How much does defense really matter?? An OOTP experiment!

[dickbaveta]

Originally posted by Skipaway
By Linear Weight:

With 150 SB and 74 CS, Florida's running games probably worth 5 runs for the season. Lots of show, very little substance.

For Piazza's whole career, he allowed 1186 SB and 388 CS. That's a little less than 100 runs for about 10 years of catching. League average catcher is about 3~4 runs a year.

Exactly my point. The stolen base issue is of marginal effect, and Piazza's bat so outweighs Wilson's, that it is crazy to think the Mets are better off with Wilson catching and Piazza in Baltimore. Yeah, the Mets are better off with a superior defensive catcher, but this stemmed from a comment that Mike costs the team 40 runs a year, which is just beyond reality.

And yes, Henry, the rules do use unearned runs, but I am trying to weigh a contribution here, and the rules are wholly inadequate at that...to take this to the extreme:

2 outs, nobody on. Shortstop charged with error that allows runner to reach first. Pitcher proceeds to give up homeruns to the next 12 batters. Are we really going to say that the pitcher has no fault here? All 13 runs allowed are someone else or--gasp!--nobody's fault? I think that is pretty silly. We can't pretend the pitcher didn't give up the homeruns, and while all 13 runs are unearned, that doesn't mean we can't say that the SS was x% culpable while the P was y%.

[Slant]

An earned run shall be charged every time a runner reaches home base by the aid of safe hits, sacrifice bunts, a sacrifice fly, stolen bases, putouts, fielder's choices, bases on balls, hit batters, balks or wild pitches (including a wild pitch on third strike which permits a batter to reach first base) before fielding chances have been offered to put out the offensive team. For the purpose of this rule, a defensive interference penalty shall be construed as a fielding chance. (1) A wild pitch is solely the pitcher's fault, and contributes to an earned run just as a base on balls or a balk. (b) No run shall be earned when scored by a runner who reaches first base (1) on a hit or otherwise after his time at bat is prolonged by a muffed foul fly; (2) because of interference or obstruction or (3) because of any fielding error. (c) No run shall be earned when scored by a runner whose life is prolonged by an error, if such runner would have been put out by errorless play. (d) No run shall be earned when the runner's advance is aided by an error, a passed ball, or defensive interference or obstruction, if the scorer judges that the run would not have scored without the aid of such misplay. (e) An error by a pitcher is treated exactly the same as an error by any other fielder in computing earned runs. (f) Whenever a fielding error occurs, the pitcher shall be given the benefit of the doubt in determining to which bases any runners would have advanced had the fielding of the defensive team been errorless. (g) When pitchers are changed during an inning, the relief pitcher shall not be charged with any run (earned or unearned) scored by a runner who was on base at the time he entered the game, nor for runs scored by any runner who reaches base on a fielder's choice which puts out a runner left on base by the preceding pitcher.

Explain to me how those 12 home runs are unearned runs? Sure, the guy who reached on the SS error is unearned, but not the next 12.

[dickbaveta]

Well, you did leave out the first part of the rule, which by my understanding does it:

10.18: An earned run is a run for which the pitcher is held accountable. In determining earned runs, the inning should be reconstructed without the errors (which include catcher's interference) and passed balls, and the benefit of the doubt should always be given to the pitcher in determining which bases would have been reached by errorless play.

If you reconstruct the inning I gave without errors, the grounder to the SS that was booted was out #3, therefore everything subsequent is unearned.

Also my reading of "before fielding chances have been offered to put out the offensive team" would confirm this in clause A.

[jdw]

dsvitak wrote:

> Ozzie, one year, had 621 assists....by FAR the highest
> total in the history of major league baseball.

It was 1980, and it's just 20 more than Glenn Wright had which was the prior record. That's not a big gap. In contrast, Frisch held (at least as of 1997) the record for 2Bs, and his gap to #2 was 53 assists.

Of course it's Ozzie's only appearance in the Top 20 as of 1997, so it's not like he was averaging 621 a year.

> Most shortstops are going to be considered for a gold glove
> if their assist totals are above 450.
>
> So, lets assume that Ozzie made one play per game more than
> his share...in the hole, over by second. This seems reasonable,
> I saw a couple of hundred games in person with him playing,
> and he did indeed make plays no other human could make.
>
> How many runs saved is 170 fewer hits? I would estimate 60
> or so. This would work out, using standard sabremetric values
> of about 7 wins.
>
> Seems like a lot of wins, for one defensive player, above normal.

This all sounds sort of "logical", but it was debunked several decades back.

Think of it "logically" like this. In 1980, when Ozzie was saving those 60 runs, the spread between the best run prevention team and the worst run prevention team was 139 runs - 728 vs. 589. Now think all all the things that could go into have the worst vs. best. How much of the credit for that goes to the pitching staff? How much of that gets split up among the 8 fielders out there? And how much of that goes to the park?

It shouldn't surprise anyone that the top two "run prevention" teams in the NL in 1980 were the Astros and the Dodgers, or that the worst was the Cubs. The Dodgers and Astros were good teams - they went to the 163rd game to determine who took the NL West. The Cubs were a bad team - they lost 98 games. The Astro Dome played as a strong pitchers park that year, while Dodger Stadium was a "better than average" park for pitchers. Wrigley was Wrigley back then, and an excellent park for hitters that year. A *large* chunk of that 139 gap between the Astros and the Cubs was because of the parks the teams played in. That's before we get to splitting the runs between pitching and fielding, and then the fielding part between the various positions. Do you really think that any single fielder could be responsible for a 60 run gap in there?

Maybe if the Astros had Ozzie at SS and the Cubs had HoJo at SS. But HoJo was a horrific SS, so he's not a reasonable replacement level SS.

There just aren't 60 runs there to save above the average.

I don't think that any of the newer high end sabermetric methods of analyzing defense have ever been able to locate 60 defensive runs of saving at SS vs. the average or replacement level. Bill James' win shares method certainly doesn't.

On Piazza in him prime as a catcher for the Dodgers, he wasn't costing the Dodgers as many runs with defense as some people would claim in the thread. I doubt for the Mets he's ever cost them 40 runs in a season over a replacement level catcher's defense.


John

[ChuckB]

Originally posted by jdw
dsvitak wrote:

On Piazza in him prime as a catcher for the Dodgers, he wasn't costing the Dodgers as many runs with defense as some people would claim in the thread. I doubt for the Mets he's ever cost them 40 runs in a season over a replacement level catcher's defense.

John

I dont think that anyone's saying that he is, has, or will. All I was saying is that I would prefer him at first -- where his defensive liabilities might be even more obvious and costly -- or in another uniform. I prefer a catcher who has some defensive abilities, especially when I see pitchers using the slide step to try to get the ball to the plate quicker and thereby affecting the quality of their pitches; pitchers throwing more fastballs in non-fastball counts (1-2, 0-2, even 2-2) in order to get the ball to the plate faster in hopes that it will help Piazza throw out the baserunner; pitchers throwing over to 1st much too often, and thereby affecting their own routine and momentum, trying to keep the baserunner -- even the medium ability baserunner -- closer to the bag so that Mike has a shot. And sometimes just watching their faces when a baserunner steals second and Mike fires one into centerfield tells me that they're not comfortable with him there.

At all levels I've seen effective catchers make a pitcher life easier. What more important position is there than the pitcher? If he's not comfortable, for whatever reason, then you have a less effective pitcher on the mound. You think Eddie Perez functioned as Greg Maddux' personal catcher all those years for nothing? Bobby Cox knew that a less effective Maddux made for a less effective Braves starting rotation, and you don't mess with your team's strength.

Right now, the Mets strength is their pitching (as old as it is). Why would you ever mess with their comfort level on the mound?? It's just not good baseball sense to me.

[dickbaveta]

Originally posted by ChuckB
I dont think that anyone's saying that he is, has, or will.

Well actually, someone did earlier in the thread which is how this whole conversation got started.

You think Eddie Perez functioned as Greg Maddux' personal catcher all those years for nothing? Bobby Cox knew that a less effective Maddux made for a less effective Braves starting rotation, and you don't mess with your team's strength.
[/B]

Well, Maddux did win 3 Cy Youngs before Perez was even in the majors, and also had his biggest negative outlier season in Atlanta (1999) the year Perez had his most playing time, so I'm not sure what you're saying there. (Last year was more age related I think we can all agree.)

[jmaness84]

Maddux had Charlie O'Brian before Eddie Perez

[Slant]

dickbaveta:

I guess you're right, but I always figured that in the reconstruction subsequent plays would be counted as if they were in the next inning. I've watched many baseball games and I don't think I recall that all runs scored after 2 outs in an inning with an error are unearned. I could be wrong, though.

[Crapshoot]

Originally posted by Henry
Obviously, the position your taking is 180 degrees away from the concepts and ideas from which MLB rules were based.

An error IS, in fact, NOT the pitchers fault... period. There is no way you can convince anyone otherwise. The fielders have a responsibility too - the pitcher can't field - nor is he responsible for - what happens on the field around him.

The pitcher is responsible for keeping the ball in the park - and if at all possible - keeping it near his fielders. Once the ball reaches the fielder, it's the fielder's responsibility to get the out - and if he errors, then the responsibility is on him - not the pitcher. Same goes for catcher errors.

I find it hard to believe you've taken a position contrary to baseball rules... unearned runs are not stupid - they are necessary to place responsibility where it belongs.

Henry

I was fully in agreement with you until the last point- which I strongly disagree with. THe problem with unearned runs is the effect they have on a pitcher's ERA - a pitcher who loads the bases and then has a fly ball dropped (leading to all three runs scoring) will get zero earned runs, which is a ridiculous evaluation of the pitcher's performance there. At worst, he should be held responsible for the one run that would have scored- which he is not here. A good rule of thumb as such when using ERA is to check the number of unearned runs- RA as a whole is probably a slightly better indicator of performance (once its park adjusted of course).

[Spielman]

Originally posted by jdw
dsvitak wrote:

> Ozzie, one year, had 621 assists....by FAR the highest
> total in the history of major league baseball.

It was 1980, and it's just 20 more than Glenn Wright had which was the prior record. That's not a big gap. In contrast, Frisch held (at least as of 1997) the record for 2Bs, and his gap to #2 was 53 assists.

Of course it's Ozzie's only appearance in the Top 20 as of 1997, so it's not like he was averaging 621 a year.

> Most shortstops are going to be considered for a gold glove
> if their assist totals are above 450.
>
> So, lets assume that Ozzie made one play per game more than
> his share...in the hole, over by second. This seems reasonable,
> I saw a couple of hundred games in person with him playing,
> and he did indeed make plays no other human could make.
>
> How many runs saved is 170 fewer hits? I would estimate 60
> or so. This would work out, using standard sabremetric values
> of about 7 wins.
>
> Seems like a lot of wins, for one defensive player, above normal.

This all sounds sort of "logical", but it was debunked several decades back.

Think of it "logically" like this. In 1980, when Ozzie was saving those 60 runs, the spread between the best run prevention team and the worst run prevention team was 139 runs - 728 vs. 589. Now think all all the things that could go into have the worst vs. best. How much of the credit for that goes to the pitching staff? How much of that gets split up among the 8 fielders out there? And how much of that goes to the park?

It shouldn't surprise anyone that the top two "run prevention" teams in the NL in 1980 were the Astros and the Dodgers, or that the worst was the Cubs. The Dodgers and Astros were good teams - they went to the 163rd game to determine who took the NL West. The Cubs were a bad team - they lost 98 games. The Astro Dome played as a strong pitchers park that year, while Dodger Stadium was a "better than average" park for pitchers. Wrigley was Wrigley back then, and an excellent park for hitters that year. A *large* chunk of that 139 gap between the Astros and the Cubs was because of the parks the teams played in. That's before we get to splitting the runs between pitching and fielding, and then the fielding part between the various positions. Do you really think that any single fielder could be responsible for a 60 run gap in there?

Maybe if the Astros had Ozzie at SS and the Cubs had HoJo at SS. But HoJo was a horrific SS, so he's not a reasonable replacement level SS.

There just aren't 60 runs there to save above the average.

I don't think that any of the newer high end sabermetric methods of analyzing defense have ever been able to locate 60 defensive runs of saving at SS vs. the average or replacement level. Bill James' win shares method certainly doesn't.

On Piazza in him prime as a catcher for the Dodgers, he wasn't costing the Dodgers as many runs with defense as some people would claim in the thread. I doubt for the Mets he's ever cost them 40 runs in a season over a replacement level catcher's defense.


John

No freaking kidding.

I love these things that come up regularly where people start throwing around impressive numbers of runs or hits that some player has saved defensively without any underlying logic.

Another, more pithy argument against dsvitak's numbers: In that year of 621 assists, Ozzie averaged 3.93 assists per game. To believe that he was getting to one more ball per game than an average shortstop, we'd have to believe that a full QUARTER of the balls he reached were plays that an average SS couldn't have made.

I'm a Cardinal fan. I love Ozzie. But that assumption is just plain insane.

-Spielman

[Henry]

Originally posted by Aadik
I was fully in agreement with you until the last point- which I strongly disagree with. THe problem with unearned runs is the effect they have on a pitcher's ERA - a pitcher who loads the bases and then has a fly ball dropped (leading to all three runs scoring) will get zero earned runs, which is a ridiculous evaluation of the pitcher's performance there. At worst, he should be held responsible for the one run that would have scored- which he is not here. A good rule of thumb as such when using ERA is to check the number of unearned runs- RA as a whole is probably a slightly better indicator of performance (once its park adjusted of course).

The problem (in my mind) with this is that the action of the fielder is something the pitcher can do nothing about. Had the fielder caught the ball in your example, the pitcher "did his job". If that's true, then the only reason the runs scored at all is because the fielder "did not do his job" - thus the pitcher isn't held responsible.

Look at this example....

Two outs and the batter hits a bouncer to the SS who bobbles it. Next guy hits a homerun - two unearned runs. Had the SS handled the ball and thrown the third guy out, the homerun would have not occured - or would it have ?? Assuming everything being equal, the 4th batter would have hit his homerun batting first in the next inning - making it an earned run.

I could see an arguement that ONLY batters who reached base on an error, or equivalent mishap, should be unearned. In that case, my homerun example would have resulted in one earned, and one unearned run. In your example, the first of the 3 runs scored would be unearned (due to the dropped fly) but the other two would be earned.

Six of one - half dozen of the other... I think the real issue is that earned runs (regardless of how they are evaluated) are not a very good way of rating a pitcher. It takes into account the fielders - which really have little to do with the quality of the pitcher. I think BAA or OBPA combined with strikeouts and ground balls is probably the best way to see the true value of a pitcher.

Henry

[Crapshoot]

Henry, I think we agree- my emphasis was only on the one run that would have been earned in my example even if the fielder had caught the ball (the sac fly). As for the correct way to evaluate pitchers- you're sounding more and more like a stathead- what you're describing is basically DIPS- Defense Independent Pitching Statistics- and Voros McCracken's work. Check out www.baseballprimer.com for a lot more information on that...