Defensive Ratings Overhaul

[teachermike]

I'd like to see an overhaul of the defensive ratings system. The current system works fine until you start trying to move players around the diamond, and specifically from catcher to any other position, or from the infield to the outfield. Since catchers often have very low infield defensive ratings, it's not always clear if they'll be able to make the successful transition to 1B; many slick-fielding, athletic infielders have terrible outfield defensive ratings, so making that transition (even to a low-intensity defensive position like LF) can be unsuccessful even for great infielders.

The current root (i.e. editable) defensive ratings, IF/OF Range, Error, and Arm, and Turn Double Plays, should all become derived ratings like the Contact rating, or be merged. The error ratings should become a single rating; it seems intuitive that a player's ability to catch and hold onto the ball should translate pretty well across positions, given an equal amount of experience at each position. Similarly, the three arm ratings (Catcher, Infield, Outfield) could be replaced by or derived from two universal arm ratings, an Arm Strength and Arm Accuracy. (Maybe these could even tie directly into pitchers' velocity/stuff and control ratings.) OF Range should be largely derived from speed and a dose of intelligence (picking the right route to the ball), and also be affected by a quickness rating; IF Range should be mostly dependent on quickness. Turn Double Plays might be some combination of quickness, intelligence, and arm ratings.

On a related note, it would be great to have an option to lock a player into a spot on a minor league lineup and/or depth chart, in the same way that you can disable AI promotion/demotion. I wanted to shift Mookie Betts to the outfield, so I changed his position to CF and put him in the starting CF role in the depth chart. I don't like micro-managing the minors, so I had some AI help there, and when I came back two weeks later, Mookie was the starting 2B again. This seems like it could be an easy change to make.

[MarkInCincy]

Going from memory here so;
Open player profile
Top right dropdown (Actions I think)
Game strategy
for hitters, bottom left section, usr Force start for position, for pitchers bottom right there is similar for locking role

NOTE: AI will start them but may also put it's preferred choice in depth playing every 2nd game

hope that helps, I use it for my minors a lot

[RchW]

Originally Posted by teachermike

The current root (i.e. editable) defensive ratings, IF/OF Range, Error, and Arm, and Turn Double Plays, should all become derived ratings like the Contact rating, or be merged. The error ratings should become a single rating; it seems intuitive that a player's ability to catch and hold onto the ball should translate pretty well across positions, given an equal amount of experience at each position.

Not sure I agree with the above. A lot of players have trouble catching LD at 3B and 1B even if they are accomplished middle infielders. Judging FB is different than GB and are definable IF/OF skills etc. As for errors remember that errors are judged by an official scorer so except for throwing errors they often have little to do with skill and more with circumstance. Even then a sure handed 3B may be an error prone 2B unable to make a DP. Many failed 3B fail because of throwing not fielding. Edwin Encarnacion comes to mind.

Originally Posted by teachermike

Similarly, the three arm ratings (Catcher, Infield, Outfield) could be replaced by or derived from two universal arm ratings, an Arm Strength and Arm Accuracy. (Maybe these could even tie directly into pitchers' velocity/stuff and control ratings.)

The throwing action at 3B (possibly SS) is radically different than the OF and C. Several years ago Jose Bautista took a week to get his arm ready for a move from RF to 3B to protect his arm. I'd say that 2B and 1B arms may be weaker than many OF but hard to compare with the different action.

Originally Posted by teachermike

OF Range should be largely derived from speed and a dose of intelligence (picking the right route to the ball), and also be affected by a quickness rating; IF Range should be mostly dependent on quickness. Turn Double Plays might be some combination of quickness, intelligence, and arm ratings.

Range is about getting to the ball even when it is on the edge of the normal space for a position. Speed is often not a big factor and intelligence has little to do with the geometric skills required to pick up a ball in various environments. If IQ mattered over skills I'd be waiting for my HoF vote.

Turning DP (talking 2B SS) seems to be like a ballet dance of timing and grace. Arm matters but it wouldn't be a top consideration IMO.

[teachermike]

Originally Posted by RchW

A lot of players have trouble catching LD at 3B and 1B even if they are accomplished middle infielders.

Are you suggesting that this persists once the player has spent a significant amount of time at a corner infield position? Because I'd certainly agree with you that this happens when a player is first making the transition; in my mind, this is the purpose the positional experience ratings serve.

Originally Posted by RchW

Judging FB is different than GB and are definable IF/OF skills etc.

Of course they're different skills. I'm suggesting that the error rating is a measure of ability, and those skills are honed with experience at the position.

Originally Posted by RchW

The throwing action at 3B (possibly SS) is radically different than the OF and C. Several years ago Jose Bautista took a week to get his arm ready for a move from RF to 3B to protect his arm. I'd say that 2B and 1B arms may be weaker than many OF but hard to compare with the different action.

Again, I think the difference here is one of ability vs. skill. The fact that it only took Bautista a week to get his arm ready suggests that, while there was a difference in what he was doing, the arm was trainable, and rather easily so. To be sure, not all players can make the transition from one position to another; maybe the solution here would be to randomly (and infrequently) cap a player's position experience rating at non-natural positions.

Originally Posted by RchW

Range is about getting to the ball even when it is on the edge of the normal space for a position. Speed is often not a big factor and intelligence has little to do with the geometric skills required to pick up a ball in various environments.

I've always understood the Intelligence rating to mean baseball intelligence, a player's ability to pick up new things. That absolutely relates to one's ability to find the best route to a fly ball, since every one is a little different than the one before it. Quickness/Acceleration has more to do with range than Speed, and Speed is pretty much irrelevant in Infield range, but you can't tell me that the ability to make up ground quickly doesn't add to an outfielder's ability to reach a ball even when it is on the edge of the normal space for his position.

[frangipard]

Originally Posted by teachermike

I'd like to see an overhaul of the defensive ratings system. The current system works fine until you start trying to move players around the diamond, and specifically from catcher to any other position, or from the infield to the outfield. Since catchers often have very low infield defensive ratings, it's not always clear if they'll be able to make the successful transition to 1B; many slick-fielding, athletic infielders have terrible outfield defensive ratings, so making that transition (even to a low-intensity defensive position like LF) can be unsuccessful even for great infielders.

The current root (i.e. editable) defensive ratings, IF/OF Range, Error, and Arm, and Turn Double Plays, should all become derived ratings like the Contact rating, or be merged. The error ratings should become a single rating; it seems intuitive that a player's ability to catch and hold onto the ball should translate pretty well across positions, given an equal amount of experience at each position. Similarly, the three arm ratings (Catcher, Infield, Outfield) could be replaced by or derived from two universal arm ratings, an Arm Strength and Arm Accuracy. (Maybe these could even tie directly into pitchers' velocity/stuff and control ratings.) OF Range should be largely derived from speed and a dose of intelligence (picking the right route to the ball), and also be affected by a quickness rating; IF Range should be mostly dependent on quickness. Turn Double Plays might be some combination of quickness, intelligence, and arm ratings.

Agree with this concept very much. I get so tired of seeing players with great OF range but 40 speed, or 80 speed and poor range, or guys with 80 arms at 3B but 40 in RF.

No, throwing isn't the exact same at every position, but when teams scout players, they grade throwing as a single tool; in the same way, a guy who is a fast runner on the basepaths isn't going to become slow in CF, so it's a single tool. It doesn't guarantee he'll have great range, but it's a good bet.

In real life, if teams see a guy with a 60 arm and 70 speed making a lot of errors at 3B, they're going to move him to the outfield. If a 6'4" power hitting OF has poor range, he's going to get moved to 1B. If he's 5'11" and can throw, they may try him behind the plate.



Here's a slight amendment to your suggestion. There are currently 10 defensive ratings generated for each player ... I think you could make a better model by generating 7, plus using speed and IQ.

Throw strength
Throw accuracy (affects throwing errors)
Fielding Error
Fly ball judgement (affects range for both OF and IF)
Reflexes (affects IF range, also matters a bit for catchers)
Turn DP
C ability (affects throwing at C position)
Speed
Intelligence


That's a more accurate reflection of the kind of tools scouts look at, and a system built around them is going to build a more realistic model than one that says that Ozzie Smith would have been a disaster in LF because he didn't have an "OF range" skill.

[Matt Arnold]

I can say this is not on tap for 16, but certainly something to consider for later releases. We'd have to be sure that doing this wouldn't suddenly make everyone into a Zobrist type player, but it would be nice to have a guy like Hanley Ramirez make a transition to thy outfield, or Mauer to 1B, without them being worse than dreadful at the new position.

[frangipard]

Originally Posted by Matt Arnold

I can say this is not on tap for 16, but certainly something to consider for later releases. We'd have to be sure that doing this wouldn't suddenly make everyone into a Zobrist type player, but it would be nice to have a guy like Hanley Ramirez make a transition to thy outfield, or Mauer to 1B, without them being worse than dreadful at the new position.

Cool.

It's interesting; OOTP is in some ways a stat-generating-machine kind of game, but playing it has made me more interested in scouting and traditional evaluation methods, and I'm becoming convinced that the best way to make a better stat-generator is to create realistic models of player skills.

[teachermike]

Originally Posted by Matt Arnold

I can say this is not on tap for 16, but certainly something to consider for later releases.

That's great to hear!

Originally Posted by Matt Arnold

We'd have to be sure that doing this wouldn't suddenly make everyone into a Zobrist type player, but it would be nice to have a guy like Hanley Ramirez make a transition to thy outfield, or Mauer to 1B, without them being worse than dreadful at the new position.

This is the balance that's hard to strike, I'd imagine. There has to be a way to account for the difficulty of mastering multiple positions, while still allowing for players to realistically be able to move to new positions (especially the easier, low-priority defensive spots). Maybe the experience portion of the positional ratings could slowly deteriorated when players haven't played in a position for a season or more; that way someone can't go from 1B back to C after five years and still be as good. And maybe some positions (all OF to a degree, but certainly middle infield and corner outfield) could have their experience ratings somewhat tied together--much the same way that, in the Road to the Show mode of the MLB The Show games, when you boost one skill, related skills may get a smaller increase as well.

Originally Posted by frangipard

I get so tired of seeing players with great OF range but 40 speed, or 80 speed and poor range, or guys with 80 arms at 3B but 40 in RF.

No, throwing isn't the exact same at every position, but when teams scout players, they grade throwing as a single tool; in the same way, a guy who is a fast runner on the basepaths isn't going to become slow in CF, so it's a single tool. It doesn't guarantee he'll have great range, but it's a good bet.

Exactly. This is a better explanation for why these changes are necessary than I could have come up with. Thank you.

Originally Posted by frangipard

In real life, if teams see a guy with a 60 arm and 70 speed making a lot of errors at 3B, they're going to move him to the outfield. If a 6'4" power hitting OF has poor range, he's going to get moved to 1B. If he's 5'11" and can throw, they may try him behind the plate.

It would be great to see more realistic movement of players between positions, especially in the minors. So many times, players are signed or drafted and they play one position, but scouting reports note that they project to a different position in the majors.

Originally Posted by frangipard

There are currently 10 defensive ratings generated for each player ... I think you could make a better model by generating 7, plus using speed and IQ... That's a more accurate reflection of the kind of tools scouts look at, and a system built around them is going to build a more realistic model than one that says that Ozzie Smith would have been a disaster in LF because he didn't have an "OF range" skill.

I like your list. Basing more ratings around tools/abilities, and then combining intelligence and experience to determine current skillset, makes a lot of sense.

Originally Posted by frangipard

OOTP is in some ways a stat-generating-machine kind of game, but playing it has made me more interested in scouting and traditional evaluation methods

I definitely hear what you're saying. It's the inclusion of these types of things that gives a primarily text-based game life.

[RchW]

Originally Posted by Matt Arnold

I can say this is not on tap for 16, but certainly something to consider for later releases. We'd have to be sure that doing this wouldn't suddenly make everyone into a Zobrist type player, but it would be nice to have a guy like Hanley Ramirez make a transition to thy outfield, or Mauer to 1B, without them being worse than dreadful at the new position.

Zobrist types (not as good defensively) exist and there are way too many already. The AI forces bats, often power bats into inappropriate defensive positions such as 2B. This causes a cascade throughout a league where good defensive 2B cannot compete for that position because they are up against big bat OF 3B and less often 1B with power who have marginal 2B skills. When these bats reach the threshold to play 2B you end up with 20-30 of the 40 or more MLB 2B being bad fielders. The few successful "true" 2B tend to be unusually good hitters for 2B as that is the only way they can compete. The poor defensive 2B get platooned often, so you will find two players splitting 2B and playing OF (LF RF) or 2B and 3B or 1B plus OF. The moved 2B is sometimes replaced by the SS or 3B or OF partner in this bizarre dance. Infield GP and GS are a mess.

This is completely unrealistic as most real life 2B play only 2B. In 2014 MLB the top 30 2B started 70% of all games at 2B. If you consider a backup then the top 60 2B started 87% of all games at 2B. That is not a lot of sharing. SS is even more exclusive; top 30 76% of all starts top 60 95% of all starts.

2014 Major League Baseball SecondBase | Baseball-Reference.com

[RchW]

See this post in another thread. Essentially the AI took a power bat OF made him into a very bad 2B. He has started most games at 2B for 4 seasons and is 6'6" 235lb.

http://www.ootpdevelopments.com/boar...ml#post3793144

[teachermike]

RchW, I disagree with your premise from the other thread that it's too easy to switch positions. Out of curiosity, I simmed the 2014 season nine times, moving Mookie Betts to each OF position three times each. I used MarkInCincy's suggestion to make sure he got the bulk of his playing time at the desired position, and each time he started 90+% of his games where I wanted to. And at the end of each of those nine seasons, he still had zeroes in all three OF defensive ratings, and he hadn't developed a positional rating at all.

I do agree that the AI doesn't do a good job of determining when it should prioritize defense when determining lineups (it's pretty much geared for maximizing offense), and that it has too many players switching in and out of positions when it doesn't make sense. All in all, the positional ratings system feels incoherent, and often doesn't seem to adhere to the defensive spectrum.

[RchW]

Originally Posted by teachermike

RchW, I disagree with your premise from the other thread that it's too easy to switch positions. Out of curiosity, I simmed the 2014 season nine times, moving Mookie Betts to each OF position three times each. I used MarkInCincy's suggestion to make sure he got the bulk of his playing time at the desired position, and each time he started 90+% of his games where I wanted to. And at the end of each of those nine seasons, he still had zeroes in all three OF defensive ratings, and he hadn't developed a positional rating at all.

I do agree that the AI doesn't do a good job of determining when it should prioritize defense when determining lineups (it's pretty much geared for maximizing offense), and that it has too many players switching in and out of positions when it doesn't make sense. All in all, the positional ratings system feels incoherent, and often doesn't seem to adhere to the defensive spectrum.

I didn't say it was too easy to change positions.

I said quote "defensive ratings and how easy it is to learn new positions". The threshold level of defensive ratings that allows the AI to start a player is too low hence it is too easy to learn a position. Further to that it is too easy to learn a position contrary to the defensive spectrum.

First baseman – Left fielder – Right fielder – Third baseman – Center fielder – Second baseman – Shortstop – Catcher

The player in question was a LF. It should be exceedingly rare for a LF to develop middle infield skill later in life. In this same league there are several other OF with multiple but bad infield ratings playing as MI. They exist because the AI sees only the bat and starts them over qualified 2B and/or other infielders.

A 2B should be able to learn OF more often than above but with limited range and arm.

Then AI minimum threshold for a player to be used regularly at SS/2B should be much higher than any other position, maybe CF should also be included.

In addition the AI should have an internal limit where certain players can be used only in an emergency. This means that in the 8th inning the AI must use the best 2B available not a bench OF or DH who happens to have minimal ratings.

[teachermike]

Originally Posted by RchW

I didn't say it was too easy to change positions.

Originally Posted by RchW

hence it is too easy to learn a position

You are splitting hairs, my friend. I don't even know what the distinction is that you're trying to make.

Originally Posted by RchW

The threshold level of defensive ratings that allows the AI to start a player is too low . Further to that it is too easy to learn a position contrary to the defensive spectrum...In this same league there are several other OF with multiple but bad infield ratings playing as MI. They exist because the AI sees only the bat and starts them over qualified 2B and/or other infielders.

And I already agreed with all of this in my previous post, much of it almost verbatim:

Originally Posted by teachermike

I do agree that the AI doesn't do a good job of determining when it should prioritize defense when determining lineups (it's pretty much geared for maximizing offense), and that it has too many players switching in and out of positions when it doesn't make sense. All in all, the positional ratings system feels incoherent, and often doesn't seem to adhere to the defensive spectrum.

I'm not sure, but I think you're incorrect about where this problem comes from. To be clear, I agree with you that part of the problem comes from the AI making poor lineup choices, and that for SS/2B/CF (I'd add catchers) the AI should do more to prioritize defensive players, when available. But, and this is the crux of the change I'm suggesting, I don't think these guys initially developed their 2B rating because the AI erroneously played them there to get their bat in the lineup. If they didn't already have a 2B rating (or at least infield ratings), I'm convinced from my Mookie-to-the-OF trials that they wouldn't develop a 2B rating at all. I think the AI, when generating new players, tries to account for the fact that some players move around throughout their careers, or are simply utility players, by randomly assigning a low rating at a secondary position when the player is generated. I see a lot of young OF prospects with a 2/20 SS rating; I think these are the same prospects you're seeing play out of position.

The solution to this problem is two-fold, and it's probably a pretty significant change to the engine. First, the root defensive ratings should be more general and tool-based, as they represent ability; positional ratings should come from a combination of root defensive ratings and positional experience--to an extent intelligence should be factored in, and to an extent experience might go up in "clusters" (when you play a lot at SS, it's going to make it easier to transition to 2B later; if you can master 3B, you can probably play 1B; etc). Second, the AI should have some basic limits set that keep the defensive spectrum in mind, such that it won't start a player with a 5/20 SS rating and a great bat if there's someone with a 10/20 SS rating or higher on the bench. These might become customizable, either having an offense vs. defense slider for each position for the user team, or having a league-wide setting for computer teams, but that's another layer of complexity on top of what's already becoming a pretty significant suggested change.

[SirMichaelJordan]

Originally Posted by RchW

Zobrist types (not as good defensively) exist and there are way too many already. The AI forces bats, often power bats into inappropriate defensive positions such as 2B. This causes a cascade throughout a league where good defensive 2B cannot compete for that position because they are up against big bat OF 3B and less often 1B with power who have marginal 2B skills. When these bats reach the threshold to play 2B you end up with 20-30 of the 40 or more MLB 2B being bad fielders. The few successful "true" 2B tend to be unusually good hitters for 2B as that is the only way they can compete. The poor defensive 2B get platooned often, so you will find two players splitting 2B and playing OF (LF RF) or 2B and 3B or 1B plus OF. The moved 2B is sometimes replaced by the SS or 3B or OF partner in this bizarre dance. Infield GP and GS are a mess.

This is completely unrealistic as most real life 2B play only 2B. In 2014 MLB the top 30 2B started 70% of all games at 2B. If you consider a backup then the top 60 2B started 87% of all games at 2B. That is not a lot of sharing. SS is even more exclusive; top 30 76% of all starts top 60 95% of all starts.

2014 Major League Baseball SecondBase | Baseball-Reference.com

Yeah, this is my only problem with the game to date.

I cringe every time I see the AI start a poor & overweight second baseman (who should be playing corner OF by then, but he's bumped by another poor defensive player because of his bat) just because he can hit instead of the good defensive 2b

In most scenarios, the good defensive 2B can play multiple positions and is usually playing musical chairs between many different positions instead of starting everyday at an important position like 2B.


I wish they raise the threshold for C, CF, 2B, SS in 16.

[SirMichaelJordan]

Granted this guy was the #1 pick in the my recent draft but this problem is throughout the later round also (Super Utility player with horrible offense) here is an example of how the game generates super utility players. I hope this is looked at also. There should be some guys who can do it all but not a good chunk of the draft. This is another reason why the game rarely produce any Jeter types (guys who mostly played 1 position for an entire career)



The actual ratings are good but players shouldn't come out the draft with this muh experience at multiple positions. This guy have almost every position experience at 200! Throughout history, a good utility players is usually good at everything but great at nothing.

Is this only exclusive to non feeder league drafts? I've notice that high school and IFA players aren't like this at all and it's mainly the college players.


This is more believable (same guy just edited his stats) he has great defensive ratings so he should be able to easily learn a different position during Spring Training or his career in the minors...

[darkcloud4579]

It could be the way your league is setup, too. That's pretty unusual to see, but it's really a question of your settings versus say, a default setup of the game without knowing more about the variables controlling your league.

I do think the defensive ratings require more nuance and even more detail or whatever, but...one guy doesn't really strike me as that strange.

[SirMichaelJordan]

Originally Posted by darkcloud4579

It could be the way your league is setup, too. That's pretty unusual to see, but it's really a question of your settings versus say, a default setup of the game without knowing more about the variables controlling your league.

I do think the defensive ratings require more nuance and even more detail or whatever, but...one guy doesn't really strike me as that strange.

That was using default settings. But I wonder if this problem is only exclusive with using generated draft classes.

Even If I lowered the fielding PCM, guys (mainly generated college prospects) will just be poor at 3+ different positions instead of being good at 3+ different positions. My main beef is that there doesn't seem to be a primary position in the game. Guys jump in and out of the different positions they have a good enough rating at (which is more often than not a lot of positions)..

The position that is listed on there player profile doesn't mean much in the game outside of the AI deciding what they are in need of.

If the AI is constantly switching positions of the players then the AI really can't judge where they are weak at. I think that is a reason why some can't find a happy median with the trade settings and why things like the posting system and Rule 5 draft was not active enough.

I have a sense that roster management would be much better if this problem was addressed.

[RchW]

Originally Posted by SirMichaelJordan

That was using default settings. But I wonder if this problem is only exclusive with using generated draft classes.

Even If I lowered the fielding PCM, guys (mainly generated college prospects) will just be poor at 3+ different positions instead of being good at 3+ different positions. My main beef is that there doesn't seem to be a primary position in the game. Guys jump in and out of the different positions they have a good enough rating at (which is more often than not a lot of positions)..

The position that is listed on there player profile doesn't mean much in the game outside of the AI deciding what they are in need of.

If the AI is constantly switching positions of the players then the AI really can't judge where they are weak at. I think that is a reason why some can't find a happy median with the trade settings and why things like the posting system and Rule 5 draft was not active enough.

I have a sense that roster management would be much better if this problem was addressed.

Good point. The AI switching positions and essentially forcing players who have no business being infielders to become very bad infielders is a major problem.

Other people should check the GP by position at the end of each season and they will see that this is very unrealistic when compared to real life. IRL players 90%+ of the time have a single dominant position as starting batters.

[Hajew86]

A amendment to your amendment would have the catcher position be more dynamic. Calling games should be a HUGE part of the position, in my opinion. Throwing ability, fielding, pitch framing, arm strength. Or am I to believe all of that stuff is consolidated into the "ability" rating?

[SirMichaelJordan]

Originally Posted by RchW

Good point. The AI switching positions and essentially forcing players who have no business being infielders to become very bad infielders is a major problem.



Other people should check the GP by position at the end of each season and they will see that this is very unrealistic when compared to real life. IRL players 90%+ of the time have a single dominant position as starting batters.

Yea, 15 was better than 14 in this regard but it was pretty common to see an overweight low range 2B start at 2B over a guy who is much better at 2B but has a worse bat.

I would like to see height/weight have much more effect on ratings as the player grow into his body.

If these things are fix, drafting in the later rounds would be important as you'll be looking for depth at varies fielding positions rather than a future bat.

As of now, pretty much every generated player can play everywhere and when they regress in age/weight the ratings don't have much effect on the fielder rating (range). It'll be nice to see fielder rating experience slowly decrease when a player have not played at a specific position for awhile.