Personnel Personalities

[dcordash]

Originally Posted by kidd_05_u2

You are so obtuse that you believe the game is scripted despite all the evidence to the contrary, evidence that you can test yourself very easily. At least you are now being more honest with yourself and admit there is no manual that would satisfy you, only an open-source game.

Whatever. Your statement about my statement is scripted. Does that make sense? I'm expecting about as much as what you just said. And please don't refer to me as an angle. That's rude.

[ubernoob]

Originally Posted by dcordash

I never had that luxury. It worked exactly as advertised or it didn't and suddenly I was being called into a "come to Jesus" meeting with the Directors.

The code I supported was of the state government variety and any citizen of the state was actually entitled to look at it if they so choose. I didn't hide anything because I couldn't. I would have loved to have been able to call bugs..features. But, it didn't work that way. I was accountable and completely transparent.

Guess I'm coming from that "old-school" philosophy and standard. Likely not relevant anymore.

Great. Except none of this applies to making a video game in the year 2022, let alone even the year 2000 or 1980.

[Baby Ruth]

Surely, there must be someone that writes the code and can answer questions about what it does. Or does the code magically get written and compiled by secret programmer fairies in the middle of the night? Good programmers write good code with plenty of commentary for those that might come after them or have to review it.

[low]

Originally Posted by dcordash

Ha, yeah you don't have to be a cynical a$$-hole to find yourself wishing you hadn't dropped another $40-50 bucks for a game you keep getting disappointed by. Just a trusting a$$-hole that is vocal about his convictions and feelings of being letdown.

You know what the definition of insanity is - doing the same thing over and over expecting it to have a different result. S'pose buying OOTP every year is me being an insane a$$-hole. I dunno.

I'm definitely not disappointed as much by it as you are.

[locuspc]

Originally Posted by dcordash

But, no offense, with OOTP product I keep seeing statements from key people (I guess developers) that suggest it's all random and you're liable to see just anything. And, we'll back up the egregiously rare yet prevalent (contradictory statements I'm aware and my point) outcomes with points of time in baseball history that this transpired.

I never had that luxury. It worked exactly as advertised or it didn't and suddenly I was being called into a "come to Jesus" meeting with the Directors.

The funny thing is, I'm also very disappointed in this year's release. Prospect development spent the first two months totally broken and providing ridiculous outcomes, making the game functionally unplayable. That's not something that should be happening in a twelve month release cycle. Even now there are serious unfixed bugs in the game such as the name database loading bug which prevents me using mods to fix the low-quality built-in name and ethnicity database. I am a frustrated customer, I am certainly not interested in blindly defending this product which was released untested and non-functional.

But that being said, I still think your complaints that you've posted across all the various threads lately are way offbase and stem from a philosophical disagreement or a misunderstanding of what the game is trying to simulate. You want the game to work exactly, repeatably, in a specific way. Jack Johnson has the same BABIP rating as John Jackson so they should have the same BABIP at the end of the season. Things should work out exactly according to spec, and if they don't work out exactly that way they've been specifically rigged to work out a different way.

This sort of straightforward perspective may have been what your programming job called for, but it's not the underlying philosophy of OOTP. Baseball has too much variance. Ratings aren't promises, they're stochastic centers of a probability distribution, and every time you run the simulation you end up with a different point on the distribution. That's an explanation that won't fly in a lot of programming contexts, but specifically in the context of designing a simulation? It's exactly what you want. You need to account for the possibility of unobservable factors in your simulation, and add an appropriate amount of variance to simulate them. It's a different world and you should expect different outcomes even when every observable factor stays the same.

[PSUColonel]

I know I am comparing to FM again (but why not if there are some good ideas to steal) but in that game, anytime you basically hover over any of the game concepts, a pop up floating window (much like what OOTP has when you hover over a players name) explaining the concept in question. I think it’s a great way to attack the problem. To me, it would be far more useful and practical than any kind of tutorial.

Just an idea.

[Brad K]

Originally Posted by PSUColonel

I know I am comparing to FM again (but why not if there are some good ideas to steal) but in that game, anytime you basically hover over any of the game concepts, a pop up floating window (much like what OOTP has when you hover over a players name) explaining the concept in question. I think it’s a great way to attack the problem. To me, it would be far more useful and practical than any kind of tutorial.

Just an idea.

It is a great idea. If it was in effect right now you'd hover Real Stats and get the wrong answer. You'd hover Coaches and get no answer.

As I posted earlier, no one knows how this game works. There was no contemporaneous documentation done when code was written.

I'd say this thing has gotten a lot bigger than was ever imagined at the beginning and didn't adapt. The Henry Ford problem, when the company outgrew being able to be run from notes on an envelope in his pocket and they estimated accounts payable by weighing the supplier invoices.

[Scipio Africanus]

Originally Posted by low

I've been convinced for years that a lot of this stuff is just for show. Some of these things are superficial features that don't do anything at all. I'm also a cynical, untrusting a-hole, so take it as you will.

That's my conclusion as well. Even if some of the things in the game (like personality conflicts) have an effect, it's something like a 0.01% change. Not worth messing with unless you're into the role playing aspect of it all.

And a large amount of the things people think they see happening in the game are just random streakiness or a result of the human habit of seeing patterns where none actually exist.

[Scipio Africanus]

On a similar topic, I don't think most people appreciate just how random this game is. I've got a fictional league, 20 teams divided into 2 subleagues with one 10-team division each. Each team gets the same amount of money to spend each season.

Out of curiosity, I turned off injuries and trades, saved the league at the start of the season, and ran that save through a full season 10 times with the same schedule, etc.. The differences in just the standings alone were huge. Only one team managed to win their league pennant 3 times. That same team finish in the middle of the pack several times and near the bottom of their league a couple of times.

Most other teams finished anywhere from 1st to 10th in their league in various run throughs of the same season. Of course, the better teams usually finished in the top half of their league and the bad ones usually finished in the bottom half, but the variation was still wide with some teams finishing in the bottom half most of the time but then winning the pennant in another run through.

It's a very random game.

[Brad K]

Originally Posted by Scipio Africanus

It's a very random game.

The devs are committed to randomness to the point they won't put a floor or a ceiling on outlier performances in the exe. However at the same time they help cold and hot streaks along and in real stats downgrade outlier real life HR performances.

In my current save I had two teams hit .300. Entire teams. In the 90s!

[dannibalcorpse]

Originally Posted by Brad K

The devs are committed to randomness to the point they won't put a floor or a ceiling on outlier performances in the exe. However at the same time they help cold and hot streaks along and in real stats downgrade outlier real life HR performances.

In my current save I had two teams hit .300. Entire teams. In the 90s!

I mean...every MLB season from 1994-1999 had at least one team hit better than .290 for the season, peaking with the Rangers at .293 in 1999. The difference between .293 and .300 with the amount of ABs they had that season (5,651) is 42 base hits over a 162 game season - so basically an extra base hit every 4 games. I've forgotten most of my statistics 101, but I'm pretty sure that should be within the standard deviation for a full season, i.e., not completely insane to have happened.

[dcordash]

Originally Posted by locuspc

But that being said, I still think your complaints that you've posted across all the various threads lately are way offbase and stem from a philosophical disagreement or a misunderstanding of what the game is trying to simulate. You want the game to work exactly, repeatably, in a specific way. Jack Johnson has the same BABIP rating as John Jackson so they should have the same BABIP at the end of the season. Things should work out exactly according to spec, and if they don't work out exactly that way they've been specifically rigged to work out a different way.

This sort of straightforward perspective may have been what your programming job called for, but it's not the underlying philosophy of OOTP. Baseball has too much variance. Ratings aren't promises, they're stochastic centers of a probability distribution, and every time you run the simulation you end up with a different point on the distribution. That's an explanation that won't fly in a lot of programming contexts, but specifically in the context of designing a simulation? It's exactly what you want. You need to account for the possibility of unobservable factors in your simulation, and add an appropriate amount of variance to simulate them. It's a different world and you should expect different outcomes even when every observable factor stays the same.

I believe you've got a very good message. Well thought and articulated. Good job.

But let me ask you something..if OOTP, any sports sims, can just start slavering numbers all over the place, ratings that often aren't relevant to stats generated, settings that "appear" to do nothing, players that can spike, plummet, spike, plummet. Basically just produce a bunch of stats that look like baseball stats, they smell like baseball stats, but are they nothing more than the byproduct of little more than a roulette wheel?

locuspc is suggesting there is no benchmark. MLB is chaos so there is absolutely nothing in OOTP that really needs to make sense. Because, well, MLB doesn't make sense.

Christ, I wish I had a programming job like that. Just spin out numbers and tell customers, don't try to make sense of them or match them to anything you've actually seen in real life because, well, they just might or might not be that. But it's ok. Trust me that it's exactly what you should expect..the unexpected.

You know because we all know MLB has no foundation in physics and talent..MLB/Milb players are little more than generics all capable of producing the same numbers. All capable of having the same ceilings and tanks. Those star ratings..fugheddit about 'em. Mean nothing. But, they sure look good.

Don't be concerned at all with any results you see. If your standings generated from an actual season are nothing close to what happened..if your real-life players are putting up numbers unlike they ever had in their long careers..just play Fictional and ignore it. Try not to put too much emphasis on your fictional players producing numbers that resemble consistency. Cause' it doesn't matter. There is no consistency in MLB so it's ok for this game just to produce whatever number the 'ol stat engine want's to belch out.

Yeah, I'm hard-headed and probably a little too critical of a game I've been playing for a very long time. It's surely better than the 2nd best MLB simulation game..which perhaps there isn't one so maybe not a great point of honor.

However, I've dropped anywhere from $500 to $1k on this game over the many seasons. I'm still seeing things I bitched about when I dropped my initial installment many moons ago hanging around and driving me ape****.

Lot of great things this game does..But 90% of the folks on this board (and I realize that's probably 5 ppl using 50 profiles..so goes the anonymous internet) can chatter about the wonder this game is. That's perfectly fine. Get after it.

But you're going to get some like myself that aren't motivated to giddily promote the great..we're only taking time out because the not-so-great is really giving us the ****s. Perhaps if we shine a light it will be addressed. Hasn't yet, but I'm not one to give up easily.

[Brad K]

Originally Posted by dannibalcorpse

I mean...every MLB season from 1994-1999 had at least one team hit better than .290 for the season, peaking with the Rangers at .293 in 1999. The difference between .293 and .300 with the amount of ABs they had that season (5,651) is 42 base hits over a 162 game season - so basically an extra base hit every 4 games. I've forgotten most of my statistics 101, but I'm pretty sure that should be within the standard deviation for a full season, i.e., not completely insane to have happened.

The peak numbers in real life are due to situations where skill coincided with everything going right. IOW, randomness. OOTP takes peak performance events that include randomness and treats them as if they are skill only events, allowing randomness on top of that.

[Matt Arnold]

As someone who has worked both in fields where numbers need to match exactly and also in the baseball world, I can tell you that this is tougher.

When you have exact numbers to match, you get it right or wrong. If the numbers shift by 1%, you know something has changed, and while it's tough to track down sometimes, it has an answer.

Trying to handle baseball statistics isn't that easy. Yeah, tolerances are larger - 1% shifts I'm batting average aren't big. But if the league batting average rises by 1%a year, you're going to have big problems later.

Never mind that you are constantly trying to balance all the points of a game. If the league average is high, is that because players are rated too highly? Defense didn't save enough hits? Player development started to run off the rails? Not enough LH reliever? AI managing mistakes? Autocalc miss something? Or was it actually not a problem because your baseline values you were looking at were wrong in the other direction. Or is this because your league is 24 teams and not 30, or because you skipped a minor league level, or changed a player dev setting?

It's not perfect. If we truly knew how to build everything perfectly we'd all be working for the Yankees or the Dodgers and telling them what to do. We're always looking to fix up anything that we find isn't right. But it's easy, especially when anything we do has to work for 150 years of history and an infinite amount of fictional universes.

[jcard]

Originally Posted by Brad K

The peak numbers in real life are due to situations where skill coincided with everything going right. IOW, randomness. OOTP takes peak performance events that include randomness and treats them as if they are skill only events, allowing randomness on top of that.

WaitÂ… isnÂ’t this the EXACT opposite argument as that at the heart of your Roger Maris remonstrance?

[kidd_05_u2]

Originally Posted by jcard

WaitÂ… isnÂ’t this the EXACT opposite argument as that at the heart of your Roger Maris remonstrance?

I thought the same thing when I read that.

[Brad K]

Originally Posted by jcard

WaitÂ… isnÂ’t this the EXACT opposite argument as that at the heart of your Roger Maris remonstrance?

No. I want a player to have a chance to reach his peak year but not a chance of exceeding it by a lot. The way the game is set up with Real Stats Maris doesn't have a chance to hit 61 but Mantle has a decent chance to hit 70.

Thanks to both who posted about my past posts on Roger Maris. I appreciate that you remembered.

[dcordash]

Originally Posted by Matt Arnold

As someone who has worked both in fields where numbers need to match exactly and also in the baseball world, I can tell you that this is tougher.

When you have exact numbers to match, you get it right or wrong. If the numbers shift by 1%, you know something has changed, and while it's tough to track down sometimes, it has an answer.

Trying to handle baseball statistics isn't that easy. Yeah, tolerances are larger - 1% shifts I'm batting average aren't big. But if the league batting average rises by 1%a year, you're going to have big problems later.

Never mind that you are constantly trying to balance all the points of a game. If the league average is high, is that because players are rated too highly? Defense didn't save enough hits? Player development started to run off the rails? Not enough LH reliever? AI managing mistakes? Autocalc miss something? Or was it actually not a problem because your baseline values you were looking at were wrong in the other direction. Or is this because your league is 24 teams and not 30, or because you skipped a minor league level, or changed a player dev setting?

It's not perfect. If we truly knew how to build everything perfectly we'd all be working for the Yankees or the Dodgers and telling them what to do. We're always looking to fix up anything that we find isn't right. But it's easy, especially when anything we do has to work for 150 years of history and an infinite amount of fictional universes.

Never did any gaming code, but probably written in excess of half a million lines of biz code. With that said, it's all a balancing act. I absolutely commend the developers that have to maintain this game. I'm sure you do your best.

I probably shouldn't get on the soapbox so often.

It's more the arguing and often inane replies from those I know couldn't spell the word c-o-d-e..believe Java is something you drink to wake you up in the morning, Ajax is what you wash your clothes with and, anything with the word "script" in it they believe is an insinuation the game is "scripted".

They speak SABR fluently and can quote formulas and believe that makes them an expert. That just scratches the surface, You then have to be able to practically apply those formulas into the code, maintain and modify that code w/o upsetting the applecart.

Any hack can write a new program. Much easier to write new code than inherit and have to modify a huge plate of spaghetti code with no documentation. The pro's that can take on that unenviable task and make it work are the real cream of the crop in the programming world.

So relating your SABR expertise to designing and developing game code is like someone that understands a little German suggesting they could translate Mein Kampf, deduce the underlying theme, and explain what was in Hitler's twisted mind that led him down the road to antisemitism. Best acronym I could come up with in the moment.

And then they want to use this "foundation of nothing" to start ripping into someone that may not have been a champion bull-rider, but surely has participated in their share of rodeos.

Amen and good luck, is all I can say to this one.

[jcard]

Originally Posted by Brad K

No. I want a player to have a chance to reach his peak year but not a chance of exceeding it by a lot. The way the game is set up with Real Stats Maris doesn't have a chance to hit 61 but Mantle has a decent chance to hit 70.

Thanks to both who posted about my past posts on Roger Maris. I appreciate that you remembered.

Well, to be fair, it is hardly like you have given us much choicee Ed in the matter.

[Brad K]

Originally Posted by jcard

Well, to be fair, it is hardly like you have given us much choicee Ed in the matter.

LOL.

People tend to remember those who are either charming or obnoxious. But not in that order!