Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Hall of Merit > Discussion
Hall of Merit
— A Look at Baseball's All-Time Best

Tuesday, June 01, 2004

1927 Ballot

Sorry it’s late . . . unexpectedly out of town for a few days - pesky holidays getting in the way of important Hall of Merit business . . .

Joey Numbaz (Scruff) Posted: June 01, 2004 at 03:05 PM | 205 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

Page 3 of 3 pages  < 1 2 3
   201. John (You Can Call Me Grandma) Murphy Posted: June 10, 2004 at 04:38 PM (#668912)
John, theer we'll have to disagree.

Well, maybe we can agree a little. I was planning on placing Beckley at the bottom of my ballot, but I have reconsidered and have decided to place Caruthers there. While I haven't changed my views on the overrating of pitchers from the 1880s, I think I have been unfairly (but not intentionally) underrating that era nevertheless. So Bob will make it at #15 this week.

I don't know how long he'll be on my ballot, but at least I'm not an "enemy" of him now.
   202. Joey Numbaz (Scruff) Posted: June 11, 2004 at 06:32 AM (#671071)
Jim - any response to my post 90 on the fact that all pitchers are overrated for their offense in WARP? I don't mean that sarcastically, I'm curious as to whether or not you agree with me, and if not, why. Thanks!
   203. Joey Numbaz (Scruff) Posted: June 11, 2004 at 06:33 AM (#671072)
Make that #190, or #90 on the #2 page of this thread . . .
   204. jimd Posted: June 11, 2004 at 09:56 PM (#671898)
all pitchers are overrated for their offense in WARP

Overrated when compared to who, and by how much?

Stepping back ...

The impression we have is that BP calculates independent replacement levels for hitting, pitching, and fielding (presumably at each position independently). Every player is rated on at least two of these three scales, and we surmise that someone who is at replacement on both scales is really below replacement because replacement hitters have some glovework going for them and replacement fielders have a live bat. So I think we all agree that BP may have replacement level set too low. The question is by "how much" is it too low and where is the excess value located?

Tango's opinion is that position players are given excess value of (FRAR-FRAA). Your (Tango's?) opinion is that pitchers are given excess value of (BRARP-BRAAP). I'm not sure it's that easy.

Removing an amount of runs from a position proportional to "playing time" is appropriate, and would not affect the ranking of players at the position relative to each other. Both of those corrections mentioned above meet that criterion. The problem comes in determining how much to remove at each position; should that amount be different at each position? Tango's method says "Yes". (See our earlier discussion and the example of Mo Vaughn vs Rey Ordonez.) This is where I question because I'm not so sure.

I think one way to investigate this is to look at the delta between "average" and "BP replacement level" by position for players at different positions with comparable playing time. (Pitchers are a problem here; what is comparable playing time?) Is the delta the same? If the delta is different at each position, does it correspond to the defensive spectrum? Should the delta be flattened, or is there a reason that an average SS is more valuable than an average 1Bman (or vice-versa)? Once this is worked out for position players, then maybe pitchers can be tackled.

I know this is a lot of work, and I don't have time for it at the moment. Maybe during vacation next month...

****

On the issue of BRAR vs BRARP in calculating WARP-1. You would expect that the conversion factor of Runs to Wins to be fairly constant through time (at least you would if the run environments have been scaled to a constant 4.5; it may not be exactly constant if some kind of linear fitting is being done each season). So I calculated the conversion factor for each of Jackson's 9 full-time seasons and for each of Caruthers 7 pitching seasons. And I calculated it using BRAR and I calculated it using BRARP instead. The results:

JJ: BRAR: run conversion factor ranges between 9.12 to 9.48, range of 4%
JJ: BRARP: run conversion factor ranges between 7.72 and 8.63, range of 12%
BC: BRAR: run conversion factor ranges between 8.97 to 9.57, range of 7%
BC: BRARP: run conversion factor ranges between 9.30 and 10.14, range of 9%
JJ/BC combined: BRAR: run conversion factor ranges between 8.97 to 9.57, range of 7%
JJ/BC combined: BRARP: run conversion factor ranges between 7.72 to 10.14, range of 31%

To me, the much wider spread on the "constant" conversion factor using BRARP says that BRARP is NOT used when calculating WARP-1. IOW, you hit as a hitter; position information is recorded in the fielding/pitching runs.
   205. Jeff M Posted: June 20, 2004 at 04:49 PM (#688896)
To me, the much wider spread on the "constant" conversion factor using BRARP says that BRARP is NOT used when calculating WARP-1.

Just getting around to this, but I agree with the statement above, based on a few seasons I examined. It is easier to see in seasons of low WARP1. As an example, compare Bench in 1982 to Yeager in 1982. Bench was mostly playing 3b, but I used a year like this intentionally because it has a nice gap between BRAR and BRARP in a low performance season.

1982.......................Bench............Yeager
BRARP+FRAR+PRAR.......12...............25
BRAR+FRAR+PRAR........17...............24
WARP1.......................1.8..............2.7
RunConvFact #1..........6.7..............9.3
RunConvFact #2..........9.4..............8.9

First, that run conversion factor for BRARP in the formula is impossibly low given the run environment in 1982.

Second, if WARP is using 4.5 r/g, then the BRAR formula makes some sense, when you factor in that WARP1 is rounded to one decimal point and the components of the formula are rounded to whole numbers.

To wit: If Bench's formula with BRAR is 16.5 (rounded up to 17), then dividing by 9 r/g would give you 1.8. The point is that the run conversion factor of 9.4 might be skewed by the rounding. If you do the same thing to BRARP, and use 12.49 (which BP rounds back down to 12) and assume the WARP is really 1.75 (which BP rounds up to 1.8), you still only get a run conversion factor of 7.1, which is still too low.

But that's just one season, so I picked 1978 and did Morgan and Lopes to see if the theory holds up for two guys playing the same position at around the same time.

1978.......................Morgan..........Lopes
BRARP+FRAR+PRAR.......43...............84
BRAR+FRAR+PRAR........41...............81
WARP1........................4.5..............8.9
RunConvFact #1..........9.6..............9.4
RunConvFact #2..........9.1..............9.1

A few more:

1956.......................Mays............Snider
BRARP+FRAR+PRAR.......90...............88
BRAR+FRAR+PRAR.......100...............97
WARP1......................11.0.............10.6
RunConvFact #1..........8.2..............8.3
RunConvFact #2..........9.1..............9.2

1956.......................Musial...........Snider
BRARP+FRAR+PRAR.......69...............88
BRAR+FRAR+PRAR........79...............97
WARP1.......................8.6.............10.6
RunConvFact #1..........8.0..............8.3
RunConvFact #2..........9.2..............9.2

1943.......................Musial...........Galan
BRARP+FRAR+PRAR......114...............80
BRAR+FRAR+PRAR.......124...............85
WARP1......................13.4..............9.2
RunConvFact #1..........8.5..............8.7
RunConvFact #2..........9.3..............9.2

1943.......................Vaughn...........Galan
BRARP+FRAR+PRAR.......81...............80
BRAR+FRAR+PRAR........81...............85
WARP1.......................8.8..............9.2
RunConvFact #1..........9.2..............8.7
RunConvFact #2..........9.2..............9.2

1939.......................Dimagg...........Williams
BRARP+FRAR+PRAR.......108...............85
BRAR+FRAR+PRAR........114...............95
WARP1.......................12.2..............10.2
RunConvFact #1..........8.9...............8.3
RunConvFact #2..........9.3...............9.3

2003.......................Pujols...........Rolen
BRARP+FRAR+PRAR.......101...............77
BRAR+FRAR+PRAR........112...............77
WARP1.......................11.9..............8.4
RunConvFact #1..........8.5...............9.2
RunConvFact #2..........9.4...............9.2

1898.......................Wallace...........Long
BRARP+FRAR+PRAR........78...............88
BRAR+FRAR+PRAR.........80...............87
WARP1........................8.8..............9.5
RunConvFact #1...........8.9..............9.3
RunConvFact #2...........9.1..............9.2

There just seems to be a lot more consistency using BRAR.

Note, I can't find ANY consistency when testing pitchers. Their run conversion factors often don't come close to 9.0 and rarely do they match the competitive pitcher. For instance, on the 1971 Mets, Seaver has a run converter of 9.8 under formula #1 and 9.6 under formula #2. Ryan, on the same team, has a run converter of 14.4 under formula #1 and 12.5 under formula #2.

You guys are smarter than me about BP and WARP, so maybe I haven't added anything...but a few extra thousand characters won't kill anyone. :)
Page 3 of 3 pages  < 1 2 3

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
1k5v3L
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.2922 seconds
49 querie(s) executed