THE BOOK--Playing The Percentages In Baseball

Aside from making the hitter see a different pitcher, platoon splits could be leveraged as well.

That said, if the series went 7 games, would the relief corps be wore out from all of the warming up?

I would imagine the great performance vs. relievers mainly reflects the Yankees’ 103-59 record.  That is, the relievers they faced were mainly the mop-up guys put in when a team is losing, much more than the set-up men and closers employed by a team with the lead.

Great thinking by Smoltz.....

I posted this on primer too, but I would treat Cliff Lee like a normal starter, and then pair up Blanton/Happ and Pedro/Hamels and have them try to go 3 innings each.

Teams that are on the brink of elimination seem to do this on a regular basis.  For example, a team down 3-2 doesn’t seem to have a problem taking out their starter (say he is their third best starter) in the 4th inning when he’s down 3-1 and in trouble....whereas they’d almost never take him out during the regular season, and probably wouldn’t do it if it was game 1 or 2 in a series.

On the flip side, shouldn’t the Yankees be thinking about this when it comes to their 4th starter (whoever that will be)?

This is truly the greatest time in history to be a baseball fan.

Except of course, I’m a Pirate fan in Pittsburgh. But still...look at the information at the tips of your fingers!

Tango, what is the equivalent 1,2,3,4+ through the order in ERA?  IOW, how many points in ERA (rather than wOBA or OPS) are lost each time through the order?

It is easier to see why relievers should be brought in if you look at it in terms of ERA. IOW, lets say we have a starter with a true talent ERA of 4.00 (a pretty good starter).  What is his ERA in the 7th inning when he starts to face batter for the 4th time?  4.50?  What about a run of the mill reliever coming in for 1 inning?  3.50?  4.00?

When you look at it that way, you can see that it is not even close.  Even a run of the mill reliever is better in ERA than a very good starter the 4th time (or even the 3rd time) through the order.  Even managers should be aware of this data, shouldn’t they?

I would like to see a chart of this:

First column is the average true talent ERA of a class of starters.  Second column is their ERA the 3rd time through the order.  The 3rd column is their ERA the 4th time through the order.  The 4th column is the ERA of an average reliever for one inning.  The last column is the ERA of an above average reliever for one inning.

There should be 4 rows.  The first row is for the class of typical “4th starters” The next row is for typical 3rd starters.  Next row, number 2 starters.  Last row, elite starters.

Can anyone provide that?

8 points of wOBA, with 9 PA, and presumably 6 outs = .008/1.15*9/6*27 = 0.28 runs per 9 IP, which is 0.26 in ERA.

So, I’d say it would be something like:
3.75, 4.00, 4.25 (for an average 4.00 starter)
4.75, 5.00, 5.25 (for an average 5.00 starter)
3.75, (for an average 5.00 starter IN RELIEF)

So, a starter who is 5.50 on average, would be 5.25 the first time through the order as a starter, and 4.25 as a reliever.

So, a 4.00 starter and a 5.50 starter are equivalent, if the 4.00 starter is facing the lineup the third time through the order, and the 5.50 starter ACTS AS A RELIEVER and faced the order the first time (and only expects to face them one time).

The Phillies do pretty well as they face a pitcher more often, too, here’s their OPS (Yankees in parentheses); keep in mind the Phillies’ stats include their pitchers as hitters:

1st 0.750 (0.785)
2nd 0.782 (0.814)
3rd 0.859 (0.859)
4th 1.024 (1.128)
relievers A not-so-hot 0.757 (A hotter 0.863)

@Donald: Yea but how much does that factor in the Phillies getting to face whatever terrible reliever the Mets or Nats brought in in the 4th after their starter got knocked out?

I’d be careful looking at team stats on the “4th time through.” It’s a self-selected population of good hitters. Better to stick with it on a batter by batter basis and not aggregate it.

I’d also guess it’s a self-selected population of good pitchers (or pitchers having a good game), otherwise they probably wouldn’t make it through 27 batters.

A further guess would be that the batters who are allowed to hit a 4th time are actually worse batters than average, because I would guess that many managers would be far more willing to let Gritty Pitchalot come into an inning after 110 pitches a lot more readily if he’s facing 7-8-9 in the order than 3-4-5.

All pure speculation/guesses. Obviously it could be checked (by someone much smarter than me).

You could do matched pairs of each batter/pitcher combination, coded with how many times thru the order for each PA.