Sabermetrician/Relief Pitcher Glen Perkins Comments on Bullpen Roles

[jay]

I tend to usually look away when I see 'Article By: Jim Souhan', but this most recent piece features noted sabermetrician and current Twins closer Glen Perkins discussing bullpen roles:

Souhan: Perkins relieved to be in Twins' bullpen | StarTribune.com

 

‘‘If I were a manager, I would probably construct my bullpen in the same way we have ours constructed, because I know we can’t do what sabermetricians want us to do,’’ Perkins said. “We just can’t.’’

 

‘‘You have to get all the outs and the easiest way to do that is to say, ‘You’re pitching here and you’re pitching here,’ because we’re creatures of habit."

 

I've always been a huge supporter of the view to use your best reliever(s) in the most crucial situations. I'd be curious to see how the article or Glen's response would be different if it wasn't Souhan, but this at least makes me question the line of thought and certainly serves as a reminder that baseball players have feelings too despite the fact that sometimes we expect them to perform like machines.

[Mike Sixel]

It is an interesting theory, which certainly has validity. We don't have much data on the other side of the argument, so it's hard to draw many conclusions. Much like my belief that stats assume starts are not independent events, stats-believers (I'm one, generally) speak with a certainty about some things that may not be so certain.

 

IME, emotion and comfort and other stuff plays a great role in performance. There is probably some middle ground, here, where the vast majority of the time it does not matter in what order pitchers pitch, so give them a clear role. But I would guess that occassionally veering from that role would not be a bad idea either. To me, it is the ridigidity of the approach that is an issue, not that the approach is used most of the time, but that it is used all of the time.

 

But I'm open to the possibility that rigidity increases the likelihood of success.

[drjim]

I question how rigid it is. When things are working it can seem automatic (perhaps it should be) but if there is big struggles or it is a key game there is some adjustment.

 

Chief speaks very wisely on this in my opinion. It is difficult to always have the best reliever warmed up for every situation and most baseball games are close. The best solution is to get more good relievers.

[Mike Sixel]

Agreed, the best idea is to have more good players availalbe.

[Otto von Ballpark]

First of all, as candid as Perkins may be, I still would not expect him to publicly disagree with his manager on something like this.

 

That said, I'm guessing that Perkins and most pitchers genuinely believe this right now, but that doesn't necessarily mean it is actually optimal. They are creatures of habit, so they naturally prefer the status quo -- their current habits. Warm up your closer in the 8th inning once a week, and eventually that could become part of their habit too.

 

I admit, that may be easier said than done, but it doesn't have to be really radical or aggressive. The Phillies have some heart-of-the-order lefties due up in the 8th inning of a 1-run game? I'd have no problem warming up Perkins at that point -- either you will bring him in to squash a rally in the 8th and/or to close it out in the 9th as usual. Do we really think Perkins is incapable of shifting his warmup routine by one inning from time to time? Or incapable of starting an appearance mid-inning with possible men on base?

 

I am definitely not suggesting sitting him up or down frequently, rushing his warmups, or warming him up without using him. Those are more than "habits" and should be cardinal rules that should apply to all relievers. I just think it's sub-optimal to default to using your ace reliever/closer the least of all of your relief pitchers.

[jay]

To me, it is the ridigidity of the approach that is an issue, not that the approach is used most of the time, but that it is used all of the time.

 

Great points by mike and everyone else. I'd agree completely that the rigid approach is what irks me as well.

 

I wonder if bullpens could evolve to have a role that is centered on coming in the game for high leverage situations with guys on base. The 'old school' line of thought delineates clear roles, so why can't we compromise and have that be one? It'd be someone who can warm up quickly, high GB% and K%, and has shown they can 'handle the stress'. I'd pay more for that guy than someone who has gotten the illustrious save stat many times.

[Otto von Ballpark]

It'd be someone who can warm up quickly, high GB% and K%, and has shown they can 'handle the stress'.

I think the "warm up quickly" part might be an issue. Wondering how much warmup time really varies among relievers? Plus "high leverage with guys on base" situations can develop pretty quickly, and anywhere from the 6th-7th inning onward.

 

If you want to push the current model even further, if a reliever is pitching well, I'd much rather see him throw two innings instead of one, maybe even three innings instead of two if he's really cruising, even if that means he's usually unavailable the next day and he posts fewer overall appearances. "Ride the hot hand", if you will. I don't really understand always limiting guys to one inning so they can pitch the next day if needed, when you have 6-7 of these pitchers on your active roster. You don't really need 6-7 relief pitchers available every game if you are willing to have any shred of creativity or flexibility (but maybe Gardenhire and Perkins really mean near-100% rigidity when they talk about roles/habits).

 

But I'd be happy if, like I mentioned previously, teams weren't afraid to warm up their relief "aces" like Perkins a little early in close games. To offset any increase in appearances/innings, you could never, ever warm guys up, sit them down, or otherwise deploy them based solely on the save rule. Three run ninth inning lead should not mean automatic Glen Perkins appearance. I absolutely hate when a team stretches a 3-run lead to a 4-run lead in the 8th inning, and the announcer says that the closer is sitting down in the pen. Or the closer is loosening up forever in a tie game just waiting for a lead while a parade of lesser relievers come and go. Or the closer has to come into a blowout just to "get some work in" because a "save situation" never developed over the previous 3-4 games despite them being close.

[jay]

I think the "warm up quickly" part might be an issue. Wondering how much warmup time really varies among relievers? Plus "high leverage with guys on base" situations can develop pretty quickly, and anywhere from the 6th-7th inning onward.

 

You're absolutely right, just trying to address one of the main counterpoints I've heard as to why it won't work. Maybe it'd just be more of a mindset that I'm going to get up, warm up as soon as I can, and go in to face guys on base.

 

Maybe the role would need a stat name to really catch on? I nominate a 'stop' and the pitcher is called the 'stopper'. Pretty catchy, eh? To qualify for a stop, the pitcher has to enter the game in the 6th inning or later with a batter at the plate that can either tie the score or take the lead and he must finish the inning with the lead intact.

[nicksaviking]

First of all, as candid as Perkins may be, I still would not expect him to publicly disagree with his manager on something like this.

 

Not to mention that Perkins just so happens to have the high profile, high paying job in the bullpen. He gets to be the closer under the traditional system.

 

There are times to stray from the traditional role of the bullpen, but I would bet the two theories wouldn't stray as often as we think. For instance, there's a good chance Jared Burton pitches the 8th inning last night under either scenario. He would have blown the lead either way. The high leverage situations tend to be toward the end of the game anyway.

[Otto von Ballpark]

There are times to stray from the traditional role of the bullpen, but I would bet the two theories wouldn't stray as often as we think. For instance, there's a good chance Jared Burton pitches the 8th inning last night under either scenario. He would have blown the lead either way. The high leverage situations tend to be toward the end of the game anyway.

 

No doubt. I actually wasn't referring to last night's game at all -- earlier this week, Gardy used Duensing in the 8th when I thought he could have easily used Perkins. Why is it preferred to use a generic lefty reliever to preserve a lead with 4-5 outs to go, but not your ace dominant left reliever until there are only 3 outs to go? Can you imagine if the 2010 Twins pinch-hit with Jim Thome exclusively in the 9th inning, and would instead use Casilla/Tolbert/Punto for all 8th inning pinch-hitting opportunities even if a good situation developed in the 8th?

 

And if a team has two good arms leading the pen (like Perkins and Burton), there is no real reason to designate one exclusively for save opportunities and the ninth inning, especially if one if a lefty and the other is a righty. Currently, the setup guy probably gets overused and the closer gets underused.

 

I know these slight modifications to reliever usage wouldn't add more than a couple wins over a season max, but they do bug me (just like batting a crappy hitter at the top of the lineup instead of the bottom). Especially when the current model is linked so much to the save stat.

 

I wonder what would happen if the rule was changed to require record 4 or more outs to get a save, otherwise you'd only get a "hold"? Would we magically start to see more closers entering in the 8th inning? Would they lose significant effectiveness if forced to occasionally go more than 1 inning?