Catcher Defense (from old forums)

[cummings2]

I've read some questions re: catcher's defense, so I thought of moving here the thread in the old forums where some of this was discussed...and also for selfish reason since I like it :wink:

[cummings2]

LA Bear
Professional
Posted July 28, 2005 12:22 AM
I saw this discussed somewhat last year, but does CATCHER defense really impact the results of a game. I'd think that only catcher throwing arm would really matter and the e rating; pb rating or catcher range rating would be infrequently impact game outcomes. What do others think or probaly know about this.


Valen
Veteran
Posted July 29, 2005 04:26 PM
I have never came across anyone who even pretended to care about a catcher field rating or any of the other catcher ratings beyone throwing arm which can shut down a running game team.


Wavy
Professional
Posted July 29, 2005 05:05 PM
One advantage to having a good range rating is when trying to throw a runner out at home.


fatty
Veteran
Posted July 30, 2005 08:03 AM
I think the throwing error rating probably should be considered. I have Irod as an e(3) with 8 errors in less than 100 games. I'm assuming most of them must be throwing as I think his throwing error rating is high if I understand it correctly. Would I still take him? Definitely yes. Also, a guy like J Burke(e(15) I think) made as many errors as a decent middle infielder when I used him.


luckyman
All-Star
Posted July 30, 2005 08:22 AM
[edit: emphasis put on range]

Personally, when I looked into the matter a few months ago, I was surprised by the impact of range.

Range has a bigger impact than passed balls; not only do you have more risk of passed balls and wild pitches in the range chart, but a high range will result in singles instead of outs in some occasions.

And as wavy says, it also has an impact when a play is made at the plate. For every 20 runners that try to score, a 1-range catcher will have about 2 additional outs compared to a 5-range catcher.

The thing with arm rating is that it really works in combination with pitcher's hold ratings.

Typical example: you have a Shea team, you got four lefty pitchers in your rotation. Without having taking a look, all four lefty pitchers have great hold ratings (-4, -3).

In such a squad, the arm rating for catchers becomes almost unimportant. Whether your catcher is +1 or -3, the net impact on your opponent's running game will be almost the same. In this case, range, e-rating, T-rating, and even pb will have more impact than arm rating.


luckyman
All-Star
Posted July 30, 2005 08:35 AM
The other thing that I found is that, if your pitching squad has a hold rating of about zero, then an arm rating of +2 compared to an arm rating of 0 will hurt much more than an arm rating of 0 compared to an arm rating of -2.

I found this not by theoretical ideas, but by looking at the data that a Strat member has gathered together after many seasons.

My conclusion is that it is much more important to avoid bad (positive) hold/arm combinations than to have a strong (negative) hold/arm combinations.


luckyman
All-Star
Posted July 30, 2005 08:44 AM
For what is worth, here was my top 20 catchers, in terms of defensive ratings, once I tried to capture most important features about this position.

I assumed a pitching squad with a hold of 0.


Matheny c-1(-1)e1,T-1(pb-0) 1b-5e25
Schneider c-2(-3)e1,T-2(pb-1)
Ausmus, c-1(0)e1,T-5(pb-0)
Miller, c-2(-1)e1,T-1(pb-0)
Molina, c-2(-2)e6,T-3(pb-1) 1b-5e25
Varitek c-2(0)e1,T-2(pb-1)
Molina, c-2(-1)e2,T-4(pb-2)
Cash,K c-2(-2)e5,T-1(pb-4)
Bennett c-2(0)e3,T-6(pb-1)
Mauer,J c-2(-1)e1,T-14(pb-0)
Torrealba c-2(0)e1,T-8(pb-1)
Blanco,H c-2(-2)e2,T-13(pb-2)
Castillo c-3(-1)e1,T-5(pb-0)
Hernande c-2(-1)e3,T-7(pb-3)
Moeller c-3(0)e1,T-1(pb-0)
Rodrigue c-2(-3)e3,T-20(pb-0)
Wilson,D c-3(0)e2,T-2(pb-0)
Hall,T c-3(-2)e1,T-10(pb-1)
Mayne,B* c-3(-1)e4,T-2(pb-3) 1b-4e25
Castro,R c-3(0)e1,T-18(pb-0)



As you can see, Matheny made it above Schneider because the range had more impact than an arm rating difference of 2.

IRod was so low because of the T-rating. His value increases if pitchers have bad holds, though.

A catcher like Quiroz (not that he's popular, but just to illustrate) was among the worst despite his -1, because all other ratings were really bad.

Take note though that no one in that top 20 list has a positive arm rating. Ultimately, having a positive arm rating is the thing to avoid when you're looking for a good defensive catcher.


Dean
Strat-O ergo sum
Professional
Posted July 30, 2005 02:56 PM
Here is the thing. When there are no runners on -- which is most of the time -- catchers' range literally does not matter. In those situations, there is a 1 in 20 chance of a squib single, and that is it. If you take errors and "rare plays" (rare plays happen 3% of the time overall) out of the mix, you'll be out on a bases-empty CATCH-X in 95% of the remaining chances, regardless of whether the catcher is Pudge Rodriguez or Slowpoke Rodriguez.

On top of that, even when men are on base, all catchers (again, taking errors and rare plays out) will have an equal (20%) chance of allowing an automatic WP on a CATCH-X.

The only difference catchers' range ratings ever make is to the chance that, with men on base, the catcher will have to check his PB rating; if he fails that check, it's a passed ball, otherwise, it's a popout. For a 1 range, there's a 15% chance of this result; 2 range = 30% chance; 3 range = 45% chance; 4 range = 60% chance. (Again, those are minus errors and rare plays.)

So, to sum up, catchers' range never matters when the bases are empty. Worse range catchers have a larger chance that they will have to pass a PB check with men on base. However, even that chance doesn't differ a tremendous amount between most catchers (going up 15% for each range rating higher than 1)... and additionally, if the catcher's PB rating is good, he will pass those checks the vast majority of the time and the chances will still be outs. Basically, catcher range is not very important at all.

--
I'm sorry... the card says "moops."


1918ers
Rookie
Posted July 30, 2005 06:20 PM
quote:
Miller, c-2(-1)e1,T-1(pb-0)


so if your pitching staff pretty much has all negative holds would you rather have miller's D or Buck's O ?????


luckyman
All-Star
Posted July 31, 2005 10:31 AM
Dean,

you are saying that range for catchers isn't important because it does not impact when there is no player on base, except for the few singles that distinguish the best range from the worst.

The thing is, the same is true for arm ratings, for pb, actually for all defensive ratings for catchers except errors, which are relatively rare.

So, your argument, at best, is that catcher's range isn't as important as in other position (which is true), but your argument doesn't hold when we limit the discussion on catchers.

My point is: just like all defensive ratings for catchers, range is mostly important with men on base, and among the different ratings found for the catcher's spot, range has a surprising importance.


Dean
Strat-O ergo sum
Professional
Posted July 31, 2005 03:26 PM
Okay, so I'm guessing we agree that range at catcher is less important than at any position besides, arguably, pitcher (not sure about that one).

That said, on the relatively rare occasions that it does come into play, I still have a problem saying that it's "important." Because, even if there are men on base, and even if a CATCH-X is rolled... a 5 range catcher will still be no worse than a 1 range catcher, if his PB rating is 0. The 5 range catcher will most likely roll a PB check... which he'll always proceed to pass with his 0 rating. So PB rating is really the thing driving this. If you have a PB 20 catcher, then hell yeah, range starts mattering, because that 15% higher chance per range rating of a PB check is going to lead to a 15% increase per range rating in actual passed balls. But as long as you have a relatively low PB rating, the differences are going to be slight. It's the combo of range * PB that matters, with PB dominating the formula.

--
I'm sorry... the card says "moops."


luckyman
All-Star
Posted July 31, 2005 03:49 PM
quote:
a 5 range catcher will still be no worse than a 1 range catcher, if his PB rating is 0


I didn't catch this on your first intervention. Hence the confusion.

I don't know who's right or wrong, but that is definitively not the understanding I have of the rules. I don't have the charts nor the instructions with me, but I clearly understood that if, in the defensive chart, you read pb (passed ball), you would get a passed ball irrespective of the pb rating.

My understanding is that the pb-rating gets into action only when rolling the dices for the runner, that is, when the dices are rolled in order to check if a balk, a pb, or a cut-off has happened while a runner is standing on base.


Dean
Strat-O ergo sum
Professional
Posted July 31, 2005 04:35 PM
Right, that's the issue then. To be very specific, on the fielding charts, we're mostly talking about the P/P and P/F results. Assuming no error or rare play on those results, the chart reads:
quote:
NOTE: When using the BK/WP/PB rules, refer to the super advanced instruction sheet for procedure to determine if a PB occurs!
Then when you look at the super-advanced instructions, it basically explains that if you roll the catcher's PB rating or lower on a d20, it's a passed ball; otherwise (going back to the fielding charts), it's a popout.

So, if PB = 0, literally the only difference between 1 range and 5 range is whether groundouts to the catcher are gbA, gbB or gbC. Groundouts don't come up much though (20% of the time for the 1 range, 10% for the 5), and either way they are still outs, it's just a question of who advances.

--
I'm sorry... the card says "moops."


luckyman
All-Star
Posted July 31, 2005 05:34 PM
...while, in my understanding, the pb rating doesn't apply for the p/p p/f readings, but only in the results of the pb/wp/bk rule.

I don't want to look stubborn, but I would hesitate to give all credence on the sentence that you read on the charts.

Further, I find your understanding of the rules unadequate if one wants to reproduce the defensive stats.

Both Matheny and Ausmus have a rating of c-1 pb=0. In about 1000 innings apiece, both finished the 2004 season with 2 pb.

Both Pudge and Varitek have rating a rating of c-2 pb=1. In 1036 innings, Varitek has 5 pb, and in 1051 innings, Pudge has 3 pb.

Both Lopez and Barrett finish with a little more than 1000 innings in 2004 (1080 each, approximately). Both have c-4 pb=1 ratings. Lopez finished the season with 10 passed balls, and Barrett with 8 pb.

I have to acknowledge that I took only a selection of catchers. If you look at all catchers, you will see some discrepancies, but overall, they clearly show that Strat expects catchers with higher range to have more passed balls than catchers with low range.


Dean
Strat-O ergo sum
Professional
Posted July 31, 2005 08:52 PM
Okay, we're both wrong but I'm closer to right :P Check out the rules:
quote:
29.3 FIELDING CHART ADJUSTMENTS

The individual wild-pitch ratings also simulate the difficulty catchers have with some wild pitchers, such as knuckleballers. When using this system, the "P/P" and "P/F" readings from the Catcher X-Chart Range Section of the Super Advanced Fielding Chart are handled differently.

These readings indicate that a possible passed ball occurs, with runner(s) on base and if no error occurs on the play.

If those conditions exist, refer to the pitcher's wild-pitch rating and roll the 20-sided die. If the rolled number is less than or equal to the pitcher's rating, then a passed ball occurs and all baserunners advance one base.

If the rolled number is higher than the pitcher's rating, no passed ball, and the batter pops out to the catcher.
So, it's not a PB check as I stated... but it isn't an automatic PB either... it's a WP check. So I should say, catcher range is not at all significant unless the pitcher on the mound has a high WP. Which is more common than a high PB, probably (at least among quality players.) But is still not going to be a big issue most of the time, especially if you're choosing between a 3 vs. 4 catcher or some small difference like that.

That concludes the meaningful part of my message. I'm going to include some math here just because I wrote it... after I read the rules, I realized it was irrelevant, but I don't want to put it to waste. Everyone who isn't a numbers geek can move on :P

quote:

Probability of CATCH-X on any given roll: 1.4%
Probability men will be on base at the time: Obviously depends on the team's pitching/defense, but my data suggests 44% for the 2004 NL as a whole
Probability that a 4-range catcher like Barrett will get a P/P or P/F: 60%
Number of batters faced in a season: For the Cubs last year, it was 6,262

So, if your interpretation were correct, Barrett would have (.014 * .44 * .6 * 6,262) = 23 PB in a season, just off CATCH-Xs! Now further consider that, every time a runner is on base, there is also a PB check that will come up 2.5% of the time (in board game rules terms, rolling a 2 on the d20 and a 4, 5 or 6 on the d6). With Barrett's PB = 1 (.05 chance), that would give at least (.025 * .05 * .44 * 6,262) = 3.5 more errors, for a total of at least 26 or 27 PB on the year, clearly far too many. (I say "at least" because there are more than 6,262 rolls in this case, as a WP/PB/balk check requires another roll for the same batter, whether it succeeds or fails.)


--
I'm sorry... the card says "moops."


luckyman
All-Star
Posted July 31, 2005 09:18 PM
We're getting closer to the Truth!

Again, I can't see the chart from myself, this would help me a lot (unfortunately the site I usually refer is down).

Clearly I was wrong above.

But now that you refresh my memory with these symbols, I recalled that P/P is concerned with passed balls, and P/F is concerned with wild pitches (or perhaps it is the opposite).

So, if my memory serves well, the quote in 29.3 should be read:


quote:

When using this system, the "P/F" readings from the Catcher X-Chart Range Section of the Super Advanced Fielding Chart are handled differently.



In other words, "P/F" are handled differently, "P/P" are handled as before, that is, a straight passed ball.

(a disclaimer, perhaps I am all mixing symbols: perhaps the symbols used for passed balls are altogether different from "P/X"; do you have a list of all symbols employed in the catcher ratings range?).

Again, if my memory serves well, your mathematical formula should read as this instead:


quote:

(.014 * .44 * .6 * 6,262) *50% = 12.5 PB
12.5PB + 3.5PB = 16PB

Barrett had about 70% of catching time, thus 16PB * 0.7 = about 11.5 PB, not too far off the 8pb and Lopez's 10 pb, and certainly closer than the expected 3.5 PB *70% = 2.5 PB that we get from your own assumption


Dean
Strat-O ergo sum
Professional
Posted July 31, 2005 10:23 PM
The catcher fielding symbols are: W/S; W/G; P/P; P/F; PO; FO; G1; G2; and G3.

Well, when you start arguing that the rules are written wrong, I dunno where to go from there. I don't have a paper copy of the super advanced rules, unfortunately. But both online copies of the rules I found, the one above and this one (under "Fielding Chart Adjustments"), have the same thing about a WP check on a P/P or P/F, that results in a PB if failed.

(If you're curious what the difference between them is, basically the P/F is a foul pop. If there's an error, the same batter is up again. On the P/P, if there's an error, it's a "squib out in front of the plate", and the batter gets on base and runners advance. The P/F rare play is catcher's interference; the P/P rare play is a PB with runners on, or if bases are empty, a strikeout + PB placing the batter on first.)

--
I'm sorry... the card says "moops."


luckyman
All-Star
Posted July 31, 2005 11:02 PM
EDIT


luckyman
All-Star
Posted July 31, 2005 11:18 PM
Sorry for my confusion!!! But after reading for a fifth time the Strat rule, I finally understood.

So, if a catcher has a bad range (say "4" instead of "1"), then the team with the bad catcher will roll the dice (and check it to the wild pitch number of the pitcher) many more times than the team with the good catcher.

Even though we look at the wild pitch of the pitcher, the official scoring in the books will be "passed ball", not "wild pitch".

So then it comes back to the empirical finding that I had observed, and that, with all respect, Dean, you had recused:

catchers with bad range will allow much more official passed balls than catchers with excellent range.

Remember what I said in my first or second post: when we look at the defensive statistics, as provided by TSN-SOM, teams with bad catchers allow many more passed balls than teams with good catchers.

In this example, the team with the worst pb stats is an outlier (it had Munson c-5pb-20 for a third of the games played).

The second team with the most pb had Lopez. The team had 17 pb, even though Lopez has an excellent pb=1. But Lopez does have a bad range (c=4). So we can assume that many of the wp done by the pitchers were marked "passed ball" due to the bad range of Lopez.

The team with zero pb had IRod (pb=0, range=2). We can thus assume that many of the pitchers in the dirt were saved thanks to IRod's excellence.

It is interesting to note the team with IRod had slightly better wp numbers than the team with Lopez, probably explaining some of the difference as well.

So, bottom line, if you have pitchers with no wp, the range is not important. Range will take more importance as your pitchers will have higher wp numbers.

So, the correct sentences are:


quote:

a 5 range catcher will still be no worse than a 1 range catcher, if all pitchers WP ratings are 0


and


quote:

the pb rating doesn't apply for the p/p p/f readings, but only in the results of the pb/wp/bk rule. However, the wp rating will apply.

This message has been edited. Last edited by: luckyman, July 31, 2005 11:28 PM


GKHD11A
Veteran
Posted July 31, 2005 11:39 PM
Doesn't the range of catchers come into play on bunts?


luckyman
All-Star
Posted July 31, 2005 11:43 PM
And the correct mathematics are:


quote:

(.014 * .44 * (range*0.15) * 6,262) * 35% (average wp of 7) = k PB (coming from wp)
k PB + 3.5* PB_rating (coming from pb rating) = TOTAL PB

This falls in line with Barrett ratings


luckyman
All-Star
Posted July 31, 2005 11:58 PM
In all honesty (and thanks to Dean vigilence), I did overvalue the impact of range (I assumed that 100%, not 35%, of the P/X readings would result in pb).

I did fast, but here would be the new ranking, again assuming an average of hold=0 and wp=7 on the part of the pitcheres:

Matheny,M
Schneider,B*
Miller,D
Ausmus,B
Molina,J
Castillo,A
Moeller,C
Varitek,J+
Wilson,D
Molina,B
Mauer,J*
Cash,K
Hall,T
Bennett,G
Castro,R
Kendall,J
Torrealba,Y
Rodriguez,I
Blanco,H
Mayne,B*


luckyman
All-Star
Posted July 31, 2005 11:58 PM
quote:
Doesn't the range of catchers come into play on bunts?


Nope, at least as far as I know.


Dean
Strat-O ergo sum
Professional
Posted August 01, 2005 01:27 AM
...



I agree!

--
I'm sorry... the card says "moops."


luckyman
All-Star
Posted August 01, 2005 01:42 AM
LOL what a day!!!