Recalc question

[italyprof]

Originally Posted by thehef

I just ran a test...

OOTP 23
start year 1965; ran thru the 1991 season
no historical minors
import complete history
real stats
potential = remaining yrs
adj/wkn = 150-40-30-10
hist txns on; hist lineups & hist injuries off
no coaching or owner goals
injuries very low; no long term injuries
position player fatigue = high
no personalites or morale
base fielding on current season
pitcher stamina based on curr season
1-yr recalc
everything else is defaults (except for insignificant things like WPA graphs and saving game recaps...)

Carl Yastrzemski
year, games at 3B:
1973, 71
1974, 23
1975, 143
1976, 138
1978, 116
1979, 58
1980, 6
1981, 37
1982, 12
1983, 12

IRL, Yaz played 2 games at 3B in '64, 31 in '73

In every year above there were other options to play 3B, just like IRL. For example, in 1975, Rico Petrocelli (whom OOTP played at 2B 38 times, despite Rico not playing 2B at all IRL in '75), Juan Beniquez, Denny Doyle, Bob Heise, and Dick McAuliffe all played 3B IRL in 1975 and were available to the Red Sox most or all of that season in OOTP.

Brian Downing
year, games at C & 3B after 1981:
1982, 3, 2
1983, 0, 0
1984, 95, 1
1985, 88, 1
1986, 91, 0
1987, 13, 6
1988, 87, 2
1989, 98, 0
1990, 42, 6
1991, 87, 0

IRL, Downing did not catch a single game after 37 games behind the dish in 1981, and he only played 3B for 8 games in 1973... Obviously, there would've been other options - just as there were IRL - to play catcher on his teams during those years.

A couple others, without going into as many details:
Bill Russell plays handfuls of games at 2B every year from 1973 thru 1978, despite IRL not playing 2B at all between 1973 to 1983.
Tommy Davis plays significant numbers of games at 3B in '74, '75, '76 despite not playing the position at all IRL in those years...

Again, this is with 1-year recalc and fielding ratings based upon current season... There's a reason there's a bug report for this issue, and the above is it (And I know for a fact that this issue goes back to at least version 21.)

Yaz at 3B?!!! NO!!!

[thehef]

Originally Posted by italyprof

Yaz at 3B?!!! NO!!!

'Tis true! You can read about his real-life '73 escapades here: https://sabr.org/gamesproj/game/augu...at-third-base/

But OOTP seems to take it a bit overboard

[David Watts]

Okay, I'm still going to struggle with guys like Ozzie Smith playing 2B, but I decided to use Spring Training for the first time ever. I'm about 8 games into a season and I'm not seeing as many red rated fielders. Haven't seen a single non rated player in the field. Fingers crossed that this is a way to avoid seeing so many crap rated fielders starting games in an attempt to become Tony Phillips.

[David Watts]

I'm starting to wrap my head around this a little better. Watching Adam Jones right now and his progress to playing RF is rather interesting. This might not be bad. But, I still don't want to see Brooks Robinson playing 1B or SS.

[thehef]

Originally Posted by David Watts

I'm starting to wrap my head around this a little better. Watching Adam Jones right now and his progress to playing RF is rather interesting. This might not be bad. But, I still don't want to see Brooks Robinson playing 1B or SS.

The key to all of this is getting the bug fixed so base fielding ratings on current year works as described. I've high hopes that it will be fixed for v24!

[David Watts]

Originally Posted by thehef

The key to all of this is getting the bug fixed so base fielding ratings on current year works as described. I've high hopes that it will be fixed for v24!

Any chance you can display Yaz's ratings at 3B over the years? Does he always maintain a ratings to play the position year in and year out, even for years he didn't play there in real life, or does he lose his 3B rating, only to regain it through starting as a red1? In other words, is he relearning the position?

I started 1962, 5 year recalc, development off. defensive rating based on 3 years. I'm not using real transactions, because no matter what I do, I can't get the game to maintain full active rosters. I end up with teams playing 1, 2 or 3 players short on a regular basis. Not sure what I could be doing wrong, but it happens often enough that I just gave up on real transactions. In this game, the Yankee are playing Tony Kubek at 2B with a red rating that is slowly climbing, 25 right now. Bobby Richardson is rated a yellow 50 at 2b, but he's riding the bench. The game does move Kubek to SS, Tresh from SS to LF and subs Richardson in at 2B at the end of most games. The strange thing is Kubek and Richardson are pretty much identical in terms of offensive prowess, so I'm not sure why the game would choose to start a player rated so poorly at 2B, when it has a better defensive option available.

Okay, after 5 games played at 2B, Kubek is a red 25/65. The 25/65 is in deep maroon and then the bar turns bright red very briefly and end in a off green shade. So does this mean he has the potential to turn into a green 65 at 2B? I don't usually even pay attention to ratings, so a lot of this stuff is foreign to me.

Thing is, when setting up a straight historical, one can't run Spring Training as the season starts the day after creations.

[David Watts]

Kubek has now started 9 games at 2B. His ratings are no longer red. He is an orange 35/65 with the light shade of green extension deal.

[Garlon]

An average defensive SS will have the Range/Arm/Error/Turn DP ratings to move to 3B or 2B.

An average 2B will not have the Range/Arm/Error/Turn DP to move to SS and lay at an average defensive level. Even a great defensive 2Bmen will not be able to move to SS and perform decently. The same goes for 3B, because a 2Bmen will not have the Arm to play 3B.

An average 3Bmen will not have the Range required for SS or 2B, but they will have the Arm. 3Bmen generally do not have enough Turn DP to be good candidates to move to 2B or SS.

Any infielder can easily learn 1B.

For a given infielder, they are only rated at a single infield position since they only have a single set of component defensive ratings. If there was a player who had 1000 innings at SS and 250 innings at 3B, their ratings will be that of a SS, but they will be given eligibility at 3B. If they played 1000 innings at 3B and 250 at SS their ratings will be that of a 3Bmen and have eligibility at SS.

For example, say that a player has 1.025 Range as a SS but has eligibility at 3B. That player is going to have a great infield Range rating because the baseline Range for SS is much higher than the other positions. If that player has 1.025 Range at 3B but is eligible to play SS they will come in with a much lower Range rating because the baseline Range at 3B is much lower than SS.

There are not that many true SS in the league at any given time. There is generally 1 starting SS per team, and then sometimes you have a utility player who played more games at SS than the other infield positions. Due to scarcity of true SS, there are not many opportunities to move them over to 2B or 3B.

[David Watts]

Originally Posted by Garlon

An average defensive SS will have the Range/Arm/Error/Turn DP ratings to move to 3B or 2B.

An average 2B will not have the Range/Arm/Error/Turn DP to move to SS and lay at an average defensive level. Even a great defensive 2Bmen will not be able to move to SS and perform decently. The same goes for 3B, because a 2Bmen will not have the Arm to play 3B.

An average 3Bmen will not have the Range required for SS or 2B, but they will have the Arm. 3Bmen generally do not have enough Turn DP to be good candidates to move to 2B or SS.

Any infielder can easily learn 1B.

For a given infielder, they are only rated at a single infield position since they only have a single set of component defensive ratings. If there was a player who had 1000 innings at SS and 250 innings at 3B, their ratings will be that of a SS, but they will be given eligibility at 3B. If they played 1000 innings at 3B and 250 at SS their ratings will be that of a 3Bmen and have eligibility at SS.

For example, say that a player has 1.025 Range as a SS but has eligibility at 3B. That player is going to have a great infield Range rating because the baseline Range for SS is much higher than the other positions. If that player has 1.025 Range at 3B but is eligible to play SS they will come in with a much lower Range rating because the baseline Range at 3B is much lower than SS.

There are not that many true SS in the league at any given time. There is generally 1 starting SS per team, and then sometimes you have a utility player who played more games at SS than the other infield positions. Due to scarcity of true SS, there are not many opportunities to move them over to 2B or 3B.

Thank you. But, why is the computer moving Kubek to 2B to learn a position when Bobby Richardson rides the bench? I understand that with 5 year recalc, Richardson's .302 hitting 1962 gets brought down due to the other 4 years factored in, but Kubek is not a great hitter himself. They are actually pretty similar at the plate. The game obviously knows Kubek is not highly rated at 2B, because late in game it's making defensive switches moving Kubek to SS, Tresh to LF and inserting Richardson into the lineup at 2B.

I'm warming up to this, as it's going to make roster depth less of an issue, but I still think the AI should try to field the best defensive team possible when it can.

[David Watts]

Okay the very next game I play features the 1962 Cubs. Ron Santo is playing 1B with a red25. Ernie Banks is at SS with a red 25 and Andre Rodgers is at 3B with a red 25. All of them could just switch to their regular positions and the Cubs defense improves three fold. Maybe this just needs to be toned down a little.

[Garlon]

The game is probably trying to field the best offensive team given the roster and while doing so is sacrificing some defense.

[thehef]

Originally Posted by David Watts

Any chance you can display Yaz's ratings at 3B over the years? Does he always maintain a ratings to play the position year in and year out, even for years he didn't play there in real life, or does he lose his 3B rating, only to regain it through starting as a red1? In other words, is he relearning the position?

Unless there's something I'm missing, I'd need to re-run the test and grab his ratings each year, right? Because I don't think OOTP keeps prior year ratings. Or does it? (Assuming it doesn't, sure, I'll re-run and report back.)

[thehef]

Originally Posted by Garlon

The game is probably trying to field the best offensive team given the roster and while doing so is sacrificing some defense.

Re the above, and the discussion on learning a new position, you're not referring to the Yaz bug, correct? Because learning a new position - let alone putting a player in a position that he did not play in real life - should not occur with ideal replay settings (1-year recalc, development off, etc.). Correct?

[David Watts]

Originally Posted by Garlon

The game is probably trying to field the best offensive team given the roster and while doing so is sacrificing some defense.

Probably, but I wonder how the offense improves by playing Banks with bad ratings at SS, Santo at 1B with bad ratings and Rodgers at 3B with bad ratings, when all the AI has to do is play them at their actual positions?

[David Watts]

Originally Posted by thehef

Re the above, and the discussion on learning a new position, you're not referring to the Yaz bug, correct? Because learning a new position - let alone putting a player in a position that he did not play in real life - should not occur with ideal replay settings (1-year recalc, development off, etc.). Correct?

I just posted several examples of guys being played out of position/learning a position. Not real transactions, as I won't play a version of the game in which the AI doesn't maintain full active rosters. But, I have development off. Tony Kubek in the above example is starting at 2B over Bobby Richardson. Kubek began with zero ratings and now has a orange 35. So, yes, with development off guys are learing new positions. I also posted in regard to Ron Santo playing 1B.

[Garlon]

For Yastrzemski, he did play 272.1 innings at 3B in 1973, but he also played 896.1 innings at 1B that season. The game is going to give him ratings as a 1Bmen. He will not have a good enough arm rating to play 3B. The game should not attempt to play him at 3B because he will be a defensive liability there.

I suspect that the game is trying to move Doug Griffin and Denny Doyle off 2B between 1973 and 1977 to generate more offense.

It is not so much that anything is broken regarding fielding though. This has more to do with how the game values defense. I think that players should learn new positions more slowly so that there is a greater defensive cost to making those types of moves. Then there is the matter of whether moving these players to a new position to get more offense outweighs their defensive performance. Maybe OOTP thinks it can win more games by moving these players around, but maybe it is actually an inefficient move?

[David Watts]

Might be nothing more than the AI doing so because it can. I think if this out of position thing is going to become the norm, historical leagues need to start earlier than the day before opening day, as they do now. Give us time to allow the game to run Spring Training. That way the AI can move guys around with wild abandon, but by the time the season starts most of the "never playeds" will have ratings established.

[thehef]

Originally Posted by Garlon

It is not so much that anything is broken regarding fielding though.

I'm not so sure. I understand what you are saying about Yaz in '73. That's not really what concerns me.

IMO, if you are playing with replay settings (1-year recalc, development off, and - most-importantly - base fielding ratings on current season), you should not see stuff like this:

Carl Yastrzemski
year, games at 3B:
1973, 71
1974, 23
1975, 143
1976, 138
1978, 116
1979, 58
1980, 6
1981, 37
1982, 12
1983, 12

When IRL Yaz played 2 games at 3B in '64, 31 in '73, and that's all. To me, that's a CLEAR indication that base fielding ratings on current season is broken.

[David Watts]

Originally Posted by thehef

I'm not so sure. I understand what you are saying about Yaz in '73. That's not really what concerns me.

IMO, if you are playing with replay settings (1-year recalc, development off, and - most-importantly - base fielding ratings on current season), you should not see stuff like this:

Carl Yastrzemski
year, games at 3B:
1973, 71
1974, 23
1975, 143
1976, 138
1978, 116
1979, 58
1980, 6
1981, 37
1982, 12
1983, 12

When IRL Yaz played 2 games at 3B in '64, 31 in '73, and that's all. To me, that's a CLEAR indication that base fielding ratings on current season is broken.

See this is where I think OOTP can't win for trying. If they fix what you're calling a bug, things will work perfectly for those playing using real transactions. If they fix the Yaz bug, lots of "what if" players will probably be screaming bloody murder. Even for the human manager it would be awfully tedious to have to scan real life player usage before each and every transaction. I'm more in favor of players losing ratings at positions based on OOTP use, than players losing ratings altogether.

As much as I'm calling into question some of this stuff, I am at least calmed down by Garlon saying when Kubek learns 2B and gathers in game experience, recalc isn't going to erase all that experience during the offseason with a heavy dose of reality. It's kind of a once Pandora's Box is open thing for me.

If you create 1974, Yaz only gets ratings for LF and 1B, using 1 year recalc, current season. Same setup for 1973 and Yaz gets rated to play LF, 1B and 3B. I simmed the 73 season and then simmed to 1974 preseason. Yaz is still rated to play LF, 1B and 3B. During the simmed 73 season he played in 3 games at 3B none of them starts. 151 games at 1B. No time in LF.

[Brad K]

Originally Posted by Garlon

An average 2B will not have the Range/Arm/Error/Turn DP to move to SS and lay at an average defensive level. Even a great defensive 2Bmen will not be able to move to SS and perform decently.

Most of these performances look decent or better to me. 2B as starting SS in my current save.

Rennie Stennett 1974-77 (Dave Cash and Willie Randolph at 2B)
Willie Randolph 1979, 85, 88-89 (Art Howe, Jose Lind, Johnny Ray at 2B)
Johnny Ray 1983-84, 86 (Willie Randolph and Tim Flannery at 2B)
Jose Lind 1987,90 (Vance Law and Willie Randolph at 2B)


Stennett ZR -1.0, +30.8. +14.8, -4.3
Randolph ZR -2.3, -12.3, -5.1, -5.8
Ray ZR +3.2, -3.5, -10.8
Lind ZR -2.4, -5.5