Fielding Bug

[swampdragon]

For a long time, I have been concerned that all of the 1919 Texas League fielders have been showing fielding ratings that are exactly average. (50 on a 20-80 scale). I tried to start a fresh 1901 game today, and in the latest patch this problem has spread to the major leagues. Every fielder in 1901 is either average or within a couple of points of it.

This should be an easy one to duplicate. If you can't, let me know and I'll try a fresh import.

[Spritze]

Honus Wagner 1901 49.5-51.5 for his various fielding components.
Overall players in 1901 ranged from 40.5-59 for their fielding components.
The average player in 1901 averaged about 50 which is the average average 1871-2015.

Just how many slow white guys with tiny gloves do you expect to field like Willie Mays or Ozzie Smith? These ratings are based on all of baseball history as well as 1901. Things were a bit different back in the day. No such thing as good field no hit. No hit = no play.

Just sayin'.

[swampdragon]

There's no fielding difference between Honus Wagner and Bill Keister? No one back then believed that. I certainly don't. OOTP has a system that depends on fielding ability to differentiate teams and get pitchers, who didn't have many walks or strikeouts back then, generally to the ability level they showed over the years. It generally did a pretty good job of that until the last patch. 1901 Jimmy Collins at 3rd had a fielding rating of 100 pre-patch. Now it's 49 on a 100 point scale. Something is being done differently. If it's a bug, it needs to be fixed. If it's a design decision, it needs to be rethought. And the minors need to be done the same way the majors were done pre-patch. This is seriously messed up.

[Spritze]

Originally Posted by swampdragon

There's no fielding difference between Honus Wagner and Bill Keister? No one back then believed that. I certainly don't. OOTP has a system that depends on fielding ability to differentiate teams and get pitchers, who didn't have many walks or strikeouts back then, generally to the ability level they showed over the years. It generally did a pretty good job of that until the last patch. 1901 Jimmy Collins at 3rd had a fielding rating of 100 pre-patch. Now it's 49 on a 100 point scale. Something is being done differently. If it's a bug, it needs to be fixed. If it's a design decision, it needs to be rethought. And the minors need to be done the same way the majors were done pre-patch. This is seriously messed up.

Collins should be between 59 and 63 according to the data. If he was 100 that was as wrong as is the current 49. Seems like something changed in the program?

Bill Keister = 48.5
Wagner played 2B,3B,SS and OF in 1901 in almost equal measures --- I'm sure once he settles in at SS his fielding will improve?

[swampdragon]

Collins is at 49 on a 100 point scale. 50 on a 20-80 scale. Just like everybody else.

Fielding ratings have always been measured against the season being played, not baseball history. With DIPS, you will seriously screw up historical replays if you go that way.

[Spritze]

Originally Posted by swampdragon

Fielding ratings have always been measured against the season being played.

Still are. Each season averages around 50 for all the components, give or take a little bit. For all of history it averages 50 with no give or take.

Minor leagues are identical formulaically speaking with the exception of a players first year since there is no data until the season is played.

[swampdragon]

I just checked all of the starting third basemen for the American League in 1901.

20-80 point scale. Range 50, Error 50, Arm 60, DP 50. Total 50.
All eight of them. That's a bug, or it better be.

[swampdragon]

The 1901 game I started shortly after the release of the game:

20-80 point scale.

Collins 75/75/75/75 Total 85
McGraw 40/45/50/35 Total 35
Hartman 66/65/65/60 Total 65
Bradley 50/75/60/50 Total 55

Note the big differences in fielding ability and in the components. Now they are all 50/50/60/50. Total 50.

[Spritze]

I reported this bug to the Beta bug report so that whatever was done could be undone and be the undoing of the bug.

[swampdragon]

Originally Posted by swampdragon

I just checked all of the starting third basemen for the American League in 1901.

20-80 point scale. Range 50, Error 50, Arm 60, DP 50. Total 50.
All eight of them. That's a bug, or it better be.

While you're at it, the 1919-21 minor league starting 3rd basemen almost all have exactly the same rating. All but Babe Pinelli.?!?! This has been the case in the early minors since the beginning, while the problem is new to the majors. We need the major league prepatch formulas used for defense for both major and minor leagues.

[Spritze]

The same formulas are still used and have been for a few months. They were changed a bit when the minors were added. Catchers only though.

Something in game changed. Probably inadvertently.

[swampdragon]

As far as I can tell, this was not addressed in the latest patch. For historical replays, accurate ratings are critical to a believable and enjoyable experience. That includes fielding ratings. This game is unplayable until the problem is fixed.

[Spritze]

The last patch was for a few specific issues. Not this one nor any of the other historical issues posted in this forum.

Donut get discouraged. Attended to they will be.

[MightyVotto]

This seems to be happening in all seasons from around 1940 to 1953. It seems as if all fielding ratings are neutralized for those seasons. It may have started in the late 1930's, but I didn't notice it until 1940. Once I saw it in my quick historical replay, I made sample test leagues for each season to 1954 and saw the fielding ratings act the same way.

I'm using 1-year recalc with no development engine. Real lineups and transactions turned on. Prior to 1940, there are usually a wide range of player fielding ratings. Several blues, greens, yellows, etc. Beginning in 1940, the highest Outfield range is 71, and all the good fielders are given that. Each year until 1954, these ratings are not changing either. Dimaggio's range is 71 throughout the entire decade no matter what year is chosen.

Beginning in 1954, it seems, the ratings go back to being tied to true performance. There's a wide range again, and blue ratings are once again assigned.

Is this a bug or were the 40's/early 50's really this bland and boring for defense?

[Spritze]

Originally Posted by MightyVotto

This seems to be happening in all seasons from around 1940 to 1953. It seems as if all fielding ratings are neutralized for those seasons. It may have started in the late 1930's, but I didn't notice it until 1940. Once I saw it in my quick historical replay, I made sample test leagues for each season to 1954 and saw the fielding ratings act the same way.

I'm using 1-year recalc with no development engine. Real lineups and transactions turned on. Prior to 1940, there are usually a wide range of player fielding ratings. Several blues, greens, yellows, etc. Beginning in 1940, the highest Outfield range is 71, and all the good fielders are given that. Each year until 1954, these ratings are not changing either. Dimaggio's range is 71 throughout the entire decade no matter what year is chosen.

Beginning in 1954, it seems, the ratings go back to being tied to true performance. There's a wide range again, and blue ratings are once again assigned.

Is this a bug or were the 40's/early 50's really this bland and boring for defense?

1954 is the first year OF ratings are split into RF,CF,LF and I'll bet this is at least part of the issue. I'll see if I can create guesstimates of position played and see if this helps.

[Spritze]

Adjusted the affected OF'ers (now = RF, CF, LF) but this does not help explain the infielders of course.

[swampdragon]

Originally Posted by MightyVotto

This seems to be happening in all seasons from around 1940 to 1953. It seems as if all fielding ratings are neutralized for those seasons. It may have started in the late 1930's, but I didn't notice it until 1940. Once I saw it in my quick historical replay, I made sample test leagues for each season to 1954 and saw the fielding ratings act the same way.

I'm using 1-year recalc with no development engine. Real lineups and transactions turned on. Prior to 1940, there are usually a wide range of player fielding ratings. Several blues, greens, yellows, etc. Beginning in 1940, the highest Outfield range is 71, and all the good fielders are given that. Each year until 1954, these ratings are not changing either. Dimaggio's range is 71 throughout the entire decade no matter what year is chosen.

Beginning in 1954, it seems, the ratings go back to being tied to true performance. There's a wide range again, and blue ratings are once again assigned.

Is this a bug or were the 40's/early 50's really this bland and boring for defense?

If you start a 1901 replay, you'll see the same thing in 1901. Just for grins, I started up a 1919 replay and got the same thing for both majors and minors. This may be an effect of a recent patch, and the only reason you didn't see it earlier was that you were continuing a game started before the patch and wouldn't be affected until you started another season.

Start a 1925 historical replay and see what the 3rd basemen look like. Spritze is right that you get varying results with different outfield positions. But if all of the starting 3rd basemen and shortstops have the same rating, it has to be a bug.

[Spritze]

Here is the ss 3b data for the 2 st. louis teams in 1925.

Not a big range in range. Should it be larger or are most fielders in general average? If they were not at least average-ish they would be minor leaguers?

Third basemen are only compared to other third basemen in MLB in 1925. etc. range is 43-56, for SS 46-53 MLB average is 50.

This example has the best fielders 12% better than average and the worst 14% below average.

Generally no one gets rated below 20 nor higher than 80. Hey! that matches OOTP's 20-80 default scale!

[MightyVotto]

Admittedly, I've never understood quite how defensive metrics actually worked, but I've always had fun watching how a past player's zone rating and efficiency ratings in the game appeared since we never really had those metrics for them. Going back several versions of the game and during the early months of this version, the best players at each position always stood out on a 1-100 scale as being given the high end ratings. The 1927 Yankees, for example, had a great defense with blue ratings around the infield. It was one of the things that made it clear why this team's pitching was so good since the pitchers on their own aren't that special. Now, when I begin a 1927 season replay game with no development on, the ratings look completely different. The best infielders are only given a 68 range no matter what. The best outfielders are only given a 71.

Is it supposed to look like this based on a change to how the game handles defense now? If so, that's cool. I just don't get it since it is very different from what it used to display for single season replay. Were the results we were getting before for these metrics completely insane? And now they'll be more realistic? Not that they seemed too off before.

[Spritze]

Originally Posted by MightyVotto

Is it supposed to look like this based on a change to how the game handles defense now? No If so, that's cool. I just don't get it since it is very different from what it used to display for single season replay. Were the results we were getting before for these metrics completely insane? Yes And now they'll be more realistic? Yes Not that they seemed too off before.

Something probably changed in game I think. The raw ratings are still pretty much the same as they ever were.

It might just be a display issue since the actual ratings have not changed.

Just tested this and it is just a display issue. The underlying ratings are correct and play as such.