A Proposal For A New Defensive/Speed/Intelligence Rating System

[Jason Moyer]

Since OOTP6 has made large strides towards realism in terms of DIPS implementation and a much more refined player development and aging system, I'd like to propose an idea for pushing the defensive and baserunning systems up to the same level as the pitching and general offense.

This rating system would replace the current position ratings and steal ratings. These ratings would also remove the need for the "taking walks" and "avoiding K's" rating, and would make the game behave in a manner more consistent with real life. There would also be much more interaction between ratings, which would make the rating system for defensive and offensive players more dynamic, and again, more realistic.

1. New ratings
- Speed: How fast a player is. Would effect GIDP, SB%, and fielding range. Would peak when a player is drafted, and decline throughout his career due to aging and leg injuries with no possibility for improvement.
- Glove: Ability to cleanly field a batted ball. Would affect fielding errors and passed balls. Would improve/decline along a normal aging curve.
- Arm Accuracy: Ability to throw accurately. Would affect throwing errors. Again, a normal improvement/declining curve.
- Arm Strength: Would act similarly to current arm strength ratings. Would have a curve similar to speed - starting high and gradually declining through age and arm injuries.
- Patience: Would effect the number of pitches a player sees in a typical at bat. This would simulate the idea of "waiting for your pitch". This would increase strikeouts as well as walks, since the odds of either occurring increase as you get deeper into counts. This would also act as a modifier to a player's hitting, gap power, and power ratings, since the player would be less likely to swing at a bad pitch.
- Intelligence: Would work as a modifier for the patience, glove, and speed ratings. Part of that would be working with the speed rating to determine a player's fielding range - this simulates the idea of "positional range", anticipating a play before it happens. It would also modify the speed rating when determining SB% and GIDP - but not the ability to get a jump when stealing (that would be left strictly to speed). Finally, it would act as a modifier for the patience rating, and also would determine a player's K/BB ratio. This could also affect a catcher's ability to handle a pitching staff. Would tend to start out on the low side and increase as a player gains experience.

2. Positional Familiarity Ratings
- Players would have ratings for defensive familiarity at each position. These ratings would start low and increase as a player gained experience playing a position. These ratings would be the primary modifier for the speed/glove/arm accuracy/arm strength ratings when determining a player's fielding ability. These ratings would increase quickly as a player accumulated playing time at a position, and slowly decrease when a player did not play a position he was already familiar with.

This is quite a bit to munch on, and I'd appreciate feedback on what you guys think of it. The benefits to this would be vastly more realistic aging patterns and an implementation of the defensive spectrum. Imagine being able to have a defensive player switch from SS to 2B in mid career as his fielding ratings decline, and having him properly fit in as a quality second baseman after gaining some spring training and early season experience. Or moving a catcher with no speed but a great arm and glove to left field. Or having a player like Davey Lopez who loses his speed as he gets older and makes up for it on the basepaths by being incredibly intelligent. Or having a player like Mike Schmidt (or thousands of others) who walk 80 times and strikeout 170 times in their early 20's and end up walking 120 times and striking out 80 times at age 38.

[Spindel]

I like it! Sounds like some great stuff.

[professor ape]

Why wouldn't intelligence effect your ability to get a jump when stealing? The ability to read the pitcher steals as many bases as raw speed.

[Jayzone]

First, I am not knocking anybody off or anything. I am just simply providing feedback.

I can not say that I favor a more complex rating system than we have now. There are several potential problems with a more complex system. First, we simply don't know what's the supposed effect of a complex system so it would be very hard to model it right. Second, it would make evaluating player a hassle rather than aiding it. I think a complex rating system is complicating things more than it should be. Third, associated with second, it would make custom player creation/roster creation a nightmare since it would be very hard to get players *right*. As it is right now, if you want increase the chance of success stealing, then you can just increase 'stealing ability'. Want to make a player with lots of SB attempts *without* changing other performance output (eg. player will still have the same fielding ability)? Simply just adjust stealing ability and speed. Going with the system that you proposed, then it would be a very tricky business. I think the first point is the most evident argument that a more complex rating system may not be the best. If we can't get it right, then we shouldn't go after it.

Simple and elegant solution itself may not represent with real world logic too well, but it could be just for the better in the end.

This is just my on the potential downside of going for a more complex rating system.

[Chappy]

Although I agree with many of Jayzone's comments (particularly the bit about it making evaluating players more troublesome), if done right, I'd be in favor of a system like this. Anything that results in a more realistic game - or feel - is good by me.

While I have no issues with the current baserunning system, I feel like something needs to be done to make realistic player movement (among fielding positions) possible without editing ratings - a system like the one Jason proposed would be a step towards this.

[Dan Theman]

I like the approach that people are taking, here - some time back I came up with a similar proposal, but there are a few shortcomings and arguments that I had to face, as well. Jayzone brought up the complexity issue, so I won't bother with that one.

The proposed system seems to break down ratings into these components:

FIELDING
Range -> speed + inteligence + familiarity
Glove -> glove + intelligence + familiarity
Arm -> arm strength + arm accuracy + familiarity
Catching ability (new aspect)

BASERUNNING
Stealing Ability -> speed + intelligence

HITTING
Eye & Avoid K's -> patience + intelligence

**First off, the rating of "Range" within the game is meant to be an effective range, which can be interpreted as already using all three components. That takes a player's related intelligence, foot speed, comfort, and experience all into account.

The concept of positional familiairty is somewhat a part of the engine, although perhaps not enough. I would like to see a smoother transition to a player's current potential fielding ability at a position. Even if it's something that we can't see as owners until the process is complete, it could still be internal and gradually give less of a chance for errors and an increasing range while the player is learning. It's a relatively small point, but I still would prefer it and think most others would as well.

**Second on the list, the ability that you propose called intelligence is unfortunately a troublesome one.

For example; Manny Ramirez is a brilliant hitter. He can break down an at bat with amazing ease. He's also as dumb as a brick in every other baseball-related aspect: he's forgotten how many outs there were more times than I can remember, run into a million baserunning gaffes, and at one point last year made an incredible diving grab to cut off a throw that never should have been cut off.

His intelligence is a singular one, not evenly distributed over all of his baseball skills. In order to properly reflect this with an intelligence rating, we'd need multiple versions of the rating. You could say we already do: Stealing Ability, Running Instincts, Contact, Eye, Avoid K's, and - as mentioned above - Range are all effected by the various facets of baseball intelligence.

**Third, your breakdown of Arm Strength is actually something I'd prefer, although the potential gain of such a switch might be outweighed by the lack of appreciation of the buying public - we may never get our way on that one.

**Last on my list, a catcher's ability to call a game. This has been debated ad nauseum in this forum and many others spanning the internet gamut. In the end, while the proof isn't there to really support any appreciable difference, I recall working on an idea with "The Artist Formerly Known as Henry" on a very basic catcher rating. The notion was that a catcher's game-calling ability would either be developed or not, and would be something that catchers would attain over time. The thought was there to also have perhaps one more setting along the lines of the "cagey veteran" that a few catchers may someday reach. The overall effect wouldn't necessarily be an improvement in pitching performance, but rather an effective improvement in the pitcher's consistency.

[Dark Horse]

Great suggestion, Jamie Moyer.

i feel no need to add nor subtract from this one. All the new ratings you described are perfect. i especially like the Patience rating. It would really work well with the new Pitch-by-Pitch feature.

[jaydh]

yea i like it.. to me, the more complex(aka the more realistic), all the better

[Jayzone]

Originally Posted by jaydh

yea i like it.. to me, the more complex(aka the more realistic), all the better

I just wanted to remind everyone that being more complex does NOT necessarily implicate being more realistic.

Until the day that we can properly model, say, IQ into baseball skills in math accurately, a more complex setup may not be beneficial. Dan has touch more about this in the previous post. This model and other proposed more complex models suffer the same key issue as the supposed catcher's game-calling ability and clutch ability: we can yet to be able to model it accurately in math. As I said before, if we can't get it right, then we shouldn't go after it.

I will admit this, however. If we can model it accurately, then I welcome the system. Though it would still mean a nightmare to roster set makers.

[jaydh]

Originally Posted by Jayzone

I just wanted to remind everyone that being more complex does NOT necessarily implicate being more realistic.

Until the day that we can properly model, say, IQ into baseball skills in math accurately, a more complex setup may not be beneficial. Dan has touch more about this in the previous post. This model and other proposed more complex models suffer the same key issue as the supposed catcher's game-calling ability and clutch ability: we can yet to be able to model it accurately in math. As I said before, if we can't get it right, then we shouldn't go after it.

I will admit this, however. If we can model it accurately, then I welcome the system. Though it would still mean a nightmare to roster set makers.

But I do believe it is the goal of this game to constantly get more realistic, and the only way to do so is to get more complex...nice graphics and endless stats can only be taken so far. I don't think it would be difficult to incorporate several features which could affect a player in various ways. Leave it to the programmers to figure out the algorithms and test them.

[Jayzone]

Originally Posted by jaydh

But I do believe it is the goal of this game to constantly get more realistic, and the only way to do so is to get more complex...nice graphics and endless stats can only be taken so far. I don't think it would be difficult to incorporate several features which could affect a player in various ways. Leave it to the programmers to figure out the algorithms and test them.

I totally agree on the part that the goal is to become more realistic. I am sure everyone agree on that and I would love to see the goal realized fully.

However, as I said, if it can not be done right, then it should not be implemented. As it is right now, I am not convinced that we have been able to model the system accurately in math. Or at least, I have yet to see any good sabarmetric analysis on the effect of IQ when it comes to baseball skills. If we want to model something *realistically*, then we will have to be able to quantify it. If an effect of a parameter can not be analyzed mathematically, then the implemenation of the model is all guess work. And, more often than not, guess work has a hard time being realistic - or correct.

[knockahoma]

Actually, they say genius is found in simplicity

I have a question on how the current system works. I've noticed that some roster-makers make range ratings specific to position. For example, no catcher has a range rating over 60. Few shortstops have a range lower than 70. It's as if the roster-makers lined up all the players at SS and said, "Let's see your range, regardless of what position you play."

Then I see other roster makers seemingly grading on a curve. Johnny Bench is a 99, all the way down to Johnny On The Bench rated at 13.

How does the engine work? I would assume the latter is how it's set up. Is there some reason to line up everbody at SS to check their range?

thanks

Rick