I'd rather have Joba or Hughes as a fifth starter than two inning relievers. For this team there's more opportunity to convert losses to wins from the 5th starter's starts than from the 7th or 8th inning onward. The 2009 Yankees lost maybe 5 games in which they were tied or ahead after 7. How many starts given to SPs #5+ did they fall behind and lose?

If you really want to rest A-Rod and Jeter, isn't a full day of rest (before or after an off-day) more effective than a couple of late innings? And, if that's so, wouldn't the Yankees' chances to win while resting them be better with the best SP possible?

(Joba was great as a 1 inning reliever and not as good as a SP. With a 40 year old Mariano, would this plan really groom Joba or Hughes in the 8th inning role as a possible successor?)

A ten man staff should be plenty whether or not Joba/Hughes are in the bullpen. If your starters are able to go six innings regularly there is no reason to carry 11 or 12 freakin' pitchers. Worst case you play an extra inning game, an early blowout or a doubleheader due to weather and when that happens you just ship Ramiro Pena or an equivalent player to AAA and bring up an 11th pitcher for a week or two.

You'd probably have to dispense with a LOOGY in this scenario but the world doesn't end without a LOOGY.

One of Hughes or Joba definitely need to be in the rotation.

The issue is, you want the other stretched out as much as possible to prepare for the inevitable injuries. You want them as your 6th SP, not Gaudin or Mitre.

Whoever is in the pen needs to be a long reliever. Let them pitch 2-3 inning stints. Robertson/Marte can handle the 8th.

Joba/Hughes should be stretched out as much as possible, even if it means not bringing Mariano in to some save situations. i.e. Joba/Hughes take over in the 7th or 8th, are pitching well after 8, and the Yankees are up by 2 or 3. Let them try and finish to get the 2nd/3rd inning under their belt.

A ten man staff should be plenty whether or not Joba/Hughes are in the bullpen. If your starters are able to go six innings regularly there is no reason to carry 11 or 12 freakin' pitchers. Worst case you play an extra inning game, an early blowout or a doubleheader due to weather and when that happens you just ship Ramiro Pena or an equivalent player to AAA and bring up an 11th pitcher for a week or two.

This only works if all RPs pitch multiple inning stints regularly. The limit is not the IP, it's that guys can't pitch 3 days in a row. You need to go back to 1970's pen usage.

This only works if all RPs pitch multiple inning stints regularly

I don't think so. You still have five relievers and presumably one of them is your closer pitching the 9th. That leaves you with two guys pitching the 7th and 8th each day. With blowouts (either way) you shouldn't expect to have to use the same pitcher more than twice in a row too often. Like I said when the schedule backs up on you for whatever reason you can make a roster move for a short time. I think the issue isn't so much having guys able to pitch multiple innings as it is having guys you can rely on throughout the bullpen. If your 4th/5th relievers are not good pitchers then you are a bit stuck.

I don't think so. You still have five relievers and presumably one of them is your closer pitching the 9th. That leaves you with two guys pitching the 7th and 8th each day. With blowouts (either way) you shouldn't expect to have to use the same pitcher more than twice in a row too often. Like I said when the schedule backs up on you for whatever reason you can make a roster move for a short time. I think the issue isn't so much having guys able to pitch multiple innings as it is having guys you can rely on throughout the bullpen. If your 4th/5th relievers are not good pitchers then you are a bit stuck.

You can't just gloss over blowouts. If you only have 5 ~one-inning RPs, a blowout means everyone pitches. Then what if the starters only go 5 and 6 IP the next two days?

To get by with 10 pitchers, you need one true long/swing-man who can go 4-5 IP when the SP is knocked out early, you need 2 other guys who can give you 3 IP on occasion, and 2 IP regularly, and 2 short guys who go 1-2 IP.

If you look at the successful bullpens from the 1970's (the 1977-80 Yankees are a great example) your RP's average ~2 IP per appearance. Plus, there are lots of RPs who are spot starters.

The Yankees could make this kind of pen work this year, although 6 RPs is probably the realistic minimum right now. Assume Joba is the 5th SP. Mariano/Marte/Robertson are the short-men, but are all capable of giving you 2 IP. Aceves, Hughes and Gaudin are all SPs, and can regularly go 3+ IP.

I'd love to see them take stress off the SP by pulling them when they're stuggling early, for one of the long men, which should then let them pitch deeper into games when they're going good w/o increasing their total workload.

Question:

If a team does go down to a 10-man staff, how many wins are they losing in terms of not being able to keep a better pitcher "in reserve" on the staff as an 11th man versus some AAAA replacement pitcher they'll have to call up in case of an emergency? It's not like MLB-decent relievers are content to go sit at Scranton and twiddle their thumbs.

I realize out on the fringes of your roster there's not a lot of traction either way in the win column, but still.

It's not like MLB-decent relievers are content to go sit at Scranton and twiddle their thumbs.

If the reliever in question is under contract and has options, he doesn't have much say in the matter.

If a team does go down to a 10-man staff, how many wins are they losing in terms of not being able to keep a better pitcher "in reserve" on the staff as an 11th man versus some AAAA replacement pitcher they'll have to call up in case of an emergency? It's not like MLB-decent relievers are content to go sit at Scranton and twiddle their thumbs.

Depends on whether you are generating your bullpen arms internally or signing FAs.

In general, I think the 5th-8th guys in the pen are pretty fungible, and pitch low leverage innings. As long as you can generate young decent arms from your minors, there should be no real drop-off. Even if there is one, I think we're talking tenths of wins here, given even a good middle reliever is usually not worth more than 1.0 WAR.

It all depends whether you think most relievers can pitch far more innings than they currently do without losing much effectiveness. A 10 man pitching staff, even with a workhorse rotation, means each reliever is throwing 90+ innings for the season. If you limit Rivera to ~70 IP, you're up to ~95 IP for each reliever. Over the last two years, there's not a single reliever in baseball who's thrown 90 innings in relief. I'm sure some guys could handle it, but you're probably going to ruin some arms finding out which one's can.

All this to carry a 3rd catcher and extra utility infielders?

York can then afford to carry a third catcher and a two slick fielding infielders to rest A-Rod and Jeter late in games.

This is where his argument falls apart. Why would you need a defensive replacement for the winner of 4 of the past 6 Gold Gloves at SS?

It all depends whether you think most relievers can pitch far more innings than they currently do without losing much effectiveness. A 10 man pitching staff, even with a workhorse rotation, means each reliever is throwing 90+ innings for the season. If you limit Rivera to ~70 IP, you're up to ~95 IP for each reliever. Over the last two years, there's not a single reliever in baseball who's thrown 90 innings in relief. I'm sure some guys could handle it, but you're probably going to ruin some arms finding out which one's can.

All this to carry a 3rd catcher and extra utility infielders?

I don't think that's the result.

First of all, no one's going to 10 right away, 11 is the short term goal. Second, I think you still use as many pitchers, they're just rotating more between AAA and MLB.

Also, if you have 2 long-men who are basically SP's, and use them in 3+ inning stints, then can give you ~120 IP each, when you combine their long-relief work, with spot starts filling in for injury.

So, if your rotation gives you 160 IP each, and your 6th-7th SPs/long-relievers give you 120 each, your 4 short-men go 70 each, that leaves 100+ IP to get from other RPs you cycle through during the year.

That seems doable.

Also, the big gains wouldn't be in 3rd C's or UIF, it would be the ability to run real platoons again and cobble together good performance out of cheap flawed players.

How many starts given to SPs #5+ did they fall behind and lose?

The Yankees had 32 starts by Aceves, Gaudin, Hughes, Mitre and Wang. They went 18-14 in those games, and got 147 innings out of the starters. Joba Chamberlain also made 32 starts. The Yankees went 21-11 in those games and got 156.1 innings out of him in that role. Replacing Wang's nine starts with a replacement level starter would probably pretty much even that out.

11 is the short term goal

Yeah, I don't see the huge problem with a bullpen of three guys who can each throw one inning three days in a row and three guys who can each throw 2-3 innings every 2-3 days. It should be perfectly workable as long as your LOOGY(s) are capable of getting a RHB out once in a while.

If your starters are able to go six innings regularly there is no reason to carry 11 or 12 freakin' pitchers.

Sorry, no.

That is the scenario these days. The typical team needs about 480-500 innings of relief. With a 5-slot pen, that's an average of nearly 100 IP per reliever slot. No team has ever come close to doing that and none ever will. Not a single reliever has topped 100 IP since Proctor in 2006 (which did wonders for him).*

It's frankly a minor miracle they're able to do it with 7 reliever slots. That's 70 innings per slot. There were only 40 relievers who even hit the 70 IP mark last year, another 49 hit the 60 IP mark. So a team's top 3 relieves pitch about 210 and a collection of flotsam and jetsam rotating between AAA, MLB, the DL and the waiver wire manage to fill up the rest.

* and if you look back at the history of 100+ IP relievers, it's not a pretty picture.

The only way to reduce the size of the pitching staff is to increase IP/start. Nobody has had healthier and more durable starters the last few seasons than the White Sox and even they have needed 400-470 relief IP. St. Louis also usually does a good job of limiting relief IP. Since the Sox in 2006, no team has used fewer than about 425 relief IP. Somewhere around 400 seems the absolute minimum (Sox 2006, Sox and Cards 2005). You need to get IP/start up to around 7 to cut back to a 10-man staff. That might be possible but it would be a radical change and it's got nothing to do with putting Joba and Hughes in the pen (where, apparently, they're both pitching about 120 innings -- yikes!).

It's frankly a minor miracle they're able to do it with 7 reliever slots.

So a team's top 3 relieves pitch about 210 and a collection of flotsam and jetsam rotating between AAA, MLB, the DL and the waiver wire manage to fill up the rest.

So why can't you rotate the flotsam and jetsam among 6 spots rather than 7?

If you put starter in the bullpen and use him for 2-3 innings and not on consecutive days, he can pitch over 100 innings. Problem with reliever usage is warm-up and back-to-back appearances, not just innings total. If you increase IP/game, innings total can increase without increasing injuries (of course, for pitchers who are properly stretched). If you have reliever who averages 2.5 IP/game, he needs to appear in less than 50 games to reach 120 IP. Make sure he is former starter and he gets 2 days off after each appearance and it should be safer than 70 IP in 80 games.

If you have reliever who averages 2.5 IP/game, he needs to appear in less than 50 games to reach 120 IP. Make sure he is former starter and he gets 2 days off after each appearance and it should be safer than 70 IP in 80 games.
it's unclear whether that is actually a safer way to go. iirc, the health record of relievers with >100ip or >2.5ip/g is poor.

It all depends whether you think most relievers can pitch far more innings than they currently do without losing much effectiveness. A 10 man pitching staff, even with a workhorse rotation, means each reliever is throwing 90+ innings for the season. If you limit Rivera to ~70 IP, you're up to ~95 IP for each reliever. Over the last two years, there's not a single reliever in baseball who's thrown 90 innings in relief.

...and...

That is the scenario these days. The typical team needs about 480-500 innings of relief. With a 5-slot pen, that's an average of nearly 100 IP per reliever slot. No team has ever come close to doing that and none ever will. Not a single reliever has topped 100 IP since Proctor in 2006 (which did wonders for him).*

Why on earth are you guys treating "reliever slots" as individual relievers? The fact that you only have 5 people in your bullpen at any one time does not mean that you'll have those same five people all year. (And hence does not mean that you'll have lots of relievers throwing 90 or 100 innings.)

Plus, we're only talking about 5 months; once rosters expand in September you're going to have more than a 10-man staff.