What's the one word that makes your eyes bleed?

[dickay]

That may seem like "common sense", but the results don't bear it out.

I could've sworn that "The Book" had a section on protection, but I can't seem to find it. At any rate, Baseball Between the Numbers does, in the chapter dealing with lineup order titled "Was Billy Martin Crazy?"

They first show a graph of the performance of all players in 2004 that hit ahead of low-quality hitters, medium-low quality hitters, medium quality hitters, medium-high quality hitters, and high quality hitters. They show average, on-base percentage, slugging percentage, walk rate, and intentional walk rate, but for simplicity's sake, I'll just show the OPS:

Hitting in front of ___ quality hitters;

Low: .728
Med-low: .744
Med: .764
Med-high: .777
High: .796

At first glance, this would "confirm" the existence of protection. However, as the book goes on to note, there's serious problems with such a study. Lineups tend to be grouped by ability. The best hitters bat at the top or in the middle, and the worst hitters bat at the bottom. Therefore, players who bat in front of high quality hitters are more likely to be high quality hitters themselves. So, when you look at the performance of ALL players depending on who hit behind them, you're going to see a higher quality of performance from players ahead of high quality hitters simply because the methods of lineup construction will more often than not place high quality hitters in front of high quality hitters.

They correct for this, however, by comparing those numbers to what we'd expect the batters to do based on their overall season numbers. After doing that, rather than a steady upward progression, there's little chance in performance. Below is a chart of the change in OPS depending on the quality of the next hitter in the order.

Low: .008
Med-low: -.002
Med: -.002
Med-high: .010
High: -.013

There's no pattern at all here, and very little change in OPS depending on the next batter. As the chapter concludes on the protection issue:

"Protection is overrated. There's no evidence that having a superior batter behind another batter provides the initial batter with better pitches to hit; if it does, those batters see no improvement in performance as a result."

There's numerous other studies all over the internet that you can turn up with simple Google searches. If you're interested, I'd be happy to link to further ones.

As you said, it "seems like common sense". That is why I can't take the stats you posted a face value. Stats can easily be manipulated, as in calling Kent a superb hitter who follows Bonds. Again, protection for Kent is equally as important IMO. I dont' know if that makes sense...but what i'm saying is I really don't think Ortiz will have that big of a dropoff without manny because they have a deep and dangerous lineup regardless. Behind Manny is other very quality bats. After Kent (who was much less than 'superb') was who, pedro ortiz?

To me...as I said...if one would believe a lineup full of Barry Bonds would cause pitchers to pitch MORE to Barry Bonds and not walk him as much, then that is clear evidence that pitchers pitch differently depending on the lineup. If they are pitching differently, its clear that the 'protection' is causing them to. Yes an extreme example which would have the most drastic and evident change....what i'm saying is if its done there, then its very likely done on a much smaller scale at times outside of fantasy land. I just don't see how that logic doesn't make sense.

[dickay]

Oh, and it just dawned on me. How bout the smartest man in baseball Tony Larussa walking Bonds with the bases loaded? Do you think he walks Bonds with Manny Ramirez, Alex Rodriguez, or a bat of that caliber hitting behind him in that situation?

[HoustonGM]

As you said, it "seems like common sense". That is why I can't take the stats you posted a face value.

I suggest you look at the facts instead of accepting what "seems like common sense" at face value. It doesn't take much research to see that protection has been debunked.

Stats can easily be manipulated, as in calling Kent a superb hitter who follows Bonds.

No stats are being "manipulated" here. They're simply showing what happened. Player performance does not show a meaningful change based on the following hitters in the lineup.

Again, protection for Kent is equally as important IMO. I dont' know if that makes sense...but what i'm saying is I really don't think Ortiz will have that big of a dropoff without manny because they have a deep and dangerous lineup regardless. Behind Manny is other very quality bats.

Ortiz wouldn't have a big dropoff regardless of his lineup. Why? Because he's a great hitter. The players around him don't make him great. He IS great. I think all these attempts to try to attribute a player's performance to factors outside of...his performance....is a disservice to the player and disrespectful to his abilities.

After Kent (who was much less than 'superb') was who, pedro ortiz?

In 2003 and 2004, Bonds hit mostly in front of Edgardo Alfonzo and Benito Santiago, not Jeff Kent as he did in 2001 and 2002. And, you know what? The only thing that changed in his performance was...his walk rate...despite hitting in front of meaningfully worse hitters than he did previously.

To me...as I said...if one would believe a lineup full of Barry Bonds would cause pitchers to pitch MORE to Barry Bonds and not walk him as much, then that is clear evidence that pitchers pitch differently depending on the lineup. If they are pitching differently, its clear that the 'protection' is causing them to. Yes an extreme example which would have the most drastic and evident change....what i'm saying is if its done there, then its very likely done on a much smaller scale at times outside of fantasy land. I just don't see how that logic doesn't make sense.

Man, nobody is saying that pitchers don't pitch players differently depending on the lineup. All anybody is saying is that that has no effect on the performance of the batters. I'll refer you again to the conclusion of that section of the chapter I referenced:

"Protection is overrated. There's no evidence that having a superior batter behind another batter provides the initial batter with better pitches to hit; if it does, those batters see no improvement in performance as a result."

Oh, and it just dawned on me. How bout the smartest man in baseball Tony Larussa walking Bonds with the bases loaded? Do you think he walks Bonds with Manny Ramirez, Alex Rodriguez, or a bat of that caliber hitting behind him in that situation?

I don't know why you keep trying to hammer home the point about walks. It's been established, and I've said repeatedly, that walk rates, particularly intentional walk rates, are affected by the next batter in the lineup. The most obvious example of this is #8 hitters being walked to face the pitcher. Nobody is disputing that.

[filihok]

http://www.whatsatararrel.com/2009/0...tatistics.html

I mean, any time you’re trying to figure out how often something happens, which is better: trying to remember how often it felt like happening, or going back and counting how many times it actually did happen? Clearly, counting, that is, figuring out what the actual facts are, rather than just what they seemed like, is the way to go.

The caveat being you have to make sure you are counting what you think you are counting

[dickay]

OK, i finally figured a way to explain this to you HGM, in a manner that I don't think even you can dispute. Providing protection is about maximizing opportunity. Obviously batting Bonds or a great bat in front of a pitcher (your example) is going to lead to alot of walks. Having someone like Manny hitting behind Bonds is, as you've already agreed upon, going to reduce the amount of walks Bonds faces.

If Bonds is a .300 hitter who hits 1HR per every 10 at bats and gets an RBI every 5 at bats (arbitrary numbers) his numbers aren't going to change much at all, but he is getting more at bats. If he walks 30 less times in a season (arbitrary) than in those 30 at bats he should still be about a .300 hitter, get an additional 3 HR's and an additional 6 RBI's. There's you're extra production. Nobody has said it definetly makes the batter more talented, but they get more opportunity. Just as in basketball when a guy is double teamed, another is open and benefits....you asked for a correlation in baseball and there it is! More opportunity due to the less walks.

"Protection is overrated. There's no evidence that having a superior batter behind another batter provides the initial batter with better pitches to hit; if it does, those batters see no improvement in performance as a result."

The argument wasn't that it is overrated...i wouldn't be disagreeing on you there because I really don't have an opinion either way. The argument was that it doesn't exist...ie, is a "myth". Clearly it exists...as the protection for the batter provides more opportunity for production.

Now, I feel protection is fact, not myth and I think what I posted above makes it clear.

[HoustonGM]

If Bonds is a .300 hitter who hits 1HR per every 10 at bats and gets an RBI every 5 at bats (arbitrary numbers) his numbers aren't going to change much at all, but he is getting more at bats. If he walks 30 less times in a season (arbitrary) than in those 30 at bats he should still be about a .300 hitter, get an additional 3 HR's and an additional 6 RBI's. There's you're extra production. Nobody has said it definetly makes the batter more talented, but they get more opportunity. Just as in basketball when a guy is double teamed, another is open and benefits....you asked for a correlation in baseball and there it is! More opportunity due to the less walks.

The argument wasn't that it is overrated...i wouldn't be disagreeing on you there because I really don't have an opinion either way. The argument was that it doesn't exist...ie, is a "myth". Clearly it exists...as the protection for the batter provides more opportunity for production.

Now, I feel protection is fact, not myth and I think what I posted above makes it clear

Ah, but, you're forgetting one key component - he also will be making more outs. Thus, in the end, the overall production remains the same. Using your numbers, if he walks 30 less times, that means 30 plate appearances divvied up between singles, doubles, triples, home runs, and outs. If he's a .300 hitter that hits a homer once every 10 at bats, that'll mean 10 hits, 3 of which are home runs and 7 of which are of another sort, and 20 outs.

[dickay]

Ah, but, you're forgetting one key component - he also will be making more outs. Thus, in the end, the overall production remains the same. Using your numbers, if he walks 30 less times, that means 30 plate appearances divvied up between singles, doubles, triples, home runs, and outs. If he's a .300 hitter that hits a homer once every 10 at bats, that'll mean 10 hits, 3 of which are home runs and 7 of which are of another sort, and 20 outs.

the .300 average, extra HR's and RBI's are more far more valuable than the walks IMO. You asked for an example of increased production....i think i've made that point clear. And there is a such thing as a 'productive' out as well. Can't really guage that in this argument.

[HoustonGM]

the .300 average, extra HR's and RBI's are more far more valuable than the walks IMO. You asked for an example of increased production....i think i've made that point clear. And there is a such thing as a 'productive' out as well. Can't really guage that in this argument.

If those 30 plate appearances were made up solely of hits, than it'd be more valuable, but 30 walks is more valuable than 10 hits and 20 outs.

You have not shown that protection increases a player's production. You've continued to speak in terms of how things "seem" like they should be. The fact of the matter is that, in terms of getting hits and power production, players show no change based on who follows them in the order. I've shown you the exact numbers, at least in terms of the 2004 season. I'd be happy to dig up more articles and link you to them if you wish to explore the issue further. It goes without saying that I suggest you do research the issue instead of being beholden to what seems like common sense, but I'm not sure what else could be shown to you to convince you that player production is not affected by the following hitters in the order.

[dickay]

If those 30 plate appearances were made up solely of hits, than it'd be more valuable, but 30 walks is more valuable than 10 hits and 20 outs.

You have not shown that protection increases a player's production. You've continued to speak in terms of how things "seem" like they should be. The fact of the matter is that, in terms of getting hits and power production, players show no change based on who follows them in the order. I've shown you the exact numbers, at least in terms of the 2004 season. I'd be happy to dig up more articles and link you to them if you wish to explore the issue further. It goes without saying that I suggest you do research the issue instead of being beholden to what seems like common sense, but I'm not sure what else could be shown to you to convince you that player production is not affected by the following hitters in the order.

lmao.....and i am thinking the same thing. I don't know what more can be shown to you to prove that production is increased. Players get more hits, and more power production (more HR's and RBI's) due to increased opportunity. Thats logic, plain and simple. Yes it comes at the sacrifice of walks, but you can't simply call it a wash. If walks were so much more valuable, Manny & Bonds would bat in front of the pitcher where they'd be walked so much more. If walks were so much more valuable, teams wouldn't be so quick to walk someone like Bonds instead of let him bat. You seem to be refusing pure logic so you can hold onto your premise, that protection is a myth. Thats what I mean by closed mind.

[Swampdog]

OK, i finally figured a way to explain this to you HGM, in a manner that I don't think even you can dispute. Providing protection is about maximizing opportunity. Obviously batting Bonds or a great bat in front of a pitcher (your example) is going to lead to alot of walks. Having someone like Manny hitting behind Bonds is, as you've already agreed upon, going to reduce the amount of walks Bonds faces.

If Bonds is a .300 hitter who hits 1HR per every 10 at bats and gets an RBI every 5 at bats (arbitrary numbers) his numbers aren't going to change much at all, but he is getting more at bats. If he walks 30 less times in a season (arbitrary) than in those 30 at bats he should still be about a .300 hitter, get an additional 3 HR's and an additional 6 RBI's. There's you're extra production. Nobody has said it definetly makes the batter more talented, but they get more opportunity. Just as in basketball when a guy is double teamed, another is open and benefits....you asked for a correlation in baseball and there it is! More opportunity due to the less walks.



The argument wasn't that it is overrated...i wouldn't be disagreeing on you there because I really don't have an opinion either way. The argument was that it doesn't exist...ie, is a "myth". Clearly it exists...as the protection for the batter provides more opportunity for production.

Now, I feel protection is fact, not myth and I think what I posted above makes it clear.

Flawed logic. First, even Dickay admits that the alleged "protection" will not make Bonds better, per se. Thats a good first step. It's supposed, though, that by giving Bonds another 30 at bats, he will be more productive, thus helping the team score more runs, making the team better, overall. Not true, though.

In the suggested scenario, Bonds will not draw the 30 extra walks per year, but will get 30 additional at bats with better "protection". This will, in itself, not provide more runs for the team. It provides less runs. (A difference of 2 runs seems small, and it is, but we're talking about an incredible hitter. But the underlying point is that runs scored actually decrease with the "protection", instead of increase).

Based on 2007 stats (all I have handy), 30 walks has an expected value of about 10 runs, by linear weights. In 30 at bats, a .300 hitter who hits 3 homers, 2 doubles, and 4 singles (a .300 average) and makes 21 outs will provide about 8 runs. Thus, the 30 walks are of more value.

Again, numerous studies have shown that "protection" is a myth. Talking about what "seems logical", or what "makes common sense" will not change this. It's a fact. Protection does not exist. Compulsively walking a hitter to pitch to another simply results in MORE runs scored, in the long run.

I also suggest that you two give up on this debate. HGM is not going to accept speculative theories about protection in the lineup, when he has seen much research that shows it basically does not exist (for that matter, neither would I). Dickay does not place any stock in statistical analysis; he just believes that he "knows better", because it seems to be right.

It's become an exercise in futility.

[HoustonGM]

lmao.....and i am thinking the same thing. I don't know what more can be shown to you to prove that production is increased.

Anything that shows it?

Players get more hits, and more power production (more HR's and RBI's) due to increased opportunity. Thats logic, plain and simple.

Can you show me any evidence of that? I've seen plenty of evidence that that is NOT the case, and have provided you with some, and could provide you with plenty more.

Yes it comes at the sacrifice of walks, but you can't simply call it a wash.

It's not "at the sacrifice of walks" but rather "at the sacrifice of outs."

If walks were so much more valuable, Manny & Bonds would bat in front of the pitcher where they'd be walked so much more. If walks were so much more valuable, teams wouldn't be so quick to walk someone like Bonds instead of let him bat.

I have no clue what you're arguing against here. One hit is more valuable than one walk. 30 walks is more valuable than 10 hits and 20 outs.

You seem to be refusing pure logic so you can hold onto your premise, that protection is a myth. Thats what I mean by closed mind.

I'm not refusing "pure logic." I'm refusing flawed logic, which Swampdog eloquently put much better than I have.

The closed mind is not the one that refuses speculative logic based on what "seems right", but rather, the closed mind refuses facts.

[dickay]

Based on 2007 stats (all I have handy), 30 walks has an expected value of about 10 runs, by linear weights. In 30 at bats, a .300 hitter who hits 3 homers, 2 doubles, and 4 singles (a .300 average) and makes 21 outs will provide about 8 runs. Thus, the 30 walks are of more value.

And thus the problem. Logic IMO supercedes stats which scew rational thought. Stats can be very misleading....30 walks of one player is not the same as 30 walks of another player. I'd much rather prefer to have 30 at bats in which Manny is pitched to than 30 at bats in which the bat is taken out of his hands.

Over manny's career, he's averaged a HR every 14 at bats. and an RBI every 4 at bats approximately. Over 30 at bats, he'll average over 7 RBI's. That doesn't include the runs he'll score, just the RBI's, of which only 2 came from him scoring on a HR. He's also a career .314 hitter. I find it hard to believe that with only 2 at bats being a HR, the other 7-8 hits he gets won't be equivalent to at least 2-3 runs scored which would equal your 10 runs for 30 walks stat. I also find that 30 walks equals 10 runs stat extremely high, and am curious how it accounts for walks of these type of situations in which there is often nobody on base, or an open base to walk the player to.

You can play with stats however you like...there are obvious flaws with stats as well, and if you are honest with yourself you will at least agree to that.

[HoustonGM]

And thus the problem. Logic IMO supercedes stats which scew rational thought. Stats can be very misleading....

I suggest you stop throwing around the term "closed minded" then, considering that rather than examine why a stat may be misleading, you have a blanket distrust of any stat that goes against what you feel is right.

30 walks of one player is not the same as 30 walks of another player. I'd much rather prefer to have 30 at bats in which Manny is pitched to than 30 at bats in which the bat is taken out of his hands.

Whether not you yourself would "rather" it is irrelevant to the question of which provides more value.

Over manny's career, he's averaged a HR every 14 at bats. and an RBI every 4 at bats approximately. Over 30 at bats, he'll average over 7 RBI's. That doesn't include the runs he'll score, just the RBI's, of which only 2 came from him scoring on a HR.

Can we not use just a context-dependent stat as RBI in this case? It doesn't give us any information about Manny's value as a player, as it is heavily dependent on the other players that bat ahead of him. Manny's career "RBI rate" doesn't tell us anything about how many RBI he's "expected" to have over any sample of at bats. Let's focus on those things that are purely of Manny's creation.

He's also a career .314 hitter. I find it hard to believe that with only 2 at bats being a HR, the other 7-8 hits he gets won't be equivalent to at least 2-3 runs scored which would equal your 10 runs for 30 walks stat. I also find that 30 walks equals 10 runs stat extremely high, and am curious how it accounts for walks of these type of situations in which there is often nobody on base, or an open base to walk the player to.

Linear Weights.

You can play with stats however you like...there are obvious flaws with stats as well, and if you are honest with yourself you will at least agree to that.

I'll gladly agree that stats can be flawed. But, see, I don't just say "Stats are flawed" and dismiss those that don't agree with my gut instinct. I'll examine the stats and see if they are, indeed, flawed. If they're not, and they disprove my gut instinct, I'll believe the facts and change my mind.

If you've examined the issue, feel free to show me the evidence. Hell, instead of you providing me evidence that protection exists, I'll even accept "disproving" the multiple studies that show protection doesn't exist.

Here, I'll link you to more:

The Myth of Protection
The Protection Externality: It Doesn't Exist
Protection Study
Pitching Around Batters (This is an excerpt from "The Book" which I couldn't find earlier)
The Protection Mini-FAQ
Subtle Aspects of the Game

I could probably keep going if I really wanted. The fact of the matter is that every substantial study done on the issue has come to the same conclusion. It is accepted fact in the baseball analysis community that lineup protection does not exist. You're free to disregard the mountains of evidence and continue holding on to your preconceived notions and gut feelings, but is that something an open-minded person would do?

[filihok]

Because this thread has given me so much entertainment, and because I have a lot of time on my hands, I tried an experiment.

I used Baseball Musings Lineup Analysis Tool to compute the runs per game for the 1993 Giants given different scenarios.

The first scenario was to use each of the Giants regular players at each position in their regular batting order (as determined by looking at baseball reference). Using Barry Bonds (batting 5th) with OBP = .458 and SLG = .677

The result for this team was: 5.013 runs per game

I then substituted Barry Bonds with a fictional Barry Bonds who did not walk, but maintained the same slugging percentage over the same number of plate appearances. This Bonds hit: OBP = .336 and SLG = .677

The result for this team was: 4.781 runs per game

Then, I substituted that Barry Bonds with another fictional Barry Bonds who walked every time he was up to the plate. This Bonds had an OBP = 1.000 and a SLG = 1.000. I also moved him to 8th in the order hitting before the pitcher, as this was the only way I could even make a semblance of sense of Bonds walking 664 times per season.

The result for this team was: 5.706 runs per game

[HoustonGM]

The slugging percentage of a player that walks every time he comes to the plate would technically be "undefined", since Total Bases = 0 and At Bats = 0, and 0/0=undefined, which may change the calculations some, but shouldn't by too much. The point is the same - never underestimate the power of not making outs.

Also, just as to make a point about the accuracy of that lineup analysis tool. The actual Giants scored 4.99 runs per game, a near dead ringer for the 5.01 the estimator comes up with. The difference is entirely explained by the fact that the real life Giants employed more than just their 8 regulars and they didn't always bat in the exact same order.