This will be the first version I buy in 3 years

[Cactusguy21]

Assuming this is stable, this will be the first version I buy in 3 years, and the second in 5. Post-PT, progress on the single player game has slowed to a crawl. I'm tired of the half-baked ideas like scouting, the endless obsession over graphics, and general lack of new features.


This year changes that. It looks like a lot of the core gameplay features are finally being fleshed out. The IFA changes look like enough of a sell already. Finally something that isn't just haphazardly tossed in. The trading changes seem to make the game deeper and more strategic, I've been begging for a trade deadline screen for years. The in-game stats are a nice fun feature too, I hope they are more than cosmetic and that things like launch angle, ev are actually predictive.


Good work folks, this one looks like an EXCELLENT addition to the series. You got me back.

[Matt Arnold]

I'd argue we've had lots of nice changes in every version recently . But there's definitely a lot of really nice pieces here too that we're excited about. Glad to have you "back".

[Chimaera]

Originally Posted by Cactusguy21

The in-game stats are a nice fun feature too, I hope they are more than cosmetic and that things like launch angle, ev are actually predictive.

The game would have to move to a physics based game engine for these to mean anything, and after the re-write last year, I can't see that happening. If they had done that for this version, it would have been in the trailer.

[chriskelly]

$35 a year? its worth it just for updated rosters. these guys pour their heart into this game and I will buy every year, like I have for at least a dozen years.

[Markus Heinsohn]

Thanks for your support guys!

[jrgst3]

Originally Posted by chriskelly

$35 a year? its worth it just for updated rosters. these guys pour their heart into this game and I will buy every year, like I have for at least a dozen years.

Could not agree more!
I happily pay my 35 dollar annual dues

[Syd Thrift]

Originally Posted by Chimaera

The game would have to move to a physics based game engine for these to mean anything, and after the re-write last year, I can't see that happening. If they had done that for this version, it would have been in the trailer.

That will be a massive overhaul when they do it and I wouldn’t be surprised if they took an entire year off to get it into place a la 6.5. Also it’ll practically kill saved games, probably - it’ll take some time to figure out how to translate ratings properly - and will wreak havoc on historical games in general as well.

[Bobbyraz49]

Originally Posted by chriskelly

$35 a year? its worth it just for updated rosters. these guys pour their heart into this game and I will buy every year, like I have for at least a dozen years.

And this post is why I miss the LIKE button.

[dsthomp]

On the other hand, it's looking like this will be my last purchase. Not an exact count but 24 was the 15th to 20th time I have bought the game. Losing the ability to create a realistic baseball world has always been the game's redeeming feature for me, thus end of the line.

[uruguru]

Originally Posted by Matt Arnold

I'd argue we've had lots of nice changes in every version recently . But there's definitely a lot of really nice pieces here too that we're excited about. Glad to have you "back".

Pitcher BABIP was a great change in OOTP24 for historical games. Thank you for adding that!

[uruguru]

Originally Posted by Chimaera

The game would have to move to a physics based game engine for these to mean anything, and after the re-write last year, I can't see that happening. If they had done that for this version, it would have been in the trailer.

I think switching to a physics-based game would lose a lot of customers unless it could be turned off.

[Syd Thrift]

Originally Posted by uruguru

I think switching to a physics-based game would lose a lot of customers unless it could be turned off.

I'm not at all a fan of the game carrying 2 different engines; however, this is idle speculation as the devs have as much as said that they simply do not have the resources to do the kind of major overhaul that switching to a physics-based engine would employ. I would go a step further and say that perhaps when that physics-based baseball game comes, it won't be OOTP.

[uruguru]

Originally Posted by Syd Thrift

I'm not at all a fan of the game carrying 2 different engines; however, this is idle speculation as the devs have as much as said that they simply do not have the resources to do the kind of major overhaul that switching to a physics-based engine would employ. I would go a step further and say that perhaps when that physics-based baseball game comes, it won't be OOTP.

I agree about the 2 different engines. It would be a nightmare to maintain...

That said, an alternative version of OOTP that is physics-based could still make sense because:

1) they already have the technical "baseball sim" expertise in house

2) they have a built-in brand name and customer base

3) they only have to pay for the MLB rights one time

4) they could bundle both games together as a slightly-more-expensive package. Basically getting a $20 or so premium price from many existing customers who are curious to try out both versions.

If they did it, they would need to branch off the codebase to a separate team. That team would inherit all of the existing db and ui stuff and just recode the internal engine. A monumental task to be sure, but at least less than having to additionally recreate the ui and db as if it were a completely new game.

I'm assuming that, over time, the OOTP codebase has become more modular so that its main components (ui, db & engine) are loosely coupled.

[LansdowneSt]

Why would you change everything learned in 25 years of fine-tuning the engine and go from statistical probability model to physics-based model... just to try and hope to get the same stats for the players with 5,000 new variables at play? I don't understand the desire for this. So, a different algorithm can tell you some formula resulting velo off the bat or launch angle? That's just color. Like the velo of a pitch or it hitting the inside corner. The AB result is what matters.

It's like having my kid show me some new math way of doing a problem with 12 steps and I just take the pencil, carry the one, and get the same answer in two steps. It's the answer that matters. Yet there is always some thread like this wanting the Rube Goldberg-ian physics stuff. Zero upside imho. Just carry the one.

[Syd Thrift]

Well, the answer is, it’s because that’s the direction the next gen stats are going and old school, results based stats are going to be less and less informative and eventually less and less used. Also because a physics based game is more likely to get little things like K rate being a thing that becomes statistically significant faster than BABIP for hitters right, and also of course because the next gen stats get at how a given player plays better than plate appearance based results.

It’s not Strat or APBA at all of course. OOTP is already not those games. I suspect that a move to physics will turn a lot of “old school” players who want the counting stats to be the basis of things off, especially if/when the physics game tries to extrapolate physics based ratings from old school stats. I would expect a physics based engine to take several versions to get historical play “right” and at that would never quite be reeeeeally right because we simply do not have that kind of data from before the 2000s. But it would play more like real baseball, and that’s kind of the whole point.

[MathBandit]

Originally Posted by LansdowneSt

Why would you change everything learned in 25 years of fine-tuning the engine and go from statistical probability model to physics-based model... just to try and hope to get the same stats for the players with 5,000 new variables at play? I don't understand the desire for this. So, a different algorithm can tell you some formula resulting velo off the bat or launch angle? That's just color. Like the velo of a pitch or it hitting the inside corner. The AB result is what matters.

It's like having my kid show me some new math way of doing a problem with 12 steps and I just take the pencil, carry the one, and get the same answer in two steps. It's the answer that matters. Yet there is always some thread like this wanting the Rube Goldberg-ian physics stuff. Zero upside imho. Just carry the one.

For me the appeal would be the introduction of meaningful predictive stats, particularly for hitters. Right now if a guy who has always been good has a brutal April/May, or you call up a AAAA-type kid who rips the cover off the ball for three weeks, there is basically no way to know how much of it is legitimate vs fluke- because the game doesn't give you access to the batted ball data so you don't know if they're getting lucky/unlucky and all you can do is look at the Ks and BBs and try to judge everything about them off those two things alone.

Compare that to a RL example, say Brandon Belt. Last year in his age-35 season at the end of April he was hitting .169/.246/.288 with 1 HR (48 wRC+) and many people wondered if he was washed and had fallen off a cliff due to age. Looking at the underlying batted ball data was encouraging though, and I remember at one point having a conversation with a Jays fan who wanted him DFA'd by pointing out the underlying numbers as well as FanGraphs projections for his rest of year numbers (I don't remember the slashline obviously but remember it was 17 HRs). That person told me they'd bet their house he wouldn't hit 17 HRs because his 'numbers' (the ones you would see in OOTP) made it clear he was done- of course, he went on to hit 18 more.

[LansdowneSt]

Originally Posted by MathBandit

For me the appeal would be the introduction of meaningful predictive stats, particularly for hitters. Right now if a guy who has always been good has a brutal April/May, or you call up a AAAA-type kid who rips the cover off the ball for three weeks, there is basically no way to know how much of it is legitimate vs fluke- because the game doesn't give you access to the batted ball data so you don't know if they're getting lucky/unlucky and all you can do is look at the Ks and BBs and try to judge everything about them off those two things alone.

Compare that to a RL example, say Brandon Belt. Last year in his age-35 season at the end of April he was hitting .169/.246/.288 with 1 HR (48 wRC+) and many people wondered if he was washed and had fallen off a cliff due to age. Looking at the underlying batted ball data was encouraging though, and I remember at one point having a conversation with a Jays fan who wanted him DFA'd by pointing out the underlying numbers as well as FanGraphs projections for his rest of year numbers (I don't remember the slashline obviously but remember it was 17 HRs). That person told me they'd bet their house he wouldn't hit 17 HRs because his 'numbers' (the ones you would see in OOTP) made it clear he was done- of course, he went on to hit 18 more.

In the probability model of OOTP, that same Brandon Belt has a Power rating that implies 17 HR. So, if he has a terrible April, you can infer that he will, as the probabilities sort over time, get the other 16 HR. It's the same result and thought process. The battled ball data in your example is just the Expected Stats less the Achieved Stats in the probability model. You could just dress it up with different words like "Scout: He's barreling the ball per the [insert non trademarked Statcast name] he's got the power in him, I'd keep playing him as there's more gas left in the tank, boss". It's the same math on predictive results, just presented differently. So, there is a "include stats in scouting reports" toggle. Okay, create an option to "utilize OOTPCast metrics in Scouting Reports" toggle. Rework the reports with text that reads that way. There'd be no difference as to what the user sees, and the statistical engine continues on as-is. That's a far more achievable approach to trying to get the kind of modern immersion you seem to be driving at.

[Cod]

Originally Posted by MathBandit

For me the appeal would be the introduction of meaningful predictive stats, particularly for hitters. Right now if a guy who has always been good has a brutal April/May, or you call up a AAAA-type kid who rips the cover off the ball for three weeks, there is basically no way to know how much of it is legitimate vs fluke- because the game doesn't give you access to the batted ball data so you don't know if they're getting lucky/unlucky and all you can do is look at the Ks and BBs and try to judge everything about them off those two things alone.

Some of the best projection systems like Marcel, PECOTA, and ZiPS do not require predictive stats to develop their projection models. And all three tend to be very accurate. ZiPS started incorporating a few predictive stats recently, but only when the data is believed to be very accurate. ZiPS also employs Mahalnobis Distance to help with player comparisons. This is process is explained here: https://brandon.nguyen.vc/2021/06/fi...obis-distance/. Note, the formula uses basic "old school" statistics.

The only predictive model that incorporates predictive stats and statcast metrics thoroughly is THE BAT X projection model. Admittedly, I know very little about this model other than the basics.

[MathBandit]

Originally Posted by LansdowneSt

In the probability model of OOTP, that same Brandon Belt has a Power rating that implies 17 HR. So, if he has a terrible April, you can infer that he will, as the probabilities sort over time, get the other 16 HR. It's the same result and thought process. The battled ball data in your example is just the Expected Stats less the Achieved Stats in the probability model. You could just dress it up with different words like "Scout: He's barreling the ball per the [insert non trademarked Statcast name] he's got the power in him, I'd keep playing him as there's more gas left in the tank, boss". It's the same math on predictive results, just presented differently. So, there is a "include stats in scouting reports" toggle. Okay, create an option to "utilize OOTPCast metrics in Scouting Reports" toggle. Rework the reports with text that reads that way. There'd be no difference as to what the user sees, and the statistical engine continues on as-is. That's a far more achievable approach to trying to get the kind of modern immersion you seem to be driving at.

My Scout saying he has X Power is not a reliable source, though. He could be wrong, he could be biased, Belt could have fallen off since the report, etc. If he has a Y-percentile Barrel rate though I *know* his Power hasn't fallen off.

[LansdowneSt]

Originally Posted by MathBandit

My Scout saying he has X Power is not a reliable source, though. He could be wrong, he could be biased, Belt could have fallen off since the report, etc. If he has a Y-percentile Barrel rate though I *know* his Power hasn't fallen off.

Turn off scouting. You'll know. That doesn't require throwing out 25 years of modeling. That's all color and presentation.