Defensive System

[sebastian0622]

Since I lost this suggestion in the great board crash of 2005...

The defensive system needs to be a lot more intuitive, and, therefore, skills based. To me, this is the part of OOTP that I most want to see fixed (hand-in-hand with bullpen and substitution AI). Here is what I propose:

Each player has four ratings:

1) Range
2) Hands
3) Arm
4) Experience (this goes by position, and could use the existing position graphics)

In simplest terms, they mean:
Range: The player's ability to get to balls
Hands: The player's ability to field balls without making errors
Arm: The player's arm strength (and yes, I think one rating is fine--no need for OF/IF)
Experience: How much the player has played a position. This increases as a player gets more time at a position. After all, aren't there learning curves to each position? Also, right now you can move a player to learn 8 positions without it affecting his original position. Here, there is the tradeoff that giving your phenom SS experience in playing LF takes away the opportunity to give him as much SS experience as possible. Realism, baby! Anyway, experience would basically serve as a coefficient for range and hands. So a highly-skilled player who is new at a position wouldn't perform as well as his skills might indicate, and a highly-experienced player with unimpressive skills would perform at or above those skills.

Doesn't this make sense? I mean, don't these four aspects cover 99% of what makes a fielder good and what a manager looks at when deciding to play a guy somewhere?

Markus, if you make these changes, I will buy two copies of OOTP 2006!*





*not a true statement

[andymac]

My idea is similar to sebastian's, so I will post it here also

Instinct: Basically, a player's natural ability to play a position...represents a player's ability to read & react to a ball hit from that position. This rating should not change much throughout a player's career.
Range: Purely a physical rating, once the player realizes where the ball is going, how quickly can he get to it. I would even argue for having 3 different "range" ratings...one for outfield, for middle infield, and for corner infield. Speed would play a large roll in the outfield range.
Hands: Ability to field the ball once they get there.
Arm Strength: How hard they can throw the ball.
Arm Release: How quickly a player can throw the ball after catching it...quicker releases would be important for infielders.
Rust or Sharpness: Basically a rating that goes up and down depending on how much work a player is getting at a particular position.

[sebastian0622]

Your system would produce more varied results and a larger variety of players, no doubt. It's not as intuitive, to me, as what I propose though. I have a pretty good idea of what instinct is, but how does it differ from range and why can't it be included in that? Why can't arm release just be accounted for in arm strength--so a guy with a real quick release and an average arm might have a the same OOTP arm strength as a guy with an above average arm and slower release? If the end result is the same--if arm strength measures how quickly the ball gets from point A to point B, then they can be rolled into one. Not to be nitpicky, but those are examples of how it just doesn't "feel" straightforward to me.

But, that's not my biggest concern. At first, I wanted to have something a bit more complicated, like something that seperates "jump" from range, with range being correlated to speed and jump being independent of it. I had a few other ideas that I thought would make it more realistic. Then, I started thinking of how to balance realism with being user-friendly, especially to new guys. So I tried to keep it simple, and get as much accuracy as possible from as few ratings (and as intuitive of ratings labels) as possible.

Anyway, that's just where I stand on why I proposed what I did. I guess it depends on what kind of detailed realism/player variation you want to sacrifice for some user-friendliness/intuition (my system) or what kind of detailed realism/player variation you want to include at the expense of some user-friendliness/intuition (your system).

[attackemu]

To that respect, maybe Markus can borrow from his new friends at SI another feature that's in the new FM this year, and that's players being rated at several positions, not just their "natural" one with varying degrees of "goodness."

[Faroo6]

Arm could very well be for infield and outfield.

I played the outfield for several years and would have a quicker release and stronger throw to a base from the outfield than from the infield.

The footwork is different, the arm angles aren't the same, and the same person will have two different throwing strengths from OF and IF.

[gmo]

Originally Posted by sebastian0622

So I tried to keep it simple, and get as much accuracy as possible from as few ratings (and as intuitive of ratings labels) as possible.

I think that is important. The burden should be on those who want to add complication to prove what is being added really suppliments the game more than the new "weight" drags at it. What I want to say is not new, but I just have wanted to re-inject it into the discussion of defense since system proposals have popped up again here.

As long as we are still text-based we should aim not to over-define the problem. (I think assuming the future holds a 2-D physics-driven game engine there might be some nice ways to pile in a lot of ratings, but that is still not necessary and remains at least some way off.) Since throwing in baseball can be considered more complex than merely a single rating, it may be more realistic to have instead of just an "arm" rating several numbers for footwork, grabbing the ball to throw it, strength of throw, accuracy of throw, and perhaps even more. I would argue that pile of factors does not add anything over the simple one-term arm rating. Like noted above it is still about getting the ball from one place to another, and bothering with each step in the process is not really useful when one number can account for the combination of the factors.

On the other hand, though it should correlate to some degree with range, speed alone does not define an outfielders ability to get to balls in the field. So a range factor is necessary. Again rather than range, there could some combination of speed, reading, jump, angle smarts, knowing how to play walls & caroms, etc. Once more though there does not to me seem any benefit of defining each of those actions over simply having one "range" rating.

After espousing the "keep it simple, stupid" philosophy for a while, I can give a few thoughts on the proposals here. I agree with some of the ideas, yet I think the current system in OOTP does a good job overall. I even think infielder arm could have been unnecessary considering throwing the ball ("arm") could mostly be included in making the play ("range"), though there are cases like relays where having it on its own might be a positive addition. I am not especially keen on breaking down everything into the separate tools. The more-ratings tools approach does have the nice aspect of being able to judge from traits X, Y, Z whether a player can handle any given position. It may be neat to be able to more realistically say that Joe Blow needs to play 2B because his arm & first step are not good enough for SS, that he ought to try or give up catching, or that he might give pitching a shot. But getting that end result from the current simple range rating (and maybe scout/coach e-mails that cite those skills as why so-and-so can or cannot play a position even though the skills are not actually split out independently in the game) is sufficient for me.

Where the game currently comes up short I think most can agree is in properly and realistically correlating various traits. A great CF should not have apparently no rating or ability for the corner OF spots, a player with 29 BB and 129 K a season should likely not hit 54 HR, and so on. It is not like the game constantly fails in this regard now, it just seems to too often come up with unrealistic skill sets within players. I hope that is improved, and I think do that with fielding would help that aspect of the game.

Keeping the current basics of making chances (range), throwing (arm), and fielding percentage seems fine with me. Perhaps making errors on fielding the ball and in throwing the ball are distinct enough to break that down or to separate or fold it into the range & arm rating. The notion of sharpness or experience at positions is useful I think, but not as a separate rating rather just as the arc in the ratings already there. How exactly ability at a position should develop, how it should erode if not exercised, how to deal with utility players as opposed one-spot regulars, and other aspects are over my head. But I do not feel a need to to view sharpness/experience beyond the overall (presumably in most cases range) rating showing that influence.

My thinking on all this seems to fit with the idea of all players having basic ratins for all positions. Again, that could be constructed from skill sets using that approach, but I prefer doing it more like how it works now, maybe also taking a talent & ability method like hitting and pitching do (which may already occur for all I know). The key is making sure the ratings make sense overall. A great CF should most likely have LF/RF talent, though maybe that talent has not been developed to show in the ratings. Perhaps with some positions there are small additional hidden factors like being the pivot on a double play for SS/2B or digging out throws at 1B. But the system overall should remain much simpler than this convoluted and redundant post.

[ctorg]

I think it's important to separate infield and outfield range. Infield range is based on reflexes, while outfield range is based on a combo of reflexes and speed. Robin Ventura in his prime, for instance, had superb range at 3B, but no matter how hard he tried, he would never have had good outfield range. Too slow a runner.

[I Am The Game]

As much as we'd all like to make the system more in depth, this has been stated I'm just echoing it, we can't run the risk of making it complicated. There's a difference between in depth, and complicated. While there are a number of us who will understand the majority of what Markus throws in the game, we DO have to remember there are plenty of first timers that are going to play OOTP 2006, and if we make it too complicated those guys could be lost. Hell, just by reading the board you can see there aren't exactly too many bright folks playing it NOW.

As for the defensive system. If we're going to do two arm ratings, I'd rather see strength and accuracy. While a guy who throws the ball from the CF fence to home plate on one hop can easily make a throw from third to first (the argument there is how can an OF with a 10 Arm have a 2 arm for infield - which I agree with), can he do it accurately? Maybe he can, maybe he can't. I'm not a fan of having an OF and an IF arm rating. I'd like for the fielding percentages and ranges to be a little more accurate when players are learning positions on the fly. When a 8 .974 SS learns 2B, why does he seem to start off at 2 .930?? That's very similar to the 9 CF .995 becomes 1 .930 LF argument.

That's my argument with the defensive system. I can live with just having Arm/Pct/Range. I'd just rather have it be a little more accurate when learning positions.

[Fibonacci]

I like the suggestions here. However, they may not be practical for historical simulations. For these suggestions to work, (to me), historical players need to reproduce their historical defensive statistics. If Lahman's and Ankit's DBs remain as they are, and are the recommended DBs for historical simulations; there may not be sufficient statistics to implement the suggested defensive systems. The most important change for me for OOTPBM 2006 is that it utilize the available defensive statistics (in Lahman's and Ankit's DBs) so that individual players abilities will better reflect their real life abilities - as opposed to randomly assigning some of the abilities.

[attackemu]

Bump to make sure everyone saw this...

The defensive system has been overhauled as well. Each player has ratings at the following defensive abilities: Catcher arm, catcher ability, infield arm, infield error prevention, infield range, turning DP’s, outfield arm, outfield range and outfield error prevention. Also, each player has a hidden rating for his experience at each position on the diamond. So, when you have a great SS and intend to move him to 3rd base, he will be a great third baseman soon. If he’s got the arm to handle that position So, each position demands slightly different skills, just like in real life.

[sebastian0622]

Originally Posted by attackemu

Bump to make sure everyone saw this...

Awesome. Sounds like a huge improvement. I especially like the experience rating--not sure if it needs to be hidden or not, but that's no big deal...at least it's there.

[dankrzyz]

I'm glad that the defensive system is being tweaked and given more depth. I always enjoy building a team based on, among other things, defense. In previous editions, though, I was disappointed that my team of all the highest fielding ratings really didn't excell statistically at Field Pct, Errors, Putouts, etc. I hope that same system in BM'06 will really pay off on the field and in the standings.