.

Anonymous said...

Why don't we move Meche to the bullpen?

Dan said...

I'm really surprised that Rany missed Craig's point, which many have already pointed out. The best use of the best reliever in your bullpen is not necessarily as a closer. Bring him in anytime in the last three innings when the scoring threat is greatest (men are on base, middle of the order, etc). Let him get out of that jam, and whether he gets three outs or stay in for six, he got you out of the jam and you still have the lead. Worry about keeping a lead in the ninth when there is actually a lead in the ninth to worry about.

irishguy said...

Once again, 10,000 words when a few will suffice:

When you have one pitcher in an 8-man bullpen, it doesn't really matter when you use the guy.