Fundamental flaws... your chance to vent!

[RonCo]

Originally Posted by Killebrew

Yes, RonCo and Skydog are both big jerkpants some times but I have a lot of confidence in their OOTP player development logic idea's.

I am actually a medium-sized jerkpant, but I'm working on growing a little more every day.

The injury related dev hits seem too extreme now, to the point where serious dev loss after any injury is near predictable. Also, when a player loses 2+ points every single week for a year or more after an injury I wonder if it was better when the player just lost a lot of talent/ratings at once. These death from a thousand cuts injuries are physically painful to watch.

I assume this is because injuries that influence talent will continue to drain the player over time. Not sure, though. I agree that injury issues whould happen as immediate changes, and not cause future fall-off.

[RonCo]

Originally Posted by Markus Heinsohn

No, it works fine. The whole point of proper development is that the average talent level of a league stays the same over the years... and the game does that perfectly. Now the question is, how do I distribute the talent drops... random or injury-related? And I feel the current distribution is just fine... if I move more towards randomness, then we'll have a whole different discussion again

A small technicality...I think the more important point of development should be to keep the average current _ratings_ in a league fairly stable. Unless _talent_ actually influences the results engine, which I'm pretty sure it does not.

If 10 players get big talent bumps, and only two of them actually progress to any degree, then eight of those talent bumps are meaningless in context of the quality of the league.

Right?

[RonCo]

Originally Posted by RonCo

A small technicality...I think the more important point of development should be to keep the average current _ratings_ in a league fairly stable. Unless _talent_ actually influences the results engine, which I'm pretty sure it does not.

Quoting myself here...this means that the player creation algorithm and the development algorithm need to be highly tuned to work together. This isn't really the case in OOTP, and is probably a major cause of some oddities that spring up (like the ubiquitous high BABIP, low everything else guys who make it into the league).

[manooly]

Originally Posted by Malleus Dei

My name is Malleus Dei, and I fully endorse this idea.

I too would really like this idea implemented. I love to start a fictional league with everything equal at the beginning of it.

[deadringer]

Since RonCo has found success in repetition I'll ask again that the overall and potential system be looked at. I still believe that when using the real life major league scales (20-80, 2-8) that the overall and potential ratings should be centered on average. IRL the whole system is based off the league average +/-. A player who is solid at defense at has a contact of 50 shouldn't be ranked the same as a kid in rookie ball who will never have a contact of over 40.

If the Major League batting average is .275 then that's where 50 should be located. Guys that are expected to hit better than that should be rated 50 or better and guys that aren't should be lower. Same goes for all the ratings.

[injury log]

Originally Posted by Markus Heinsohn

No, it works fine. The whole point of proper development is that the average talent level of a league stays the same over the years... and the game does that perfectly. Now the question is, how do I distribute the talent drops... random or injury-related? And I feel the current distribution is just fine... if I move more towards randomness, then we'll have a whole different discussion again

But this is one aspect of development we actually have data on - it's not a matter of opinion. I've analyzed the performance loss of players who have suffered serious injuries in MLB, and OOTP is way too extreme. I'll send you the data.

The other changes sound great, by the way!

[Malleus Dei]

Originally Posted by Markus Heinsohn

Well, I look funny at times too, but I work as intended

Only at times, you think? And I would say *mostly* as intended.

[Malleus Dei]

Originally Posted by manooly

I love to start a fictional league with everything equal at the beginning of it.

A lot of us want this. Let's hope Mister Mostly-Working-As-Intended hears us.

[Malleus Dei]

Originally Posted by RonCo

I am actually a medium-sized jerkpant, but I'm working on growing a little more every day.

Trust me, RonCo, both you and SkyDog have joined me in the majors.

[mikev]

Originally Posted by Markus Heinsohn

OK, an update.

The following things have been fixed in the meantime:
- FA compensation: There is now a supplemental round in the draft, and the compensation rules follow real life.
- Arbitration: You now submit contract amounts, and the arbitrator decides. If no offer is made, the player becomes a FA. Super-2 players are supported as well. Also, 'normal' potential free agents have to be offered arbitration as well if you want to get a compensation pick, following current MLB rules.
- The AI is now more agressive claiming players from waivers.
- Pitching prospects are valued higher now
- Improved player creation code (more utility players and position ratings make more sense now)
- Improved player development (removed player performance in the calculation if a talent change happens)
- Players are now more likely to test free agency (less extensions = better FA class)
- Improved injuries... more short-term, less long-term (I'll look into the CEI later)

That's it for now... I'll continue to keep you posted on the progress

Just to clarify -- this is OOTP 10 stuff you're coding?

[Pineapple]

Originally Posted by Markus Heinsohn

No, it works fine. The whole point of proper development is that the average talent level of a league stays the same over the years... and the game does that perfectly. Now the question is, how do I distribute the talent drops... random or injury-related? And I feel the current distribution is just fine... if I move more towards randomness, then we'll have a whole different discussion again

I don't have the statistical data like RonCo or Injury Log does but I figure the control of talent average via talent drops can occur through the following:

1. Major injuries - Self-explanatory. It's easy to understand how a severe back injury can affect a hitter or how a major arm injury affects a pitcher.

2. Chronic/series of moderate injuries - Some guys just have chronic body type injuries over and over again (Kerry Wood, Mark Prior)... or just guys who seem to always be injured (Ken Griffey Jr.)

3. Consistency - Some players are just more consistent than others until old age factors in (Tony Gwynn, Greg Maddux). Other players in the prime of their careers seen vary quite a bit from their standard deviation.

A counterpoint to #2, not every player who is injury prone have seen their skills diminish considerably. An example I can think of is Milton Bradley or Rich Harden to a lesser extent. So it's not a black and white thing. Some players can overcome injuries and continue with their careers while others can't.

So as a point to #3, I think a similar system to RonCo would be good. Currently the system a potential and current rating scale. To make it realistic, the current rating scale should be more dynamic. For example, it can be further be broken down into two scales similar to Ronco's suggestion current and effective scale. Another method would be some type of sliding scale.

For example, a current contact average rating for a 27 year old Tony Gwynn would 9 (out of 10). Using a sliding scale and some type of hidden consistency rating, in a "bad year" Gwynn's current contact rating could be an 8.6. In a good year, it could be a 9.2. But for the most part, his average is around 9.

For inconsistent players, the scale would "slide" more. The scale does not have to be symmetrical either. For example, a contact rating can be 9 average with a -0.3 on the low end and +0.7 on the high end.

[endgame]

Originally Posted by mikev

Just to clarify -- this is OOTP 10 stuff you're coding?

That'd be my assumption. OOTP9 is considered completed.

[Le Grande Orange]

Originally Posted by Markus Heinsohn

- Arbitration: You now submit contract amounts, and the arbitrator decides. If no offer is made, the player becomes a FA. Super-2 players are supported as well. Also, 'normal' potential free agents have to be offered arbitration as well if you want to get a compensation pick, following current MLB rules.

Markus, just so you know, but there is available data on all the arbitration salary awards in MLB from 1974 onwards. You can find the player's previous salary, the salary he was requesting, the salary the team was offering, and the outcome (i.e. either the player's or team's salary was awarded). This real-world data should help with determining if how the arbitration results in the game are reflecting how the results have worked out in reality.

This page of The Biz of Baseball web site lists the arbitration results for the last couple of seasons. There's a link at the bottom of that page to download a PDF file which lists the arbitration results from 1974-2004.

I downloaded that file sometime ago and was entering the results into an Excel spreadsheet to determine the average percentage salary increases that resulted each year from arbitration. If this might be useful, I can continue on to add more seasons' worth of data to it.

[Kelric]

<taps heels together and wishes> Old development report, old development report, old development report...

[Malleus Dei]

Originally Posted by Kelric

<taps heels together and wishes> Old development report, old development report, old development report...

We do need it back. I was also fond of the promote/demote arrows.

[Carlton_Willey]

On injuries, I don't mind the frequency of CEIs, what I don't like is that they are instantly recognizable as the end of the player's career. I don't think that happened to anyone except Ray Chapman.

What I'd like to see is that we have to figure out if the player is ever going to make a comeback. With that in mind, we should see hits to current ability (and talent, if the player is younger). However, after the 12 months or so of the long injury goes by, the player should show up for work.

Now we can see if he can comeback. Players whom the game currently has as coming back after a 12 month injury should be mixed in with the true (disguised) CEIs. Some players will come back from the injury, others won't. We won't know until we play them at some level.

There should be a small chance of a Tony Conigliaro comeback, or Herb Score coming back to be a C- pitcher. Then we can decide if we want to keep the guy around.

[RonCo]

I think the development reporting feature should be two-fold, one that tracks an individual's entire history and another that tracks a team over the entire year.

Here's a very rough example of what an individual report ought look like for pitchers.

[RonCo]

And here's what it ought look like for a team over the year (for pitchers):

[lt2cents]

The programmer choice to use drag and drop methods is a real drag and should be dropped. And replaced with simple left click methods.

Playing the game 90% or more of the days since OOTP 2006 the track record of drag and drop leading to program crashes is very well established. Longer sequences of drag/drop lead inevitably to program crashes. Whether its the way drag/drop is used in OOTP or whether it is inherently unstable in MS operating systems ( my reference point ) it needs to be replaced with a much simpler left click system on the lineup pages.

A left click system will probably actually be faster for most OOTP players to use. Left click to select player - left click to place. There should also be fewer 'drop' errors as the user can more easily pick the drop location than at the end of a drag. If other uses of left click are already assigned they can be reassigned to other methods, or an extra 'selecting column' can be added to the data.

Replacing drag/drop with other programming methods will go a long way toward stabilizing OOTP.

Now it might be that the lineup selection pages need an even more fundamental redesign, but until then just replacing drag and drop is seriously needed.

[Balker]

How about this for a minor league DL...

If this makes it easier to code, just set it up as a level of the minors that plays no games. You can do that now, but the only addition you would need is that the AI would know to send injured minor league players there, then adjust the rest of the organization accordingly.