One Pitch vs Pitch by Pitch

[RonCo]

However, I still say that the existence od these holes do not, in themselves, say that Markus doesn't listen to the user base. Markus listens to the wave of input--and it is a wave--and picks things to implement that [in no particular order] (1) he's interested in or likes, (2) he knows how to do, and (3) are being asked for by the greatest volume, and (4) he's got time for.

Some times he uses beta testers as a priority sounding board. Others he makes the calls himself. So I stand by the statement that I don't think Markus has any issue in listening to the user base. I think he listens to too many of the user base (spreads himself too thin trying to appease everyone a little) and occasionally weights the desires of the "wrong" set of users too highly (a statement that I fully admit carries my own personal biases to what I want the game to be).

[Malleus Dei]

I get my time estimates from my wife, who writes and maintains serious software. She said maybe five minutes to add clone player and fifteen more to test it, and one morning to code mass select et al with an afternoon to test it.

[robc]

Originally Posted by Malleus Dei

I get my time estimates from my wife, who writes and maintains serious software. She said maybe five minutes to add clone player and fifteen more to test it, and one morning to code mass select et al with an afternoon to test it.

I am by no means trying to discount your wife's knowledge, but it is impossible to know how long something would take without seeing the code for the software being modified. Something that seems simple may end up being a complicated change.

Rob

[BMW]

Originally Posted by Malleus Dei

I get my time estimates from my wife, who writes and maintains serious software. She said maybe five minutes to add clone player and fifteen more to test it, and one morning to code mass select et al with an afternoon to test it.

I wouldn't ask any developer I've ever worked with to give me time estimates to modify code they've never seen, even the best ones.

In fact, the best ones would tell me it's impossible to give any reasonable estimate without ever seeing the code.

[Malleus Dei]

My wife is a highly-paid (IMO quite overpaid) expert, who works on far more complicated software every day, and part of her job is estimating how long something will take to change. Even when she hasn't seen the code. She's very good at it.

The exceptions are when she encounters bad code, and she then rewrites the code, sometimes from from scratch. She recently rewrote something that makes OOTP look like "Hello World," from scratch, in eighty hours. She was Hell to live with while doing it, though.

[RonCo]

Well, I would go all "posting resumes" here, but I figure you just can't possibly trump a guy who plays the "my wife's better than your wife" card.

[Craig Scarborough]

Originally Posted by RonCo

Well, I would go all "posting resumes" here, but I figure you just can't possibly trump a guy who plays the "my wife's better than your wife" card.

Yeah, that's up there with "triple-dog dare"!

[Matt Arnold]

Originally Posted by BMW

I wouldn't ask any developer I've ever worked with to give me time estimates to modify code they've never seen, even the best ones.

In fact, the best ones would tell me it's impossible to give any reasonable estimate without ever seeing the code.

Heck, I work with code, and my estimates for how long something takes are almost always wrong. Even from code that I'm familiar with. There's always something wrong.

But really, clone player shouldn't be hard. Personally, cloning a player is one of the first things I'd implement, since it makes testing much easier.

Mass select is something that might take a day to do, but it might take longer, depending how many other interfaces rely on only having one thing selected at once. Heck, depending on the engine, it might require a whole rewrite of the interface, which definitely wouldn't be done in a day.

And while they'd both be great to have, as what always happens in software, sometimes stuff gets put on the backburner. They'd both be "nice to have, but not essential" features. Usually, those type of things are the last to get put together, if they ever get in. And sometimes, for a feature like mass select, it might be a hassle to get in the current interface, but given that the interface is getting a redesign, it's often easier to say, "let's just wait and design the next one to make this easy" instead of hacking the current one into something that might expose a few more flaws.

[RonCo]

I agree with you UWHabs, but your argument is still clearly inferior to George Peppard's "So Easy My Wife Can Do it With Eight Fingers Tied Behind her Back" supposition.

[Malleus Dei]

Originally Posted by RonCo

I agree with you UWHabs, but your argument is still clearly inferior to George Peppard's "So Easy My Wife Can Do it With Eight Fingers Tied Behind her Back" supposition.

What you don't get is that I'm not bragging or making anything up. She does write far more complex software, she writes it well, and she writes it quickly.

Back before Markus started OOTP I was planning upon my retirement to produce my own baseball simulator, featuring more things than Earl Weaver II did, and market it. I was going to have my wife code it and my daughter the graphic designer do the design work. But then OOTP I came along and I decided that there wasn't that big a need any more.

[GCCWolverine]

Getting back to the original topic - RonCo, I believe I remember seeing a thread about this sort of thing a while ago. Can't be bothered to look it up, but the gist of it is that somebody made the hypothesis that for one-pitch mode, the game automatically came up with the outcome before the pitches were thrown, and the pitches were randomly determined based upon ratings until it got to a point where something specific needed to happen - for instance, if the pre-at-bat determination was a groundout, and on a 3-1 count, you told the hitter to take, it would be a strike to go to 3-2.

The effect of causing batting averages and walk rates to skyrocket is exactly how you described. For instance, suppose the game determines that the next at bat will be a walk. In one-pitch mode, this could take anywhere above four pitches, but will tend to average around 5 or 6. In pitch-by-pitch, if the first two pitches are strikes, and now you tell the hitter to swing away, any pitch that would have been a swinging strike three turns into a foul ball, pushing the average up to 7 or 8. That, in and of itself, doesn't cause issues, but if pitch-by-pitch is pushing up the average at bat by even one pitch per hitter, then instead of the starter having thrown 84 pitches through five innings, he has thrown 106, meaning that every hitter after that is facing a more tired pitcher than they should.

I'd be interested in seeing the average pitches-per-at-bat in those games you are presenting. Again - I don't know if this is how it works - but I distinctly remember reading a thread along this argument's lines.

[Malleus Dei]

Originally Posted by UWHabs

Mass select is something that might take a day to do, but it might take longer, depending how many other interfaces rely on only having one thing selected at once. Heck, depending on the engine, it might require a whole rewrite of the interface, which definitely wouldn't be done in a day.

If the engine requires a rewrite of the entire interface to handle mass select et al, then the whole thing needed to be rewritten anyway. But then you know that.

[RonCo]

Originally Posted by Malleus Dei

What you don't get is that I'm not bragging or making anything up. She does write far more complex software, she writes it well, and she writes it quickly.

I'm just funnin' with it a bit, MD. I've been writing software to some degree since I was 12. I'm 4x that now. So I understand a little.

[RonCo]

Originally Posted by GCCWolverine

Getting back to the original topic - RonCo, I believe I remember seeing a thread about this sort of thing a while ago. Can't be bothered to look it up, but the gist of it is that somebody made the hypothesis that for one-pitch mode, the game automatically came up with the outcome before the pitches were thrown, and the pitches were randomly determined based upon ratings until it got to a point where something specific needed to happen - for instance, if the pre-at-bat determination was a groundout, and on a 3-1 count, you told the hitter to take, it would be a strike to go to 3-2.

The effect of causing batting averages and walk rates to skyrocket is exactly how you described. For instance, suppose the game determines that the next at bat will be a walk. In one-pitch mode, this could take anywhere above four pitches, but will tend to average around 5 or 6. In pitch-by-pitch, if the first two pitches are strikes, and now you tell the hitter to swing away, any pitch that would have been a swinging strike three turns into a foul ball, pushing the average up to 7 or 8. That, in and of itself, doesn't cause issues, but if pitch-by-pitch is pushing up the average at bat by even one pitch per hitter, then instead of the starter having thrown 84 pitches through five innings, he has thrown 106, meaning that every hitter after that is facing a more tired pitcher than they should.

I'd be interested in seeing the average pitches-per-at-bat in those games you are presenting. Again - I don't know if this is how it works - but I distinctly remember reading a thread along this argument's lines.

That was probably my original hypothesis. Markus's comments later led me to believe that he didn't calculate the outcomes first, but instead modified the result calculation by count-based values for each pitch. I would assume he would do this by modifying the league totals for that individual pitch, but I'm really just making that up. However he does it, it doesn't really seem to be working well--though you would need to do a lot more work than I've done to prove that beyond doubt.

I'll pull the pitches/PA numbers when I get a chance. But I'm pretty certain we'll see a large increase.

[RonCo]

However, it would be pretty astounding if the process worked well using league totals adjustments unless you had ratings for each player based on count. So maybe that's not how he does it at all. Dunno.

[JTSMOOTH]

Originally Posted by Malleus Dei

What you don't get is that I'm not bragging or making anything up. She does write far more complex software, she writes it well, and she writes it quickly.

Back before Markus started OOTP I was planning upon my retirement to produce my own baseball simulator, featuring more things than Earl Weaver II did, and market it. I was going to have my wife code it and my daughter the graphic designer do the design work. But then OOTP I came along and I decided that there wasn't that big a need any more.

There is always a need for more baseball games. Maybe its time to resurrect the idea!

[greendiver]

Or a football game. If OOTP fills your needs for a baseball game, you've got to admit football market is wide opening and hungering for a great new game

[Malleus Dei]

That's not a bad idea at all. The only question is can I possibly now tear myself away from my lazy-ass fajita-eating, beer-guzzling, deer-hunting, fishing, gaming and barbecuing happy retirement to do it.

[highandoutside]

Originally Posted by greendiver

Or a football game. If OOTP fills your needs for a baseball game, you've got to admit football market is wide opening and hungering for a great new game

No doubt about it. Although I'm not a huge American Football fan, can you imagine a football game that was a good as Football Manager from Sigames. It would gain a pretty big following, I'm sure. The Madden Series take a lot of the oxygen out of the room but it is no different from the popularity of the Fifa EA series.

[StyxNCa]

Originally Posted by RonCo

I think fans would get behind any team that averaged 9 or 10 runs a game of offense.

They don't IRL now?