Closers

[David Watts]

Recently I started using the settings of, stamina-low, use of relievers-normal and use of closers-rarely. I use the DH in my league and these settings provide for a 10 man pitching staff, 5 starters and 5 relievers. I am loving this setup. It's kind of refreshing to manage a bullpen without a closer. I now have my best relief guy available anytime I need him. I actually think I'm a better manager using this strategy.

This has me thinking about whether or not a real team might benefit from trying this. What do y'all think? And what about the save as a stat? I know in the past Gossage has been very critical of the fact that so many saves are nothing more than a guy coming in to get 3 outs with a 2 or 3 run lead. Basically, something most guys should be able to do in their sleep.

Oh and thanks to Questdog for his ridicule of the 11 man bullpen. You sir are my muse/muze, well one of those at least.

[D-BacksJosh]

1 game 2 pitchers. SP 7 hard innings, reliever comes in for the rest, no closer. Closer is the GLAM position of the bullpen. More of a show then anything else.

[JohnHoward]

If you like historical play, what you have described is basically the way bullpens were managed from the mid-50's through the early 70's. Except for a few guys who profile more like closers (Perranoski, Abernathy, Wilhelm in some years), the norm was 4-5 guys who worked in tandem, with the manager often riding one or two guys when they were hot, sometimes right into the rotation (Niekro for the Braves in 1967). It really is fun to manage the bullpen in the style of these years. I have always thought that it isn't until you get to the early-mid 70's that you start to really see widespread exclusive closers. (Fingers still got a couple of starts a year as late as 1973.) Everything changed after that and even though Gossage is right, it was his performance more than any other pitcher that solidified the role of closer, even thought I always thought Ron Davis was a better pitcher on those teams (79-81 Yankees).

[GMLoophole]

I think the Giants last year stumbled onto something worth continuing. The defectiveness of the "save" stat (come in with the bases empty and give up two runs of a three run lead!) leads to some really, really overpaid ninth inning pitchers. Relief pitchers also don't get into too many games, so by nature decisions about them are based on small sample sizes and you're likely to see up and down performances from year to year.

And why save your best guy to pitch to the 7-8-9 batters if that's who's up in the ninth? If you have a small lead in the 7th and a jam to be faced, that's when the best guy should get the call.

I've ended up playing OOTP this way, more or less, and the effects on your team's budget are pretty dramatic. There's a reason Billy Beane suggested the "sell the closer" gambit in Moneyball. Unless you're talking about Mariano Rivera or Goose Gossage or someone, picking a closer is just glorifying a middle reliever.

[rpriske]

Yes. The creation of the 'Save' has really hurt managers' ability to handle the bullpen.

[Questdog]

They used to be called Firemen, because they would come in to put out the fires of a rally......now they are used more as Fire Prevention......which is boring.....nobody wants to go see a movie about a fire prevention specialist.....

[The Game]

In one fictional historical league set in 1976. I used 6 MR 3L/3R all 6 pitched in 100 games and all 6 had at least 10 saves. My SP was atrocious. No starter won more then 9 games.

[David Watts]

Well so much for me being a better manager without a closer. I still suck. Won 80 games our first season, but followed that up with a last place finish. Oh well at least we got the 4 pick in the draft.

Crazy thing is, I had a starter lose 23 games (7-23). Strange thing is, his numbers really weren't those of a 20+ game loser. The pitcher on my team that won 18 games, may have actually had worse numbers. I also had my best reliever lose 16 games(1-16). I think he only had 4-5 blown saves, so it wasn't like he was coming in and blowing it every time. He pitched in a ton of games that we were tied up late and always seemed to be the one on the hook for the loss.

[GMLoophole]

Originally Posted by David Watts

Crazy thing is, I had a starter lose 23 games (7-23). Strange thing is, his numbers really weren't those of a 20+ game loser. The pitcher on my team that won 18 games, may have actually had worse numbers. I also had my best reliever lose 16 games(1-16). I think he only had 4-5 blown saves, so it wasn't like he was coming in and blowing it every time. He pitched in a ton of games that we were tied up late and always seemed to be the one on the hook for the loss.

I'm no big sabermetrics nerd, but I can tell you that the Win as a stat is probably even less helpful than the save. Reliever wins can be especially deceiving, in that they can pitch poorly and get lucky if their team happens to score more runs while they are still pitcher of record. W/L is to my mind telling you more about the team the guy pitches on than the pitcher himself.

[pilight]

Unlike most strategic evolution in baseball, the closer did not derive from someone doing something different and being successful. For all the Gossage talk, he only pitched for one WS winner and he had no saves in that series. It's very uncommon for a league saves leader to pitch for a WS championship team (Brian Wilson in 2010 is the only one this century).

[GMLoophole]

Originally Posted by pilight

For all the Gossage talk, he only pitched for one WS winner and he had no saves in that series. It's very uncommon for a league saves leader to pitch for a WS championship team (Brian Wilson in 2010 is the only one this century).

Wilson is a funny example, for several of his years the master of allowing a single, a walk, two runs, and getting the save. Charitably I often thought he was eager to get to a specific matchup in the lineup and willing to give away bases for that. But needless to say living on the edge is a dangerous proposition.

As for Gossage, in most of his peak years he pitched much more than an inning per appearance, usually averaging 2 IP or more. So, not necessarily save stats as opposed to "finishing" stats.

I've always thought that LaRussa's A's were the big innovators with the specialized bullpen. Honeycutt in the 8th and Eckersley in the 9th, with Nelson and Burns switching off to pitch the 7th where needed. Maybe it's just the team of that era I saw more than others, but pretty soon it seemed like every team had some guy who waited for his special song to come on and jog in for the 9th. Now (at least in SF) they get their own little video montage, which was amusing last season when suddenly Casilla was "the closer" and was getting his own video tribute whenever he came on. Playing "Enter Sandman" doth not Mariano Rivera make.

[pilight]

My somewhat botched point was that there isn't any evidence to support the idea that having a dominant closer makes teams more likely to win, or even that using a pitcher specifically to close games makes a team more likely to win.

[ThePretender]

You could make a pretty good case for using your third best reliever as a CL in OOTP. SU 1 is your best, SU 2/MR1 is your second best, and CL your third best.

SU1 and SU2/MR1 if he's a good pitcher even with low stam could probably throw 80-100 innings in high leverage spots which is the best way to take advantage of your pen. If your third best RP (CL) is competent he'll still save 85-90% of the games. Plus the way OOTP uses CL he probably won't pitch more than 50-60 innings, whereas the top SU will be put in significantly more innings/games.

If you're not comfortable with using your third best RP as a CL, then just make your second best guy a CL and your top guy SU #1. Your SU #1 will still pitch in the highest leverage situations and get more innings than the closer.

[fhomess]

While I agree that the win isn't as useful for rating pitchers as some might like to think, I do think the loss is a bit more useful, particularly for relievers. Your offense may not have helped you out, but you still gave up those runs.

[GMLoophole]

Originally Posted by fhomess

While I agree that the win isn't as useful for rating pitchers as some might like to think, I do think the loss is a bit more useful, particularly for relievers. Your offense may not have helped you out, but you still gave up those runs.

Agreed, and I think the same can be said for "blown saves".

[Nunyer]

The 'win/loss' stats go both ways... You won't find many 4.50 ERA guys on dominant teams that manage to get 15 wins... and you won't find many 3.00 ERA guys on awful teams with 15 losses... So while it's a "team" stat to a degree, it's mostly a pitcher stat. It's all about the amount of weight you put behind it... the truth is always in the middle. Sure, there is Felix Hernandez who finished with 12-14 wins and right around .500 overall for some awful Mariners teams... and he was largely dominant. He would have been a 20 game winner for almost anybody else... But generally speaking for the vast majority of the league... it all comes out in the wash... good pitchers get wins, bad pitchers take losses.

I agree that the "save" stat is largely useless and/or vastly overrated. If your best reliever is needed in the 6th inning, then dammit you better use him in the 6th inning.

[ThePretender]

You should never, ever put any weight behind Win/Loss stats. They're useless. Doesn't matter if it's a starter or reliever. It doesn't tell you anything.

To the two guys above who said you like using W/L for relievers - it's irrelevant. If one RP gets 2 outs and allows a base hit, and gets replaced by another reliever, and that reliever fails to strand the base runner, then the first RP gets stuck with the loss even though the second guy screwed up. It doesn't tell us anything about either pitcher.

[Sven Draconian]

When you look at how the bullpen has evolved since the 50s until now I expect that sometime around 2020 the idea of a rotation (as we know it) will be extinct.

Eventually the best pitchers will throw 3-5 innings a game and they'll pitch ever 3 or 4 times a week. This makes sense for a lot of reasons:

- Pitchers are more effective in short bursts.
- Decreased loads under fatigue and increased frequency are better for the arm
- More cost effective.
- Changes in strategy have hurt the amount of innings a starter can throw.
- Allows for a greater flexibility with match-ups.

The bullpen has changed a lot just based on effectiveness before we had the ability to compile the data we can now. The bullpen went from guys who were not good enough to start, to the idea of a relief ace, to the closer, to the closer and the setup man, to the full 7-8-9 bridge to the present in which teams are effectively building bullpens that are 5 deep.

It won't be long before teams realize that giving 160 innings to a "5th" starter (who is really a 7th or 8th starter after injuries) is wasteful. It won't be long before the inefficiencies of using a closer are eliminated. Those are changes that I expect to see within 2-3 years.

Once teams are back to 4 man rotations and free from the "roles" in the bullpen the line will get further blurred between a starter and a reliever, albeit over several years.

[struggles_mightily]

Originally Posted by Nunyer

The 'win/loss' stats go both ways... You won't find many 4.50 ERA guys on dominant teams that manage to get 15 wins... and you won't find many 3.00 ERA guys on awful teams with 15 losses...

Right, I mean, it's rare to find an awesome 20 game loser or a 20 game winner who flat out sucks, but I think the issues are:

i) the existence of both is possible/conceivable, and
ii) it's not like there are no better stats for judging pitcher effectiveness than W/L (to say the least)...

As for closers -- I tend to use a relief ace model in which my best reliever appears whenever we're in a jam that isn't a save situation, and my closer pitches strictly in save situations. Right now I am lacking a relief ace, but I am keeping my closer (Andrew Bailey) in the closer role, because it's kind of an in-house rule for me that if I have a Proven Closer(tm) on the roster, he is used as a trad closer (unless there are mitigating circumstances, like the PC(tm) having sucked in the previous year and thus being willing to take a set-up man role, etc.). Just feels more realistic. For that reason I prefer to use closers who are either in (pre-)arb or old and available for one year, whilst my relief ace I might be willing to give a bit more to in dollars and years (although in practical terms you can find these guys before they hit FA).

Whatever your feelings are about the effect of leverage upon individual pitchers, I don't think OOTP models any such effect does it?

[pilight]

Eventually the best pitchers will throw 3-5 innings a game and they'll pitch ever 3 or 4 times a week. This makes sense for a lot of reasons:

More cost effective

Pitchers would affect a larger number of games, which would drive their cost up rather than down. Three innings three times per week over the course of a season is more innings than most starters throw now. Five innings four times a week would be over 500 innings in a season!

The bullpen has changed a lot just based on effectiveness before we had the ability to compile the data we can now. The bullpen went from guys who were not good enough to start, to the idea of a relief ace, to the closer, to the closer and the setup man, to the full 7-8-9 bridge to the present in which teams are effectively building bullpens that are 5 deep.

The data doesn't really support the change. Teams aren't allowing fewer runs than they did with the previous arrangement. The frequency with which teams blow late leads has not changed. Modern bullpen strategy is a change that has come about without evidence to support its effectiveness.

It won't be long before teams realize that giving 160 innings to a "5th" starter (who is really a 7th or 8th starter after injuries) is wasteful. It won't be long before the inefficiencies of using a closer are eliminated. Those are changes that I expect to see within 2-3 years.

Spreading those innings among many equally mediocre pitchers is just as wasteful.

Once teams are back to 4 man rotations and free from the "roles" in the bullpen the line will get further blurred between a starter and a reliever, albeit over several years.

If anything, I expect the opposite. Teams will realize that having their better pitchers throw more innings and their lesser pitchers throw fewer is more effective than throwing all of their pitchers the same number of innings regardless of their quality.