Lineup Protection Program

[Dontworry]

I submit the following to instigate some decent discussion on an oft-discussed subject that is almost universally accepted as a baseball truism, namely:
Lineup protection.
The populist theory? That Joe Goodhitter would see his production plummet if he didn't have Harry Greathitter protecting him in the lineup. We've come across this theory several times on this board, most recently in regard to David Ortiz and Jason Bay and at other times in regard to Barry Bonds et. al.
On the surface, it seems like logical reasoning. It makes sense that Goodhitter would see better pitches with Greathitter lurking on deck and he would benefit.
But does it hold up to investigation. In a word: No.
I submit that it's hogwash, that for every example of Goodhitter's production falling when he loses his protection from Greathitter, there are an equal number of examples of Goodhitter either maintaining or increasing his production without Greathitter as protection.
And as evidence, I submit these two articles:

Protection: Fact or Fiction?
http://www.baseballthinkfactory.org/files/primate_studies/discussion/wright_2003-06-30_0/

Protection Study
http://www.baseball1.com/bb-data/grabiner/protstudy.html

[Ubiquitous]

Theres actually a lot of studies and discussion on this topic.

But the one thing that always surprises me is that they never ask the pitchers or look at what is thrown at these players. The myth is that protection gives a player more fastballs and more meaty pitches to hit then the nibbling tosses he might expect in another situation. Well that is something that can be studied and top of that one could simply ask the pitchers. "Hey Joe what you throw Ortiz and why?" Now then I'm not a Red Sox follower but I have yet to see any pitcher say I threw him a fastball down the middle because Manny was on deck.

Now then what I will say about protection is that reverse-protection is real and it does exist. Though its impact is small. That is people getting on a head of you does lead to favorable situations but that the difference between the player with the most reverse protection and the player with the worst will probably only be around a 5% impact.

[Dontworry]

Theres actually a lot of studies and discussion on this topic.

But the one thing that always surprises me is that they never ask the pitchers or look at what is thrown at these players. The myth is that protection gives a player more fastballs and more meaty pitches to hit then the nibbling tosses he might expect in another situation. Well that is something that can be studied and top of that one could simply ask the pitchers. "Hey Joe what you throw Ortiz and why?" Now then I'm not a Red Sox follower but I have yet to see any pitcher say I threw him a fastball down the middle because Manny was on deck.

Now then what I will say about protection is that reverse-protection is real and it does exist. Though its impact is small. That is people getting on a head of you does lead to favorable situations but that the difference between the player with the most reverse protection and the player with the worst will probably only be around a 5% impact.

Now if this " lineup protection " myth were so true, how did maris in 1966 with Mantle behind him, put up an OPS+ of 101, and in 1967, after being traded to St. Louis, with no Mantle for protection, put up an OPS+ of 116?

[Ubiquitous]

Because players are not robots, they have variation. Having a higher OPS+ the next year does not prove or debunk anything.

[leecemark]

--Not to mention Maris hit in front of Orlando Cepeda in 67. Cepeda won the MVP.

[Edgartohof]

I haven't done any studies or anything, but I personally believe that there is definite truth to lineup protection helping.

The situation that I thought of involved the Mariner's, specifically Griffey Jr. and Edgar Martinez. Separately great hitters, and they could do well anywhere, but together, they were even better.

What is said is: if Griffey were in an unprotected lineup, as great of a hitter as he is, he would not get as many good pitches to see, as it is less of a problem if he gets walked if Dave Valle is hitting behind him, than if Edgar was. They would be willing to walk the .300+/50+ HR guy in order to face the .230 hitter with no power. But they may be more inclined to pitch to him if there is another great hitter behind him. Because if he gets walked, then there is a higher probability that the next hitter will get him in.

I know that didn't cover everything, but hopefully it helped (if it made any sense at all). Of course I can see the other side to this, I just don't particularly agree with it, but then again, I haven't spent too much time studying it, so there is a chance that someone could change my mind.

[Ubiquitous]

A player and his plate discipline is his own best protection.

[digglahhh]

Also, and this was not significant factor in Maris '66 and '67, but sometimes a rising or falling OPS+ is just a function of the league's OPS doing the same. Several players have had their highest raw OPS and adjusted OPS+ in different seasons.

This is something to be especially wary of when players change leagues when one league is particularly stronger than the other, or when a player joins a team that has the premier pitching staff in the league.

In 2001, the Oakland A's pitching staff had a combined 121 ERA+, Jason Giambi and his 202 OPS+ didn't have to hit against any of those guys. A-Rod's (164 OPS+) Rangers has combined ERA+ of 78.

That is a 43% difference. Giambi gets 17 games to hit against those pitchers, A-Rod gets none. A-Rod has to face the A's staff 17 times, Giambi, none.

So, the 38 OPS+ difference between the two wouldn't disappear no matter what, the difference is really large. But remember, Giambi gets about 70 PAs against the worst staff in the league and none against the best. A-Rod gets 70 against the best and none against the worst.

[LouGehrig]

Now if this " lineup protection " myth were so true, how did maris in 1966 with Mantle behind him, put up an OPS+ of 101, and in 1967, after being traded to St. Louis, with no Mantle for protection, put up an OPS+ of 116?


They weren't Mickey Mantle or Roger Maris in 1966.

[Dontworry]

Because players are not robots, they have variation. Having a higher OPS+ the next year does not prove or debunk anything.

Well, that was my point -- that for every example of increased production due to protection, there is a counter example of where protection didn't help (and in some cases had a negative effect on) production. From that, we can conclude that lineup protection is a myth.

[Ubiquitous]

Well, that was my point -- that for every example of increased production due to protection, there is a counter example of where protection didn't help (and in some cases had a negative effect on) production. From that, we can conclude that lineup protection is a myth.

But again that doesn't mean anything. You are not showing any actual relationship or lack of one. A counter example does not negate an example. You have to first prove that their is a relationship in the first example and then you have to prove the relationship in the counter-example.

As of right now you have not shown the increased production was because of protection, age, league, random fluctuation, or praying reallly really hard. Therefore citing examples and counter examples prove nothing because we don't even know if the examples are actually looking at protection samples.

[Dontworry]

But again that doesn't mean anything. You are not showing any actual relationship or lack of one. A counter example does not negate an example. You have to first prove that their is a relationship in the first example and then you have to prove the relationship in the counter-example.

As of right now you have not shown the increased production was because of protection, age, league, random fluctuation, or praying reallly really hard. Therefore citing examples and counter examples prove nothing because we don't even know if the examples are actually looking at protection samples.

Well obviously you didnt take the time to read both of the articles I posted. They both presented good evidence backing up my arguement.

Read them.

[Ubiquitous]

I've read them before and they don't provide plenty of evidence. For again much the same reason I mentioned above. Neither study is all that detailed and is extremely simplistic. Neither has any real controls to actually gauge performance. If protection is getting good pitches to hit then the end result shouldn't be measured but the kind of pitches being given to a player. There is no controls for the many variations that occur throughout baseball and random events that happen with BIP. These studies are similar in the criticism Bill James uses against clutch hitting in his beyond the fog article. It is possible that there is an effect but it is hidden behind the fog so that we cannot measure it with the tools and and the procedures we are currently using but that does not mean it does not exist. Protection studies right are simplistic they are not the absolute proof that you probably think they are.

[Dontworry]

I've read them before and they don't provide plenty of evidence. For again much the same reason I mentioned above. Neither study is all that detailed and is extremely simplistic. Neither has any real controls to actually gauge performance. If protection is getting good pitches to hit then the end result shouldn't be measured but the kind of pitches being given to a player. There is no controls for the many variations that occur throughout baseball and random events that happen with BIP. These studies are similar in the criticism Bill James uses against clutch hitting in his beyond the fog article. It is possible that there is an effect but it is hidden behind the fog so that we cannot measure it with the tools and and the procedures we are currently using but that does not mean it does not exist. Protection studies right are simplistic they are not the absolute proof that you probably think they are.

I guess I don't like the term "protection". Because I do believe that over the course of the season the opportunities will even out for any hitter in a major league lineup. Thus, I don't believe Derrek Lee would have had 300 RBI had A-Rod and Ichiro batted in front of him last season.

Having said that, I do think a batting order is a strange "cocktail" and one batter almost certainly does effect the other batters- the pitches they see, the pitcher's temperment knowing who is on or who is up next. But each person is different and it's an inexact science.

[Ubiquitous]

Now you are talking about something different. Now you are talking about the reverse protection. Before this you highlighted passages directed toward normal protection.

Derrek Lee would have more RBI's if he had more runners on, that is an inescapable truth.

[Sultan_1895-1948]

The pitcher being mentally grinded down after a batter, causing him to let up or be more apt to make a mistake to the next hitter. That needs to be thrown into "reverse" protection ;)

[Dontworry]

Now you are talking about something different. Now you are talking about the reverse protection. Before this you highlighted passages directed toward normal protection.

Derrek Lee would have more RBI's if he had more runners on, that is an inescapable truth.

We arent getting anywhere with this arguement, So let's leave it at that.

[iPod]

This is something to be especially wary of when players change leagues when one league is particularly stronger than the other, or when a player joins a team that has the premier pitching staff in the league.

In 2001, the Oakland A's pitching staff had a combined 121 ERA+, Jason Giambi and his 202 OPS+ didn't have to hit against any of those guys. A-Rod's (164 OPS+) Rangers has combined ERA+ of 78.

That is a 43% difference. Giambi gets 17 games to hit against those pitchers, A-Rod gets none. A-Rod has to face the A's staff 17 times, Giambi, none.

So, the 38 OPS+ difference between the two wouldn't disappear no matter what, the difference is really large. But remember, Giambi gets about 70 PAs against the worst staff in the league and none against the best. A-Rod gets 70 against the best and none against the worst.

I've never really stopped to think about that just because I always have assumed those things would be too small to make any real difference. But, when you put it that way, that does sound a little dramatic.

[digglahhh]

Yeah, it was surprising to me as well. This was an extreme case, and I just really wanted to illustrate a point. I was trying to think of an example so I was justing thinking in my head and trying to find two guys with monster OPS+ but one on a team with good pitching and one with poor pitching- in the same division so they would play eachother a lot. I thought about Giambi, and then checked out A-Rod and the awful staff the Rangers trot out perenially. Ir turned out to be the virtul apotheosis of what I wanted to communicate. I always feel really good when I look for an example of something I want to say and find, like a perfect one on the first try.

Obviously, it goes without saying that the same phenomenon happens for park factor/home and road splits, as the player who hits in the good home park doesn't have that great park factored into his road splits to draw a comparison. Colorado plays no road games at Coors, but a divisional opponent plays, what, nine.