Bruce Miles on Roles in the Bullpen

[Hairyducked Idiot]

No, it wouldn't. History has shown us that the best high-leverage relievers usually end up with roughly the same value contributed as the best starters.

 

Win Share totals disagree with you. Over the past four years, the best reliever year (2004 Brad Lidge - 17.2 pitching win shares) comes in as the 38th best pitching year over those past four years, the equivalent of 2007 Zambrano/Lilly.

 

A good example of a pitcher that was both an effective starter and reliever is John Smoltz. In 2004, he was the fourth most effective reliever in baseball (Lidge, Gagne, Benitez). He finished the year with 12.1 Pitching Win Shares. The next three years, he was an effective starter, with Win Share totals of 21.1, 19.4, and 17.5.

 

That's because win shares are awarded to pitchers based on total runs prevented, exactly the sort of system that underestimates a high-leverage reliever. You might as well just use VORP.

 

What are your specific problems with SNLVAR and WRXL? Why are they wrong when they show the best relievers and starters to be closely packed?

 

I think your assumptions lean heavily on the concepts of win probability, which are really quite arbitrary and provide little predictive value. It does provide for some fun charts to look at, though.

 

Win probability is an incredibly important concept to bullpen usage. If we ignore it, then we shouldn't have any bullpen roles at all. Just pull a name out of a hat every time the pen is needed.

 

And it's the exact opposite of arbitrary. Arbitrary would be if the numbers were made up. They aren't.

[CubColtPacer]

Keep Wood exactly where he is.

 

Move Marmol to the rotation.

 

Cut Marquis.

 

Win.

 

Marmol in his career as a starter - meh.

 

Marmol in his career as a reliever - ridiculoawesome.

 

Shrug.

]

 

 

This. Plus:

 

Marmol's Expected Wins Added Above Replacement: 2.516, second in all of baseball behind Brad Lidge.

 

Support Neutral Win-Loss Above Replacement (essentially the equivalent stat for a starter):

Zambrano: 2.6, Dempster: 1.8.

 

A high-leverage reliever is every bit as valuable as an ace starter.

 

You keep saying this, but you neglect to take into account that a reliever, even the most overworked reliever in baseball, is throwing about 90 innings. A solid starter is generally throwing at least twice that for the year.

 

So, then, to maximize overall team value, the most effective pitchers on the team should be used in situations where they throw the majority of the innings. I'd rather see Marmol get 200 innings and Marquis get 90 than the other way around. Even if a Marmol 90-inning year is hugely effective.

 

Let's make this a fair comparison (to help out the Marmol as a starter argument, as Marquis to the bullpen hurts the argument). Marmol at his current production with a league average starter, and a league average reliever + Marmol as a starter.

 

Marmol for 90 innings (1.64 ERA) + league average starter (4.47 ERA) for 200 innings=3.59 ERA

league average reliever for 90 innings (3.77 ERA)+Marmol as a starter= ?

 

Marmol would have to have a 3.51 ERA for a team to get the same production out of his as a starter as a reliever.

 

Most of the other factors then tend to balance out. Your top reliever throws more close innings than an average starter. At the same time, starters are harder to find, and more expensive to pay for.

 

It's too big of a gamble IMO. Marmol would have to pitch like a borderline ace just to get it to even out. There's not even close to a guarantee he will do that. He could be that, but he could be a lot worse. The chance of him being better than that and being one of the best 10 starters in the league is small compared to the chance that he would just be a league average starter.

[bukie]

Win probability is an incredibly important concept to bullpen usage. If we ignore it, then we shouldn't have any bullpen roles at all. Just pull a name out of a hat every time the pen is needed.

 

Bullpen roles are extremely overrated. Getting the most innings out of the best pitchers when run prevention is needed is far more important than putting your best relievers in at designated spots in the game. Just because a game is in the ninth inning, doesn't make those three outs the most significant outs in the game. There are many other factors to consider, the most important of those is using your best pitchers in situations to prevent the team from scoring runs most often. Sure, in a 7-0 game, there's less need to prevent runs from scoring, but in a close game, whether it's the 6th inning or the 9th inning, preventing runs from scoring means just as much.

 

Stats like VORP and Win Shares are cumulative, yes. They give less credit to relievers, specifically because relievers participate less than starters. You may not think that's fair, because relievers are just as valuable in their innings as starters are in theirs...and while that may be true, simply pitching 1/3 as much time as a starter is really what is limiting their value, there.

 

SNLVAR and WXRL are two stats calculated separately. They can't really be used interchangeably, as one compares starters to starters to find a replacement level, and the other compares relievers to relievers. Unless you work under the assumption that the average reliever == the average starter (or, even more of a stretch, a replacement-level reliever == a replacement-level starter), it's not a fair comparison. I think it's a more safe assumption that the average starter > the average reliever.

[bukie]

Your top reliever throws more close innings than an average starter.

 

I hope I've done enough already in this thread to show that this is a false assumption.

[CubColtPacer]

Your top reliever throws more close innings than an average starter.

 

I hope I've done enough already in this thread to show that this is a false assumption.

 

You have not. The two things I have an issue with are:

 

You have made no distinction between leads and deficits. A reliever coming in constantly with a 1-2 run deficit is not nearly as valuable as a reliever coming in constantly with a 1-2 runs lead. Giving up 1-2 runs when you're already behind simply lowers your already low chance of winning. Giving up 1-2 runs when you're barely ahead decreases your chances of winning a lot more than that.

 

There is no distinction made that top relievers come in sometimes (and if the manager is smart, a good percentage of the time) with runners already on, which makes the game closer than the score would necessarily indicate.

[bukie]

Your top reliever throws more close innings than an average starter.

 

I hope I've done enough already in this thread to show that this is a false assumption.

 

You have not. The two things I have an issue with are:

 

You have made no distinction between leads and deficits. A reliever coming in constantly with a 1-2 run deficit is not nearly as valuable as a reliever coming in constantly with a 1-2 runs lead. Giving up 1-2 runs when you're already behind simply lowers your already low chance of winning. Giving up 1-2 runs when you're barely ahead decreases your chances of winning a lot more than that.

 

There is no distinction made that top relievers come in sometimes (and if the manager is smart, a good percentage of the time) with runners already on, which makes the game closer than the score would necessarily indicate.

 

Considering a starter never comes in with a 1-2 run deficit, this only further indicates the fallacy that relievers come in with more close situations. A starter is always entering a tie game.

 

The only difference between a 1-0 game in the third and a 1-0 game in the ninth is that the team has fewer opportunities in front of it to score more runs to win the game. Plus, there exist possibilities earlier in the game that haven't come to fruition by late in the game. However, the pitcher himself has very little control over those possibilities. The pitcher can only be primarily concerned with what he does have direct control over, and that is limiting walks, HRs, and LDs and maximizing Ks and pitch efficiency. That job is the same in the first as it is in the ninth.

[CubColtPacer]

Your top reliever throws more close innings than an average starter.

 

I hope I've done enough already in this thread to show that this is a false assumption.

 

You have not. The two things I have an issue with are:

 

You have made no distinction between leads and deficits. A reliever coming in constantly with a 1-2 run deficit is not nearly as valuable as a reliever coming in constantly with a 1-2 runs lead. Giving up 1-2 runs when you're already behind simply lowers your already low chance of winning. Giving up 1-2 runs when you're barely ahead decreases your chances of winning a lot more than that.

 

There is no distinction made that top relievers come in sometimes (and if the manager is smart, a good percentage of the time) with runners already on, which makes the game closer than the score would necessarily indicate.

 

Considering a starter never comes in with a 1-2 run deficit, this only further indicates the fallacy that relievers come in with more close situations. A starter is always entering a tie game.

 

The only difference between a 1-0 game in the third and a 1-0 game in the ninth is that the team has fewer opportunities in front of it to score more runs to win the game. Plus, there exist possibilities earlier in the game that haven't come to fruition by late in the game. However, the pitcher himself has very little control over those possibilities. The pitcher can only be primarily concerned with what he does have direct control over, and that is limiting walks, HRs, and LDs and maximizing Ks and pitch efficiency. That job is the same in the first as it is in the ninth.

 

A starter is always entering a tie game, but frequently an average starter will be behind by 1-2 runs but still pitching. Those are situations that a top reliever is not frequently pitching in, and that is why his on a per inning average, a top reliever will have more innings where an effective pitcher is essential than a starter will. Effectiveness is much more needed when a team is ahead than when a team is behind.

[bukie]

A starter is always entering a tie game, but frequently an average starter will be behind by 1-2 runs but still pitching. Those are situations that a top reliever is not frequently pitching in, and that is why his on a per inning average, a top reliever will have more innings where an effective pitcher is essential than a starter will. Effectiveness is much more needed when a team is ahead than when a team is behind.

 

Effectiveness is much more needed to get a team ahead in the first place. Sure, when a team has a 1-2 run lead late, it's nice to have effective pitching to finish out the game. However, effective pitching to start the game will lead to many more 1-2 run leads late than 1-2 run deficits late.

 

I'd much rather have an 90% effective starter and a 60% effective reliever than the other way around. Sure, with an 90% effective reliever, you'd feel confident once you have the lead late, but with a 90% effective starter, you'd feel confident that you would have the lead late to begin with.

[CubColtPacer]

A starter is always entering a tie game, but frequently an average starter will be behind by 1-2 runs but still pitching. Those are situations that a top reliever is not frequently pitching in, and that is why his on a per inning average, a top reliever will have more innings where an effective pitcher is essential than a starter will. Effectiveness is much more needed when a team is ahead than when a team is behind.

 

Effectiveness is much more needed to get a team ahead in the first place. Sure, when a team has a 1-2 run lead late, it's nice to have effective pitching to finish out the game. However, effective pitching to start the game will lead to many more 1-2 run leads late than 1-2 run deficits late.

 

I'd much rather have an 90% effective starter and a 60% effective reliever than the other way around. Sure, with an 90% effective reliever, you'd feel confident once you have the lead late, but with a 90% effective starter, you'd feel confident that you would have the lead late to begin with.

 

I think we're talking in circles at this point, so let me reiterate one of my earlier points that wasn't addressed. The likelihood that Marmol can put up a 3.50 ERA as a starter is certainly less than 50%. Even if you think it is above 50, the chance that he is at least .5 below that is smaller then the chance he has to put at least .5 above that. Marmol+an average reliever would put up worse numbers overall than an average starter+Marmol most of the time. It just isn't worth it unless there was 90-100% assurance that Marmol would be an ace, and there is not anywhere close to such assurance with Marmol.

[bukie]

I think we're talking in circles at this point, so let me reiterate one of my earlier points that wasn't addressed. The likelihood that Marmol can put up a 3.50 ERA as a starter is certainly less than 50%. Even if you think it is above 50, the chance that he is at least .5 below that is smaller then the chance he has to put at least .5 above that. Marmol+an average reliever would put up worse numbers overall than an average starter+Marmol most of the time. It just isn't worth it unless there was 90-100% assurance that Marmol would be an ace, and there is not anywhere close to such assurance with Marmol.

 

Given Marmol's numbers as a starter over parts of 5 years in the minors, I'd say the likelihood he could put up a 3.50 ERA (which, honestly, is a high bar for even an above average starter) is decent, possibly even 50-50. His biggest deterrent right now from being an effective starter is his pitch efficiency. Really, if the guy's going to average 20+ pitches an inning, it'll be difficult to stretch him out into an effective starter.

[CubColtPacer]

I think we're talking in circles at this point, so let me reiterate one of my earlier points that wasn't addressed. The likelihood that Marmol can put up a 3.50 ERA as a starter is certainly less than 50%. Even if you think it is above 50, the chance that he is at least .5 below that is smaller then the chance he has to put at least .5 above that. Marmol+an average reliever would put up worse numbers overall than an average starter+Marmol most of the time. It just isn't worth it unless there was 90-100% assurance that Marmol would be an ace, and there is not anywhere close to such assurance with Marmol.

 

Given Marmol's numbers as a starter over parts of 5 years in the minors, I'd say the likelihood he could put up a 3.50 ERA (which, honestly, is a high bar for even an above average starter) is decent, possibly even 50-50. His biggest deterrent right now from being an effective starter is his pitch efficiency. Really, if the guy's going to average 20+ pitches an inning, it'll be difficult to stretch him out into an effective starter.

 

Oh, I realize it's a high bar. But it's the bar Marmol has set for himself with how unbelievable he's been as a reliever.

 

And I would agree that pitch efficiency would be a big problem. Marmol is averaging over 4 pitches per plate appearance for his major league career. He's averaging 4.3 pitches this year. You can still be effective with a number around 4, but it doesn't leave you much margin for error. Only 3 pitchers in the majors managed to hit 200 innings with the number at or above 4 (Kazmir, Cain, Peavy). They typically manage to do it because they never come out early. The fewest pitches Peavy threw all season last year was 95. For Kazmir, it was 91. Cain's had a few short outings, but also several very long outings.

[Hairyducked Idiot]

 

Bullpen roles are extremely overrated. Getting the most innings out of the best pitchers when run prevention is needed is far more important than putting your best relievers in at designated spots in the game. Just because a game is in the ninth inning, doesn't make those three outs the most significant outs in the game. There are many other factors to consider, the most important of those is using your best pitchers in situations to prevent the team from scoring runs most often. Sure, in a 7-0 game, there's less need to prevent runs from scoring, but in a close game, whether it's the 6th inning or the 9th inning, preventing runs from scoring means just as much.

 

 

That's exactly what I meant. If we use Win Shares and VORP, we ignore that a good reliever's value comes partly from the fact that he can be used in situations where run prevention is more critical than in normal situations.

 

Stats like VORP and Win Shares are cumulative, yes. They give less credit to relievers, specifically because relievers participate less than starters. You may not think that's fair, because relievers are just as valuable in their innings as starters are in theirs...and while that may be true, simply pitching 1/3 as much time as a starter is really what is limiting their value, there.

 

It's not that they are cumulative, it's that they ignore context. It's fine to ignore context for a starter or a position player, because their playing time isn't determined by context, the context is more or less random. A relief pitcher's appearances are completely determined by context, so any stat that ignores it when calculating the value of the reliever is asinine.

 

SNLVAR and WXRL are two stats calculated separately. They can't really be used interchangeably, as one compares starters to starters to find a replacement level, and the other compares relievers to relievers. Unless you work under the assumption that the average reliever == the average starter (or, even more of a stretch, a replacement-level reliever == a replacement-level starter), it's not a fair comparison. I think it's a more safe assumption that the average starter > the average reliever.

 

That's completely irrelevant. We're not trying to come up with some sliding scale where we determine who is better relative to others at their position, we are determining absolute value to the team.

The statement that "average starter > average reliever" is completely irrelevant unless you state what scale you are measuring by. Talent? ERA? Wins added to team performance? The average starter is worth more than the average reliever in terms of wins and talent, less in terms of absolute run prevention per inning. But none of that is relevant because starters are on a much narrower band than relievers. While the average starter may be worth more than the average reliever, the best relievers can be of comparable value to the best starters, unless you believe that all innings and situations have equal likelihoods to decide wins and losses, which is absurd.

[Bruce Miles]

I stumbled on this yesterday, and it's a few days old, but I thought it would be a good discussion.

 

Link

 

I think I agree with Bruce here. It's not that I don't have confidence in Wood as a closer, but the way that everyone in the bullpen has to have a pre-ordained assigned role irks me as well.

 

Sometimes it might be prudent when Marmol is throwing well to let him continue on and get the final outs. The same might be said of any pitcher in the pen.

 

And maybe sometimes it would be best to bring Wood into the eighth inning of a game.

 

Well, don't just "stumble" on to our new baseball blog. Make it a point to go to it! We have a new blog at the Daily Herald, covering both the Cubs and White Sox. I try to post almost every day. I'll throw out all kinds of topics, some and some not. Today, I did a mini book review. I talked hockey with Dempster last week. We'll get into some stats stuff, too. Check 'em out. Leave comments. Go back often. See you there and right here at NSBB.

[Ding Dong Johnson]

Well, don't just "stumble" on to our new baseball blog. Make it a point to go to it! We have a new blog at the Daily Herald, covering both the Cubs and White Sox. I try to post almost every day. I'll throw out all kinds of topics, some and some not. Today, I did a mini book review. I talked hockey with Dempster last week. We'll get into some stats stuff, too. Check 'em out. Leave comments. Go back often. See you there and right here at NSBB.

we have to stumble if you don't pimp the site here first. I'll add it to my list of daily things to do, right after checking woot.com and FJM

[Bruce Miles]

Well, don't just "stumble" on to our new baseball blog. Make it a point to go to it! We have a new blog at the Daily Herald, covering both the Cubs and White Sox. I try to post almost every day. I'll throw out all kinds of topics, some and some not. Today, I did a mini book review. I talked hockey with Dempster last week. We'll get into some stats stuff, too. Check 'em out. Leave comments. Go back often. See you there and right here at NSBB.

we have to stumble if you don't pimp the site here first. I'll add it to my list of daily things to do, right after checking woot.com and FJM

 

I appreciate being third on that esteemed list. I see FJM has had some fun with the Trib lately.

[Ding Dong Johnson]

Well, don't just "stumble" on to our new baseball blog. Make it a point to go to it! We have a new blog at the Daily Herald, covering both the Cubs and White Sox. I try to post almost every day. I'll throw out all kinds of topics, some and some not. Today, I did a mini book review. I talked hockey with Dempster last week. We'll get into some stats stuff, too. Check 'em out. Leave comments. Go back often. See you there and right here at NSBB.

we have to stumble if you don't pimp the site here first. I'll add it to my list of daily things to do, right after checking woot.com and FJM

 

I appreciate being third on that esteemed list. I see FJM has had some fun with the Trib lately.

you'll have to mention my name to get moved up

[jersey cubs fan]

Well, don't just "stumble" on to our new baseball blog. Make it a point to go to it! We have a new blog at the Daily Herald, covering both the Cubs and White Sox. I try to post almost every day. I'll throw out all kinds of topics, some and some not. Today, I did a mini book review. I talked hockey with Dempster last week. We'll get into some stats stuff, too. Check 'em out. Leave comments. Go back often. See you there and right here at NSBB.

we have to stumble if you don't pimp the site here first. I'll add it to my list of daily things to do, right after checking woot.com and FJM

 

I appreciate being third on that esteemed list. I see FJM has had some fun with the Trib lately.

 

I check the Daily Herald sports section first, but I don't remember about the blog until after I've read the SI Extra Mustard and check my bank account. You need more pictures of women to draw traffic.

[vance_the_cubs_fan]

I stumbled on this yesterday, and it's a few days old, but I thought it would be a good discussion.

 

Link

 

I think I agree with Bruce here. It's not that I don't have confidence in Wood as a closer, but the way that everyone in the bullpen has to have a pre-ordained assigned role irks me as well.

 

Sometimes it might be prudent when Marmol is throwing well to let him continue on and get the final outs. The same might be said of any pitcher in the pen.

 

And maybe sometimes it would be best to bring Wood into the eighth inning of a game.

 

Well, don't just "stumble" on to our new baseball blog. Make it a point to go to it! We have a new blog at the Daily Herald, covering both the Cubs and White Sox. I try to post almost every day. I'll throw out all kinds of topics, some and some not. Today, I did a mini book review. I talked hockey with Dempster last week. We'll get into some stats stuff, too. Check 'em out. Leave comments. Go back often. See you there and right here at NSBB.

 

I'll bookmark it for now on. Looking forward to more good stuff. :good:

[Tyrant]

 

Oh, I realize it's a high bar. But it's the bar Marmol has set for himself with how unbelievable he's been as a reliever.

 

I'm not sure about Marmol as a starter because of his lack of efficiency. He'd also need a third pitch to use occasionally.

 

Maybe he has a changeup or something, but just doesn't bother using it in relief.