Data update only for 26

[Brad K]

Originally Posted by Sweed

From your thread am I to take it that the AI turned a DP on every ground ball hit to ss or 2b with a runner on and less than two outs? For the most part there are only two situations where a DP won't be turned (if the fielders are competent). Ball hit to slow to get two. AND the runner was in motion, ie hit and run. IOW if on a gb the SS or 2B is taking the fielder's choice at 1b rather than going for the lead runner it is probably because of a H&R, with the exception of a slowly hit ball.

Are you saying that most of the time a runner makes second on GB to 4 or 6 then the hit and run was on even though "runner goes" isn't mentioned in the play by paly or shown in the animation? That makes sense from the standpoint that hit and run results can't be determined from available data and is aleady in included in SB, CS, runners taking an extra base, and double plays.

if so then there remains a peculiarity in that the animation occasionally shows a runner starting, although rarely, and more often for the opposition than for my team.

[Brad K]

Originally Posted by Syd Thrift

lmao what If I told you that 10 is a feature, not a bug

You missed the point. If it's a feature then caps and floors would be a feature.

[Syd Thrift]

Originally Posted by Brad K

You missed the point. If it's a feature then caps and floors would be a feature.

Nope, it is in fact a feature and one that's borne out by the data: ballplayers tend to be significantly more prone to streaks, good and bad, than real life. You trying to add in additional things that "should" be in there because of what your fee-fees think about streakiness does not change this.

[Brad K]

Originally Posted by Syd Thrift

Nope, it is in fact a feature and one that's borne out by the data: ballplayers tend to be significantly more prone to streaks, good and bad, than real life. You trying to add in additional things that "should" be in there because of what your fee-fees think about streakiness does not change this.

What does this mean? Ballplayers (where? real life?) are more prone to streaks than real life? Like whaaat?

[Brad K]

Originally Posted by Brad

2. A historical what if league where the talent is different from the historical league should perform according to the talent present and not be hammered by LTMs into producing historical stats.

Originally Posted by Garlon

2. You can turn off the league totals modifiers if you want.

This is a non answer. Doing this isn't a viable option. It's established the game can't produce correct output from ratings from historical stats without triple tested LTMs. That means it can't produce plausible output with changed talent. It produces fantasy.

Originally Posted by Brad

3. If retire according to history is not selected then some players should have their careers shortened to compensate for those who play longer.

Originally Posted by Garlon

3. If there are only 25 roster spots then sone players will not play anyway."

From This Thread post 8

Originally Posted by Garlon

You should also play with retire according to history enabled so that you do not have excess talent in the league which will also suppress league leaders.

Your dismissal of the idea to take some players out of the game early when retire according to history is not selected contradicts your oft stated recommendation to set retire according to history for best accuracy. Both are ways to prevent having too much talent in the game.

[Syd Thrift]

Originally Posted by Brad K

What does this mean? Ballplayers (where? real life?) are more prone to streaks than real life? Like whaaat?

Ballplayers are more prone to streaks than regular statistical distribution would have you expect. This isn't hard, although your track record is pretty consistent with you not getting it.

[Brad K]

Originally Posted by Syd Thrift

Ballplayers are more prone to streaks than regular statistical distribution would have you expect. This isn't hard, although your track record is pretty consistent with you not getting it.

Allegations are all you have that I don't get it.

[Matt Arnold]

For streaks, they're basically just giving players a slight boost to their performance, in the same ways that coaching, morale, fatigue, rust, etc... interact. The problem with applying floors and caps to player performance is that to actually keep things in line, you would need some brutally high adjustments. If 2001 Bonds is at 25 HR at the All-Star break, if we had a floor in you'd know that he'd have to hit at a crazy rate to reach the floor. Or if that's not enough, if we said "okay, he needs 50 HR in 500 AB and he's at 40 in 490" then all of a sudden your "floor" algorithm has him hitting 10 straight HR with no way to change that. That seems incredibly wrong.

As for LTMs, that's part of the game. With the heavy ratings changes this year, the goal is to make sure that LTM don't really need to come in to play other than just some fine tuning around the edges. In new fictional games, the actual league outputs are balanced enough that you don't need to use auto-calc to have stable results.

I mean, if you create a historical game with 0 in all the adjust/weaken, and you play out every game manually ensuring that everyone's usage is 100% historical, your results will likely be very close to accurate. But most players don't want to do that, and so once the AI starts in, LTM is just one way to balance that. We do always strive to improve the historical usage.

[Brad K]

Originally Posted by Matt Arnold

For streaks, they're basically just giving players a slight boost to their performance, in the same ways that coaching, morale, fatigue, rust, etc... interact. The problem with applying floors and caps to player performance is that to actually keep things in line, you would need some brutally high adjustments. If 2001 Bonds is at 25 HR at the All-Star break, if we had a floor in you'd know that he'd have to hit at a crazy rate to reach the floor. Or if that's not enough, if we said "okay, he needs 50 HR in 500 AB and he's at 40 in 490" then all of a sudden your "floor" algorithm has him hitting 10 straight HR with no way to change that. That seems incredibly wrong.

As for LTMs, that's part of the game. With the heavy ratings changes this year, the goal is to make sure that LTM don't really need to come in to play other than just some fine tuning around the edges. In new fictional games, the actual league outputs are balanced enough that you don't need to use auto-calc to have stable results.

I mean, if you create a historical game with 0 in all the adjust/weaken, and you play out every game manually ensuring that everyone's usage is 100% historical, your results will likely be very close to accurate. But most players don't want to do that, and so once the AI starts in, LTM is just one way to balance that. We do always strive to improve the historical usage.

I'm glad you are working toward a lot less reliance on LTMs. That shows a recognition that ratings aren't accurate and they need to be.

Concerning floors and ceilings it appears I've not been clear on what I mean by that. What I mean is I don't want players hitting substantially better than their best talent year with all the breaks going their way or substantially worse than their worst year when they just didn't have it and all the breaks went against them.

The example I usually use is when Tony Armas hit 325 then 179 in consecutive years on 3 year recalc not weighted. I'm not worried about things like your example with Bonds hitting 40 HRs instead of 50. That's a reasonable variation.

[Brad K]

Originally Posted by Sweed

For the most part there are only two situations where a DP won't be turned (if the fielders are competent). Ball hit to slow to get two. AND the runner was in motion, ie hit and run. IOW if on a gb the SS or 2B is taking the fielder's choice at 1b rather than going for the lead runner it is probably because of a H&R, with the exception of a slowly hit ball.

The game engine produces hits and outs. It does not produce slow rollers to 4 or 6 and neither does it produce scorching sure double play balls directed at those positions.

[Brad K]

My original comments 1, 6, and 10 are, after discussion, settled issues. Despite this being an active thread (638 views as of this writing) no one has disputed the conclusions of my counter response in post 25 on my original comments 2 and 3. I believe given the number of views the thread has received these can be considered settled issues since two days have passed and those who would be expected to oppose have been active on the board but not in this thread.

Given that five comments are settled in favor of the original statements, I'm moving on to a couple more.

Originally Posted by Brad

4. Real stats should be neutralized against the game's park factors or, if they already are as some suspect, then that should be confirmed by devs.

Originally Posted by Garlon

4. You can do a test if this yourself with a neutral park factors file and load the same season and see if the ratings are any different.

The manual up to 24 explains real stats indicating they aren't neutralized. It further endorses your neutralized stats as the most accurate way to play. Yet in the game setup in defaults to real stats.

You have stated that real stats are neutralized, contradicting the manual. The devs are silent. This is something for them to clarify. The question has been asked many times. Why should the answer be for me (or anyone else) to test this? This isn't some obscure function. This is the basis of the game where the game itself provides conflicting information (a recommendation to use neutralized while defaulting to real.) Something like this being central to the game must be resolved by a statement from the devs, not my me or someone else conducting a test to prove you or them wrong.

Originally Posted by Brad

5. Choices that don't apply to historical games should be grayed out. Effect of coaching, development budget, and psychobabble etc shouldn't apply to historical players' performance. They should perform in the ballpark of how they did historically based on randomness and the selection of TCR.

Originally Posted by Garlon

5. The game is a sandbox and sone people may want to use those options.

The point isn't that people may want to use them. The point is that LTMs hammering output into historical compliance is evidence exercising those options has little or no effect in historical saves. And they shouldn't have an effect. They are for fantasy games with fictional players. So for historical games they shouldn't be offered as choices. It's like a game at the fair where the goal is to toss a ball through a hoop but the ball is bigger than the hoop.

[Garlon]

The game uses the modifiers to account for every variable that is selected when you set up the game so that league output tracks correctly to a given season.

There are literally millions of ways to set up a historical league.

While the modifiers account for these other options it does not mean those things have no effect. The modifiers are not cancelling the effect of the options.

The game will always need the modifiers. You act like the modifiers are some sort of cheat code. They are just modifying the dice rolls in an appropriate way for a given season and set of options. They are generated dynamically each time you play and they provide accurate results.

Try Diamond Mind Baseball or some other game and compare their league results across all of those parameters and you will see that those other games are not even close to accurate.

[Sweed]

Originally Posted by Brad K

The game engine produces hits and outs. It does not produce slow rollers to 4 or 6 and neither does it produce scorching sure double play balls directed at those positions.

But it does produce ground balls that do not result in double plays with a runner NOT in motion. IE the only play is to first base. AND it produces plays where the runner is not in motion that do result in DP's.

If you go back to v6.5 and before, yes the game first determined hits and outs and then filled in the pbp. This is pre DIPS, when things like pitchers being rated for how many doubles they gave up was a part of the game . After DIPS came in the game determined K/BB/Ball in Play. If in play it then determined location and defense to determine if it was fielded and then if an out was recorded, or an error occurred as the play unfolded. (Side note: when OOTP went to DIPS many users (not you) predicted it would be the end of OOTP )

The game, after the total rewrite following v6.5, does more than produce hits or outs. After the game determines the ball is in play (not a BB, K, HBP), the engine does determine where the ball is hit and takes into account the fielders ability and positioning
(shift left/right, DP, IF in, corners in, etc.) within the engine to determine the "out/hit" result you are talking about.


The outcome of SS taking the out at 1b, without the runner being in motion, can be a play that is determined by real life data and the video shows the play in that way. The ball from a physics standpoint did not roll slowly, but the game simulates, in that one instance that it did, and creates a video that represents that play. Sure we may find fault with how the play is shown as the 3d continues to be a WIP (whether one likes it or not). Point being ball location and defender's ability lead to the FC at 1b. The game did not determine "out" and then make up the play after.

Then there are the balls hit to SS/2b/3b etc. at "normal" speed that would result in a DP if the runner was not in motion. The same runners I've seen in motion playing out over 600 games in v24 (and actually all games post v6.5), where the SS had to take the out at 1b.

Look at it this way.. The game gives both teams the option to set the IF at DP depth. This results in more DP's, but at the expense of more balls getting through since the ss and 2b cheat in a bit from normal depth. The counter to this move from the offensive side is to send the runner in motion. The H&R you want removed from the game since the AI (according to you) doesn't use it. However the AI clearly does. This is shown when the SS only has a play at 1b, or the runner from first is close enough to break up the DP. So back to my setting of the AI controls baserunning and me seeing DP's being broken up (when I did not send the runner) , or there only being a play at 1b shows the AI will send runners to combat the DP. Not just for it's team, but also when it has control of baserunning for the human user.

Originally Posted by Brad K

Are you saying that most of the time a runner makes second on GB to 4 or 6 then the hit and run was on even though "runner goes" isn't mentioned in the play by paly or shown in the animation? That makes sense from the standpoint that hit and run results can't be determined from available data and is aleady in included in SB, CS, runners taking an extra base, and double plays.

if so then there remains a peculiarity in that the animation occasionally shows a runner starting, although rarely, and more often for the opposition than for my team.

I don't see anything not being mentioned in the pbp as a reason to think it isn't represented in the game. Some plays are described in the text file some aren't. For those that are some are in great detail, others aren't. Whoever worked on the text had "x" amount of time and had to make decisions on what situations were described in depth, in short, or not at all? IDK, the reasons one can think of are endless, be it "time to add" or "didn't think of it". When OOTP was text only and a gb ended up as an out at 1b it just seemed to me to mean there was not a play to be made at 2b for whatever reason, be it location of batted ball, runner in motion, etc. IE you read the text and saw the play in your head.

I'd argue the runner is seen in motion, but in an incomplete (for lack of a better word) way? I play out all games and when there is not a play at 2b the runner is clearly closer to the bag then what I see on a DP. If the out is only made at 2b the runner is clearly closer and breaking up the play with his slide.
In both cases the runner had to be in motion. He may not be leaving 1B in a "visually realistic way", but he is closer to 2b when the play is made. IE if as the SS throws towards 1b, I look towards 2b, I'll see the runner is close. Same for the ss only getting the out at 2b as the runner will be sliding.

There are lot's of peculiarities in the animations. I look at that one as an attempted steal where the runner didn't get the jump. Could be the AI manager has a tendency to steal more than you do? IDK.

[Sweed]

Originally Posted by Garlon

The game uses the modifiers to account for every variable that is selected when you set up the game so that league output tracks correctly to a given season.

There are literally millions of ways to set up a historical league.

While the modifiers account for these other options it does not mean those things have no effect. The modifiers are not cancelling the effect of the options.

The game will always need the modifiers. You act like the modifiers are some sort of cheat code. They are just modifying the dice rolls in an appropriate way for a given season and set of options. They are generated dynamically each time you play and they provide accurate results.

Try Diamond Mind Baseball or some other game and compare their league results across all of those parameters and you will see that those other games are not even close to accurate.

This. The LTM's are the backbone that allow the seasons from all eras to be played and give realistic results.

I like Matt saying

As for LTMs, that's part of the game. With the heavy ratings changes this year, the goal is to make sure that LTM don't really need to come in to play other than just some fine tuning around the edges. In new fictional games, the actual league outputs are balanced enough that you don't need to use auto-calc to have stable results

I'd imagine this has been on his plate for awhile and was planned as a part of increasing the ratings scale. Finetuning the LTM's so they "only" tweak your game compared to large changes is a great thing.

[Brad K]

Originally Posted by Garlon

The game uses the modifiers to account for every variable that is selected when you set up the game so that league output tracks correctly to a given season.

There are literally millions of ways to set up a historical league.

Your first paragraph admits LTMs over ride strategy settings.

My example is not of the millions of ways to set up a historical league but rather that with everything historical, lineups, retire, etc, the ratings cannot produce historical output without thrice tested and adjusted LTMs. And also that when things aren't set exactly historical that LTMs over ride those settings instead of allowing plausibly alternate output.

Originally Posted by Garlon

The game will always need the modifiers

That was contradicted by Matt before you posted.

Originally Posted by Matt

With the heavy ratings changes this year, the goal is to make sure that LTM don't really need to come in to play other than just some fine tuning around the edges. In new fictional games, the actual league outputs are balanced enough that you don't need to use auto-calc to have stable results.

Right in this thread, before your post, Matt stated a goal to not have LTMs and also that he considers the effort most of the way there. Do you think the devs can't do that?

Originally Posted by sweed

After the game determines the ball is in play (not a BB, K, HBP), the engine does determine where the ball is hit and takes into account the fielders ability and positioning

Originally Posted by sweed

Then there are the balls hit to SS/2b/3b etc. at "normal" speed that would result in a DP if the runner was not in motion. The same runners I've seen in motion playing out over 600 games in v24 (and actually all games post v6.5), where the SS had to take the out at 1b.

You are using the animation of the play to say what the game engine did. In reality it is clear that the animation is not an accurate representation of what the game engine did. Of course, it feels "right" to remove a pitcher who just gave up a pair of warning track fly outs with the idea he is losing it.

Originally Posted by sweed

This. The LTM's are the backbone that allow the seasons from all eras to be played and give realistic results.

LTMs are what allowed realistic performances for the era with random debut, and allows the 1927 Yankees to play in 1968. And your following quote of Matt shows a desire to get rid of them, not that they are essential. I quoted Matt's statement above in response to Garlon so I won't requote in here.

[Garlon]

There are 43 parameters that the modifiers are modeling. Those are not going anywhere if you want the game to produce accurate results in any given season.

We have actually added MORE modifiers for OOTP25 than we had in OOTP24.

[Scoman]

LTMs are your friend.

[Garlon]

They work in tandem with the millions of combinations of settings you can use when setting up your league. They are not overriding anything.

[Sweed]

Originally Posted by Brad K

Your first paragraph admits LTMs over ride strategy settings.

My example is not of the millions of ways to set up a historical league but rather that with everything historical, lineups, retire, etc, the ratings cannot produce historical output without thrice tested and adjusted LTMs. And also that when things aren't set exactly historical that LTMs over ride those settings instead of allowing plausibly alternate output.




That was contradicted by Matt before you posted.





Right in this thread, before your post, Matt stated a goal to not have LTMs and also that he considers the effort most of the way there. Do you think the devs can't do that?

He did not say his goal was to not have LTMs. He said..

Originally Posted by Matt Arnold

the goal is to make sure that LTM don't really need to come in to play other than just some fine tuning around the edges.

That's not removing them, is it? I then simply said I liked what Matt was saying.

Would you like to hear the first conspiracy theory if LTM's were to disappear?
They are still there, just hidden under the hood, with autocalc also being run under the hood. "The users will never figure it out. We are just too damn smart!" followed by an "evil laugh".

Originally Posted by Brad K

You are using the animation of the play to say what the game engine did. In reality it is clear that the animation is not an accurate representation of what the game engine did. Of course, it feels "right" to remove a pitcher who just gave up a pair of warning track fly outs with the idea he is losing it.

I am?

Originally Posted by Sweed

and creates a video that represents that play. Sure we may find fault with how the play is shown as the 3d continues to be a WIP (whether one likes it or not).

I literally said the game creates a video that represents the result of the play, and the play comes from where? The engine. I also noted that the 3d is a WIP and we may find fault in how the play is shown. I honestly have no idea if you only skimmed my post or ignored the content?

With regard to the 3d I can only say, if you play out a game and have never seen a gb to ss and thought "DP" only to see the runner from first is already to close to 2b for a play to be made you must not be watching. This happens not just for the AI team, but also for my team, with baserunning set to AI, meaning the AI started that runner, not me.

My original point of coming into the thread was to show the AI does use H&R, for both AI teams and human teams (something you claimed it does not do) with AI controlled baserunning.
Your answer, as I read it, seemed to at least give credence to my explanation. However you said something about the pbp doesn't say "the runner goes". My reply is above in an earlier post, but the general thought is the pbp doesn't go into detail for many things.

If you remove a P because he gave up a pair of flyballs to the track because it's a sign he's losing it, I don't know what to say. It's not something I'd ever do, but feel free to manage how you manage.

Originally Posted by Brad K

LTMs are what allowed realistic performances for the era with random debut, and allows the 1927 Yankees to play in 1968. And your following quote of Matt shows a desire to get rid of them, not that they are essential. I quoted Matt's statement above in response to Garlon so I won't requote in here.

As stated above he does not say there is a goal to "get rid of them".


At the end of the day v26 will not be a data update only. The masses would revolt even more than they do now, even if/when there is a big under the hood update. As they also do if there is a big update we can see. According to many, no matter what they do, the only thing a new version brings is a "roster update".

[Brad K]

Originally Posted by Scoman

LTMs are your friend.

Correct... because with everything historical the game cannot produce historical output without them. But they're not a friend to anyone playing historical what if or random debut because they hammer those saves into historical output despite the talent. Make the ratings produce correct output and LTMs aren't needed.