How are "entire career" ratings calculated? Something seems off.

uruguru · 07-13-2024, 01:44 PM

Sorry but I've been crawling down this rabbit hole of fielding ratings and noticed something very peculiar about 'entire career'.

Hank Aaron played 27 games in 2B in 1955 and a smattering of games (23) at 2B & 3B over the rest of his career..

Ernie Banks played 1125 games at SS (a full-time starter) and 58 games at 3B from 1953 to 1961. He played a smattering of games (11) at 3B later in his career.

You can see their OOTP career fielding ratings in the attached image. Not only is Aaron rated higher at 2B than Banks is at SS, but his underlying infield ratings are vastly superior to Banks' even though "single season" fielding ratings show Banks to be an above-average defensive SS in his career.

Something must be off about how career fielding ratings are calculated.

Once again, I want to apologize to devs for being so anal-retentive about stuff like this. I know it's annoying.

How are these ratings calculated? Is the fact that Banks played so many games at 1B somehow dragging down his career infield ratings? And Aaron only played 1B at the end of his career, so his ratings at 2B are relatively unaffected?

If career ratings are averages of individual seasons (if... I don't know), maybe using root mean squares would help keep these late-career stints at 1B from dragging down career IF numbers at 2B/3B/SS.

uruguru · 07-13-2024, 03:35 PM

Well, I'm still lost. Bank's "official" OOTP career infield ratings (range-error-arm-dp) are:

77-98-71-69. This equals a bad (red) 5 rating at SS on the 1-100 scale. Not just bad, but basically unplayable at SS.

But if you take the average of each of Bank's infield ratings for each season, you get:

86-130-99-83. This results in a poor (orange) 26 rating at SS.

The root mean square (rms) for those same values is:

99-135-106-92. This results in a mediocre (yellow) 34 rating at SS.

If you weight each season's infield ratings by his actual infield innings played for those seasons (which makes sense), the average becomes:

90-130-101-84. The poor (orange) rating improves slightly to 28 at SS.

If you weight the rms by innings played, you get:

102-136-108-93. Still mediocre (yellow) 36 rating at SS.

All of these are poor ratings at SS for someone who played regularly at the position, but I guess it's understandable since his career ratings would be dragged down by his many years at 1B.

But even so, all of them are vastly better defensively than the OOTP calculated career ratings for Banks, so my conclusion is that I am way off the mark in deciphering how they are calculated.

Garlon · 07-13-2024, 03:46 PM

For the career defensive ratings in the file we have range/error/arm/turn DP for a combined 2B/3B/SS and another for 1B.

For Banks, his primary infield position based on innings played is slightly more at 1B than SS + 3B combined, so the game sees him as 1B in this instance.

This is a limitation of the fact that the game only has one set of infield component defensive ratings, so when it makes those ratings, it can either pick 1B or the 2B/3B/SS version. Banks is really the only major player affected by this because it is very rare that a player is none as much for playing 2B/3B/SS but yet has more defensive innings at 1B. If you use 3yr fielding for Banks, then he will be rated properly at SS during his SS seasons.

Aaron is also seen as a 1B as an infielder because he has far more innings at 1B than 2B+3B combined. In this instance though his range/error/arm/turn DP even at 1B is enough to play some 2B.

A player needs much higher component ratings to play SS than 2B. A player needs a much better arm to play 3B than 2B.

uruguru · 07-13-2024, 04:05 PM

Quote:

Originally Posted by Garlon

For Banks, his primary infield position based on innings played is slightly more at 1B than SS + 3B combined, so the game sees him as 1B in this instance.

So there's basically a hard check as to whether he's a 1B or a 2B/3B/SS? ok, that makes sense. Kind of harsh, but it's at least an understandable rationale.

But if that's the case, when you set fielding ratings to "entire career" then it should probably not assign fielding experience to Banks for the SS position.

Because that's what it does. When I started a test sim in 1962 with career fielding stats, I zero'd out Banks' SS experience because I didn't want the AI to try playing him at SS. And it didn't.

But when the recalc hit for 1963, it gave him 200 fielding experience at SS again (despite the poor ratings). That's when I started investigating him in particular.

I guess my issue with this is that, in the pre-DH era, many many players gravitated from defensive positions to 1B at the end of their careers. Banks, Aaron, Killebrew, Mantle, Stargell, McCovey, Rose, Bench, Cedeno, Staub, etc.

1B was the DH position before there was a DH. But players in the DH era won't get the same positional downgrade to their career fielding ratings as historical players before it.

To be honest, if there is a 1B vs 2B/3B/SS decision to be made for a player's career, I'd suggest that 1B innings at age 35 and up are just ignored for that particular decision. In the modern era, those guys would go to DH anyway.

Garlon · 07-13-2024, 04:17 PM

For Banks though, try the 1959 season with career fielding and see how what his ratings are at SS. They should be better since he was a SS in that season. The game I think will still give him experience at 1B and SS, but it will use the SS career ratings. This at least is how it used to work.

OutS|der · 07-13-2024, 04:27 PM

Feilding ratings need a massive overhaul or at the very least give us the ability to create ratings based on stats.

uruguru · 07-13-2024, 04:44 PM

Quote:

Originally Posted by Garlon

For Banks though, try the 1959 season with career fielding and see how what his ratings are at SS. They should be better since he was a SS in that season. The game I think will still give him experience at 1B and SS, but it will use the SS career ratings. This at least is how it used to work.

Thanks for the suggestion. I tried 1959 but the career fielding ratings were the same. And frankly it would feel like a bug if career ratings changed based on the season you were in.

Rain King · 07-13-2024, 09:00 PM

Quote:

Originally Posted by OutS|der

Feilding ratings need a massive overhaul or at the very least give us the ability to create ratings based on stats.

Fielding ratings literally just did get a massive overhaul. This just sounds like a (somewhat) edge case bug from that process.

uruguru · 07-13-2024, 09:48 PM

Quote:

Originally Posted by Rain King

Fielding ratings literally just did get a massive overhaul. This just sounds like a (somewhat) edge case bug from that process.

Yeah, based on what Garlon said it sure sounds like Ernie Banks is an edge case. It's like they have a pretty straightforward algorithm that works 98% of the time, but unfortunately that 2% where it doesn't work well includes a Hall of Famer, lol.

jcard · 07-14-2024, 12:45 AM

Interesting discussion. A couple of related comments—

1) Regarding 2B/SS/3B position group: In the past I have sometimes wondered if a player like Manny Machado with elite defense at one spot but had both the experience at and capability for a more difficult position was “punished” a little bit in terms of his ability at his primary position. Given that there is one set of component ratings shared by all position ratings formulae and that the player had real experience at a more difficult position, is it necessary to decrease his component ratings to prevent the AI from making the more difficult position his primary spot (with the simultaneous unintended effect of lowering his rating at his primary position).

2. I do wish that the “Turn DP” component rating was instead a “DP Pivot” rating that primarily applied to 2B defense. This would allow 2B defense to be more nuanced than just basically the SS rating against a lower relative baseline.

uruguru · 07-14-2024, 06:08 AM

Quote:

Originally Posted by jcard

I have sometimes wondered if a player like Manny Machado with elite defense at one spot but had both the experience at and capability for a more difficult position was “punished” a little bit in terms of his ability at his primary position. Given that there is one set of component ratings shared by all position

In a similar fashion, Banks is sort of inadvertantly being "punished" in his career defensive ratings for being a good hitter.

If he were just an average hitter at best, the Cubs would have not kept trotting him out at 1B for his bat and he would have retired earlier, not having racked up so many old-man innings at 1B that the OOTP career algorithm sees him as a 1B.

Garlon · 07-14-2024, 07:23 AM

For infielders the game wants SS to have the highest ratings for the range, error, arm, and turn dp ratings.

If you have a player with a great arm but limited range and turn dp, they are a good candidate for 3B.

If you have a player with decent range and decent turn dp but a weaker arm, they are a good candidate for 2B.

If you have a player with limited range and arm, they are a good candidate for 1B if they can provide offense.

07-13-2024, 01:44 PM	#1
uruguru All Star Starter Join Date: May 2022 Posts: 1,300	How are "entire career" ratings calculated? Something seems off. Sorry but I've been crawling down this rabbit hole of fielding ratings and noticed something very peculiar about 'entire career'. Hank Aaron played 27 games in 2B in 1955 and a smattering of games (23) at 2B & 3B over the rest of his career.. Ernie Banks played 1125 games at SS (a full-time starter) and 58 games at 3B from 1953 to 1961. He played a smattering of games (11) at 3B later in his career. You can see their OOTP career fielding ratings in the attached image. Not only is Aaron rated higher at 2B than Banks is at SS, but his underlying infield ratings are vastly superior to Banks' even though "single season" fielding ratings show Banks to be an above-average defensive SS in his career. Something must be off about how career fielding ratings are calculated. Once again, I want to apologize to devs for being so anal-retentive about stuff like this. I know it's annoying. How are these ratings calculated? Is the fact that Banks played so many games at 1B somehow dragging down his career infield ratings? And Aaron only played 1B at the end of his career, so his ratings at 2B are relatively unaffected? If career ratings are averages of individual seasons (if... I don't know), maybe using root mean squares would help keep these late-career stints at 1B from dragging down career IF numbers at 2B/3B/SS. Attached Images Last edited by uruguru; 07-14-2024 at 06:09 AM.

07-13-2024, 03:35 PM	#2
uruguru All Star Starter Join Date: May 2022 Posts: 1,300	Well, I'm still lost. Bank's "official" OOTP career infield ratings (range-error-arm-dp) are: 77-98-71-69. This equals a bad (red) 5 rating at SS on the 1-100 scale. Not just bad, but basically unplayable at SS. But if you take the average of each of Bank's infield ratings for each season, you get: 86-130-99-83. This results in a poor (orange) 26 rating at SS. The root mean square (rms) for those same values is: 99-135-106-92. This results in a mediocre (yellow) 34 rating at SS. If you weight each season's infield ratings by his actual infield innings played for those seasons (which makes sense), the average becomes: 90-130-101-84. The poor (orange) rating improves slightly to 28 at SS. If you weight the rms by innings played, you get: 102-136-108-93. Still mediocre (yellow) 36 rating at SS. All of these are poor ratings at SS for someone who played regularly at the position, but I guess it's understandable since his career ratings would be dragged down by his many years at 1B. But even so, all of them are vastly better defensively than the OOTP calculated career ratings for Banks, so my conclusion is that I am way off the mark in deciphering how they are calculated.

07-13-2024, 03:46 PM	#3
Garlon Hall Of Famer Join Date: Jun 2004 Posts: 4,271	For the career defensive ratings in the file we have range/error/arm/turn DP for a combined 2B/3B/SS and another for 1B. For Banks, his primary infield position based on innings played is slightly more at 1B than SS + 3B combined, so the game sees him as 1B in this instance. This is a limitation of the fact that the game only has one set of infield component defensive ratings, so when it makes those ratings, it can either pick 1B or the 2B/3B/SS version. Banks is really the only major player affected by this because it is very rare that a player is none as much for playing 2B/3B/SS but yet has more defensive innings at 1B. If you use 3yr fielding for Banks, then he will be rated properly at SS during his SS seasons. Aaron is also seen as a 1B as an infielder because he has far more innings at 1B than 2B+3B combined. In this instance though his range/error/arm/turn DP even at 1B is enough to play some 2B. A player needs much higher component ratings to play SS than 2B. A player needs a much better arm to play 3B than 2B.

07-13-2024, 04:17 PM	#5
Garlon Hall Of Famer Join Date: Jun 2004 Posts: 4,271	For Banks though, try the 1959 season with career fielding and see how what his ratings are at SS. They should be better since he was a SS in that season. The game I think will still give him experience at 1B and SS, but it will use the SS career ratings. This at least is how it used to work.

07-13-2024, 04:27 PM	#6
OutS\|der Hall Of Famer Join Date: Apr 2010 Location: In A Van Down By The River Posts: 2,721 Infractions: 0/1 (1)	Feilding ratings need a massive overhaul or at the very least give us the ability to create ratings based on stats.

07-14-2024, 12:45 AM	#10
jcard All Star Reserve Join Date: Oct 2015 Posts: 603 Infractions: 0/1 (1)	Interesting discussion. A couple of related comments— 1) Regarding 2B/SS/3B position group: In the past I have sometimes wondered if a player like Manny Machado with elite defense at one spot but had both the experience at and capability for a more difficult position was “punished” a little bit in terms of his ability at his primary position. Given that there is one set of component ratings shared by all position ratings formulae and that the player had real experience at a more difficult position, is it necessary to decrease his component ratings to prevent the AI from making the more difficult position his primary spot (with the simultaneous unintended effect of lowering his rating at his primary position). 2. I do wish that the “Turn DP” component rating was instead a “DP Pivot” rating that primarily applied to 2B defense. This would allow 2B defense to be more nuanced than just basically the SS rating against a lower relative baseline.

07-14-2024, 07:23 AM	#12
Garlon Hall Of Famer Join Date: Jun 2004 Posts: 4,271	For infielders the game wants SS to have the highest ratings for the range, error, arm, and turn dp ratings. If you have a player with a great arm but limited range and turn dp, they are a good candidate for 3B. If you have a player with decent range and decent turn dp but a weaker arm, they are a good candidate for 2B. If you have a player with limited range and arm, they are a good candidate for 1B if they can provide offense.