Fielding Ratings

[Thomas359]

Love OOTP 9, with one issue - the fielding stats. I created a league with the best alltime players in their best years for each team (Cardinals with Medwick 1937, Musial 1948, etc...). The batting ratings all make sense, but a lot of the fielding ratings seem completely wrong. For instance, Mike Piazza has a catcher arm and defensive ability rating of 39, while Johnny Bench is rated 34 in both categories. Reggie Jackson, a notoriously bad fielder, has an outfielder range rating of 45, and an outfield error rating of 65 - Vlad Guerrero is rated 28 and 35 in these 2 categories. Brooks Robinsonhas an infield range of 60 and an infield error rating of 63 - Darrell Evans, mediocre at best, has a ratings of 63 and 67. Scott Rolen and Mike Schmidt, 2 of the best fielding 3b in real life, are rated well worse. ozzie Smith and Luis Aparicio are rated well below Arky Vaughan in both infield range and ability to avoid errors. I'm sure I've bored everyone already, but the list goes on and on. Any idea why these strange things happen, and/or more importantly how to fix them without manually redoing them all?

[Giants44]

hopefully someone can corroborate this, but I believe the problem is the lahman databases that is used to create the historical rosters does not contain defensive statistics, thus the defensive ratings are random.

I do think the gambo database fixes this, but is not built into the game.

again need someone with more knowledge to reply - but I believe you can point the game to the gambo database and it would solve the problem.

me, I am such a control freak that when I play historical, I tend to go and and do a lot of editing to get things right.

[StyxNCa]

Originally Posted by Giants44

hopefully someone can corroborate this, but I believe the problem is the lahman databases that is used to create the historical rosters does not contain defensive statistics, thus the defensive ratings are random.

I do think the gambo database fixes this, but is not built into the game.

again need someone with more knowledge to reply - but I believe you can point the game to the gambo database and it would solve the problem.

me, I am such a control freak that when I play historical, I tend to go and and do a lot of editing to get things right.

Lahman has fielding, however something I have discovered is since fielding stats don't include outs played at position, which OOTP translates into innings played, it is defaulted by OOTP to 9 innings per game. This really distorts players who played a lot of defensive games but few innings and TC/G (game in OOTP terms). Ken Berry comes to mind, one of the best CF's, especially his range. I experimented and found when I put in his outs played, his ratings came up dramatically to a level I would have expected to see. I have been adding outs played to everyone in the Lahman database based on information found in retrosheet and baseball prospectus which turns the defensive numbers into TC/9 innings rather than TC/G.

Thirdbase is a position, however, that always seems to produce low ratings. I guess part of the equation in determining defensive ratings is the position. Shortstops tend to get higher ratings than any other infield position. I have also noticed how Johnny Bench gets rather average ratings. I am not sure how the game determines arm since there is no stat in Lahman for SB/CS for catchers or pitchers. It probably uses just the assist numbers, and for a player like Bench, those numbers will be lower than most catchers because only imbeciles would try to run on him so the opportunities for him are low.

I am like you in that I like historical things so accurate it takes me quite a while to set up a season because I go through and edit a lot of the ratings and stats and things like uniform numbers. Since I found a site that has pitchers GB%, that gives me one more thing to edit. How's that for picky.

[Giants44]

thanks for the correction - I was under the assumption the DB had no fielding, but I stand corrected.

I did go look at my historical league tonight after the patch and was amazed at how much better the def looked (not from the first build, but from previous versions)

They guys I expected to be top rated, were top rated - I specifically remember this not being the case before. I do think Bench was one of the guys that was not right and I edited him after reading this.

[Thomas359]

Thanks for the responses - seems like an issue OOTP should address at some point. Baseball-reference.com has defensive range and fielding % ratings by player for each year, as well as the league average for the position for that year. For catchers, also has SB against and CS for each player. Given this information is available, seems like OOTP could find way to incorporate the data into Lahman (or some other database) to be used in their game.

In the meantime, I will simply have to use this data to manually update players. Might be annoying, might be fun and educational!!

As I do that, I can make sure the relative ratings for players are at least somewhat relevant. Anyone have thoughts on what the range and ability to avoid errors rating should be set at for the typical player - i.e. one who meets the league average for his position for the year - so I can set the base case? Also any thoughts on outfielder arm ratings, which I believe will have to be adjusted subjectively. (Got to be good for some arguments!) I know Parker, Guerrero, Clemente, Mays, even Reggie Jackson, had great arms, but any thoughts on the best (and worst) outfiled arms for the better players in history? Also using Negro League players from the Sprtiz DB, so thoughts on any of the great players inthat league as to fielding/strength of arm?

[swat-spas]

I haven't actually tested the new version but the old 07/8 versions had the following breakpoints and values per rate based on positions, which results in things like the variation between 3b and SS raw fielding ratings. I generated 'averages' by taking the average ratings of a randomly generated league and adjusting it down to a 100 average per position. Those should be seen as approximations, nothing more. These ratings are from the player editor and use a 250 scale. edit: I checked and its usually a 110 average, but that's still a decent baseline.


C
Min Arm: 10, value per point >min 1.7, average ~93
Min Rng: 10, .78, 100
Min E: 1, .12, 50

1b:
Arm: 1, .18, 40
Rng: 1, 2.18, 50
E: 1, 1.08, 66
DP: 1, .18, 33

2b
A: 40, .14, 70
R: 40, 1.72, 115
E: 50, .66, 110
DP: 60, .72, 115

3b
A: 70, .59, 150
R: 40, 1.72, 105
E: 40, .63, 105
DP: 20, .28, 85

SS
A: 60, .16, 115
R: 70, 1.88, 140
E: 50, .6, 125
DP: 60, .64, 125

CF
A: 40, .14, 100
R: 90, 3.4, 140
E: 40, .28, 115

LF
A: 40, .36, 100
R: 40, 2.68, 100
E: 40, .54, 105

RF
A: 40, .58, 100
R: 55, 2.69, 100
E: 40, .58, 105

I myself go in and edit the ratings, regardless of things like 'range factor vs era' available at B-R, which doesn't actually tell us much about the player's defensive abilities (but is better than nothing)---I do use the F% vs era. I usually end up using an adjusted average of DWS and FRAR (baseball prospectus) to get an overall rating and then go from there. I'm not saying these are the best fielding ratings, but they seem to be the only ones available across history and they pass the 'feel' test by using both.

SB/CS for catchers is available only after '57 (play by play data is needed). For players before that all we have is assists and assumptions. Rodriguez and Bench had the best arms when I did my runs with actual catcher data. Piazza was the worst. Keep in mind that often LOWER assist totals are paradoxically better therein.

[Thomas359]

Thanks - exactly what I was looking for.