(OT) What is protection?

DWKB · April 18th, 2003, 01:37 PM

Well, maybe it'll work steve.........

With all of this discussion going on about our lineup and the thinking that we need another bat in the lineup to protect Gonzo, I was wondering:

What is protection?
If it's there, how do we prove that?
If it's not there, how do we prove that?
What would be an acceptable test to gather data from?

I do not want to address the need for another bat (anyone could use another bat), but I want to know how this other bat would help Gonzo in his personal production? Would we expect to see Gonzo's production go up with another bat in the lineup?

schillingfan · April 18th, 2003, 03:20 PM

I don't know that protection is proveable on any gross league wide basis. It certainly is a truism that is routinely accepted because it is another common sense principle.

I'd love to know what an actual pitcher like Curt says about protection, after all he is the one the theory is targeted towards. The theory is that with Larry Walker hitting after Todd Helton, Curt Schilling is more like to challenge (i.e. throw strikes) Helton because he doesn't want to walk him to be on base with a big also left handed bopper like Larry there to knock him in. OTOH with someone like Payton hitting behind Helton, Curt will have less incentive to try to get him out, so he will throw more off the plate, being willing to take the risk of him walking.

So, since that is the common wisdom, I don't believe the burden of proof exists to prove that it's true. I think any pitcher will tell you that he pitches that way. So I suggest that the burden is on you to explain and prove why you believe the commonly accepted wisdom is not true.

I suppose if I wanted to prove it was true, I'd try to see the numbers of a particular player and if they varied depending on the numbers of the player hitting behind him. It seems to me that protection is an individual thing.

DWKB · April 18th, 2003, 03:36 PM

Quote:

Originally posted by schillingfan

So, since that is the common wisdom, I don't believe the burden of proof exists to prove that it's true. I think any pitcher will tell you that he pitches that way. So I suggest that the burden is on you to explain and prove why you believe the commonly accepted wisdom is not true.

Well lots of things are percieved to be common wisdom that aren't true. There are many in history such as the earth and sun, the earths shape, etc...

There are a few in baseball (if you believe them) such as DIPS, etc.. (I'll leave the baseball ones short as I don't want to cause a ruckus).

I don't believe that something that is believed to be common wisdom should be absolved of having the burden of proof.

What I don't want to do (to late!) is have everyone resolve themselves to a certain side and then look at all the data that supports that side and write off the other stuff.

Quote:

Originally posted by schillingfan

I suppose if I wanted to prove it was true, I'd try to see the numbers of a particular player and if they varied depending on the numbers of the player hitting behind him. It seems to me that protection is an individual thing.

This is a great way to start the test and the individuals we'll look at are the ones SF used as an example. I'm going to take the last 3 years, because they're readily available and because Walker was injured for some of 200 and only got a lil over 300 ABs.

(We'll work on league wide later)

RLakin · April 18th, 2003, 04:16 PM

Quote:

Originally posted by schillingfan
It certainly is a truism that is routinely accepted because it is another common sense principle.

I concur. It seems logical to me that a pitcher would be more likely to challenge Luis Gonzalez w/ Mike Lowell behind him than with Matt Williams. It is a statistical fact that Lowell has better numbers than Williams therefore he is much more likely to do some damage. That being said, it is unlikely that a pitcher would walk or give Gonzalez bad pitches with a good hitter behind him more than w/ someone that isn't.
I'm not sure that the Schilling analogy is the best one in this case, however. Curt is baseball rarity in that he doesn't seem to pitch around anybody, including Barry Bonds.
I see your point DWKB but it seems logical to me to think that a hitter would have more success with a good hitter behind him rather than a mediocre one. And I'm not even factoring in the late inning scenarios when such situations like the one I gave you become even more magnified.

BC867 · April 18th, 2003, 05:53 PM

In addition to who hits behind your best batter (in the 3rd or 4th slot) . . . it's also important who hits in front of him.

If the bread-and-butter guys get on alot, no only will it give Gonzo, for example, a shot at more RBI's than coming up with the bases empty or a runner on first . . .

But it will aid him at bat. A pitcher with runners on base has to have his attention diverted. And, if he doesn't usually come to a stop, he has to, not pitching from his strength.

So Counsell and Baerga, versus Womack and Spivey, batting ahead of Gonzo, help him at the plate, perhaps as much as the guy hitting behind him.

RLakin · April 18th, 2003, 06:34 PM

Quote:

Originally posted by BC867
So Counsell and Baerga, versus Womack and Spivey, batting ahead of Gonzo, help him at the plate, perhaps as much as the guy hitting behind him.

I agree with everything except this. I think Spivey should be the #2 hitter on this team and he and Counsell are the best 1-2 combo for the front of this team's lineup.

DWKB · April 18th, 2003, 08:57 PM

Wow...focus people! One thing at a time!! (just kidding).

All of this talk is very interesting, but I want to know. Can any of you prove this? BC867, can you prove that having runners on base helps a batter?

DWKB · April 18th, 2003, 10:36 PM

Ok, I've done a little estimate of the Walker/Helton protection issue.

From 2000-2002, Walker hit in the #3 pos 1229 times. 867 of those times, Helton was behind him. The other 362 times, Walker was batting in front of:

Code:

2002
Player     OPS  GM
Zeile     .778   4 
Payton    .839   1 
2001
Player     OPS  GM
Cirillo   .838  20
Gant      .797   3
Brumbaugh .591   1
2000
Player     OPS  GM
Hammonds  .924  15
Cirillo   .869  51

Helton's 2000-2002 OPS was 1.161, 1.117, 1.006.

Protected represents Helton behind Walker, Unprotected represents the others:

Code:

Type	         AB	 H	 TB	 BB	 SO	 AVG	 OBP	 SLG	 OPS
Protected	867	296	547	127	156	0.341	0.426	0.631	1.056
Unprotected	362	122	208	 51	 52	0.337	0.419	0.575	0.993

If you use the Protected numbers as a guide to how Walker would have performed in the Unprotected ABs with Helton behind him instead of the other batters, the difference would look like this:

Code:

Type	H	TB	BB	SO
Raw	2	20	2	13
%age	1%	10%	4%	25%

So Walker would have 2 more hits, 20 more Total Bases (showing he seemed to hit the ball harder with Helton behind him), 2 more BBs and 13 more SOs. Seems Walker had a higher SO rate in front of Helton than he did in front of the other hitters.

RLakin · April 19th, 2003, 12:17 AM

DWKB, great stuff as usual. I've been trying to find a counter to your argument and I’ll just offer up a quick one that I found. By the way, where did you get the data for Walker/Helton? I use ESPN, but is there a quicker way to sort through those numbers (involving batting order) without revisiting lots of past box scores?

Anyway, mine involves Scott Rolen's being traded to St. Louis on July 29th, 2002. Rolen goes to StL and thus provides "protection" to #4 hitter Albert Pujols. (similar my proposal of #5 hitter Lowell to AZ for #4 Gonzalez)

Rolen's arrival seemed to affect Pujols somewhat. Pujols had his best month of the 2002 season in August (right after Rolen’s arrival) hitting .368 (previous high for any month was .305), with 9 HRs (previous high was 7), 43 hits (previous high 29), and 1.072 OPS (previous high .991). It should also be pointed out that Pujols had more at bats (117) in August than any other month but, even so, his walks dropped from 15 (in July) to 10 in August despite 25 more at bats. This suggests to me that teams weren’t pitching around him as much, as do the spike in numbers.

unc84steve · April 19th, 2003, 12:18 AM

Quote:

Originally posted by BC867
But it will aid him at bat. A pitcher with runners on base has to have his attention diverted. And, if he doesn't usually come to a stop, he has to, not pitching from his strength.

That's true. Unless it doesn't.

You also hear that having a base-stealer ahead of a hitter will "cost 30 points off a the hitter's average" because he had to take good pitches to allow the SB. (That was back in the days of big Astroturf stadia & when we paid attention to OPS--I wonder why that changed? Sorry --focus!)

BTW I totally agree about the pitching from the stretch costing a pitcher's effectiveness as an intuitive thing.

schillingfan · April 19th, 2003, 04:34 AM

DWKB I don't understand how you went from your numbers to your extrapolated numbers to the amount of extra hits, etc. From your raw numbers, even though there were more at bats, it looks like the difference should have been greater. The OPS difference was greater, which suggests more power. The strikeout increase makes sense, because he's getting more strikes pitched to him.

I suspect that when a player feels "protected" he feels less pressure to produce, because he has a stronger sense that the person behind him will pick him up (in an RISP situation).

DWKB · April 19th, 2003, 08:45 AM

I got my data through www.retrosheet.org

If you're into to information overload, this is a wonderful place to waste an afternoon.

To explain the extrapolation process, I took the amount of H, TB, BB, or SO and divided it by the # of "Protected" ABs (867). Then I multipled that fraction by the "Unprotected" ABs.

For instance, "Protected" H/ABs is .341 (hey thats the avg!) and .341*362 = 123.442 (I rounded up to 124).

124 Hs is 2 more than he actually got (122). The only one this cheats in BBs but it was late and I didn't want to do the math and we're talking a couple at the most. The major difference in the numbers is the SLG and that is certainly shown in the TB.

Now, of course, one thing that this is missing is when Walker hit in a different slot than #3 (his most common) and Helton was or was not behind him then. I plan on adding that to get a better adjustment but I thought I'd throw up some numbers to get you all excited

(edit: I decided not to add Walkers non-#3 slot stats as a lot of them were pinch hitting and such. I think adding those games would actually add noise to the study instead of taking it away)

DWKB · April 19th, 2003, 09:23 PM

Quote:

Originally posted by RLakin
DWKB, great stuff as usual. I've been trying to find a counter to your argument and I’ll just offer up a quick one that I found. By the way, where did you get the data for Walker/Helton? I use ESPN, but is there a quicker way to sort through those numbers (involving batting order) without revisiting lots of past box scores?

Anyway, mine involves Scott Rolen's being traded to St. Louis on July 29th, 2002. Rolen goes to StL and thus provides "protection" to #4 hitter Albert Pujols. (similar my proposal of #5 hitter Lowell to AZ for #4 Gonzalez)

Rolen's arrival seemed to affect Pujols somewhat. Pujols had his best month of the 2002 season in August (right after Rolen’s arrival) hitting .368 (previous high for any month was .305), with 9 HRs (previous high was 7), 43 hits (previous high 29), and 1.072 OPS (previous high .991). It should also be pointed out that Pujols had more at bats (117) in August than any other month but, even so, his walks dropped from 15 (in July) to 10 in August despite 25 more at bats. This suggests to me that teams weren’t pitching around him as much, as do the spike in numbers.

Well I looked around on retrosheet and here is what I gathered

The games go as follows

Code:

Player	        GM	 OPS
Jim Edmonds	59	0.981
Scott Rolen	39	0.860
Tino Martinez	35	0.775
Eduardo Perez	13	0.745
Edgar Renteria	5	0.803
J.D. Drew	4	0.778
Miguel Cairo	1	0.660

I don't think having Rolen after Pujols over having Edmonds after Pujols should have made a positive difference if "protection" really held form.

What I would like to do is look at all NL batters who SLG over .500 and look at the people they followed. Compare that to when the players followed somebody with a less than .500 SLG and see what I get. This will be labor intensive though so it will take awhile.

DWKB · July 1st, 2003, 08:17 AM

In case any of you are interested. The conclusion of the thread ended up with me writing an article and being lucky enough to have it posted on one of the premier baseball study sites out there (so thanks to all that participated). If any of you all would like to see it you can click here.
I have to thank Dan, Sean , and Tom over at www.baseballprimer.com for giving me this oppertunity.

Comments and questions are welcomed and appreciated.

(OT) What is protection?

The guys ahead of him, too!

Re: The guys ahead of him, too!

Re: The guys ahead of him, too!