Player development in OOTP 9.2.2

[Markus Heinsohn]

Originally Posted by Childress

It's been years since reading the Almanac, but one of Bill James' findings was that strikeout pitchers tended to enjoy longer and more productive careers- with consistent K totals- than control specialists. Also, fast base runners held up better. It seems counter-intuitive be he backed it up.

OOTP 9 reflects this...

[phenom]

I think OOTP 10 should have a checkbox for "Perform EXACTLY as the player did in real life" which triggers HTML output to redirect to Baseball-Reference.com - Major League Baseball Statistics and History

[Markus Heinsohn]

Originally Posted by phenom

I think OOTP 10 should have a checkbox for "Perform EXACTLY as the player did in real life" which triggers HTML output to redirect to Baseball-Reference.com - Major League Baseball Statistics and History

I plan the following checkbox:
[x] Read the user's mind and perform exactly as he expects, no matter if realistic or not.

[phenom]

[Corsairs]

If I can throw my in here, the message I take away from this thread really doesn't have anything to do with whether or not RonCo's modifiers are adopted by Markus. I think what's being highlighted here is the importance of communication.

I started this thread with no intention of igniting an argument between two divergent camps. I started it because there was a note from Markus about a 9.2.2 update - "Fine-tuned player development" - that was vague and in need of clarification. Even in the full patch list this change hasn't been expounded on (it simply reads "Player development has been further tuned").

Markus has since kindly explained the change, so my question has been answered this time around. Unfortunately, this thread has evolved into an argument between those who side with Markus' view on player development vs. those who side with RonCo's. Had the change been explained up front, this thread would never have been started and the argument may have been averted.

If I may offer one critique (and this reflects just my own opinion), I feel there are times when important information about the workings of the game is left unclear or unexplained entirely. It's important for us, the users, to be mindful of the fact that the developer's time is limited and he can't be posting here all day answering each and every one of our questions. I personally accept that and am glad that so much time is being devoted to development. That being said, I believe it's also important to make sure that key game information gets communicated to the users clearly and in a timely fashion. Users shouldn't have to guess what a fine-tuning of player development means; it should be spelled out (and as it turns out, it only took a sentence to do so).

Good communication has positive benefits for the developer as well as the users. While there's always going to be that vocal minority that complains loudly and bitterly when they don't get their way, I believe the vast majority of the user base is predisposed to give the developer the benefit of the doubt... if they understand where he's coming from! I know I feel much better about Markus' reasoning behind player development after reading his posts in this thread, for instance (even if I don't necessarily agree with all his opinions).

So that'd be my message: place an emphasis on communicating key game changes and addressing questions about fundamental game mechanics clearly and promptly to avoid confusion. Less confusion should equal less arguments and greater understanding of the developer's point of view.

[Markus Heinsohn]

Originally Posted by Corsairs

If I can throw my in here, the message I take away from this thread really doesn't have anything to do with whether or not RonCo's modifiers are adopted by Markus. I think what's being highlighted here is the importance of communication.

I started this thread with no intention of igniting an argument between two divergent camps. I started it because there was a note from Markus about a 9.2.2 update - "Fine-tuned player development" - that was vague and in need of clarification. Even in the full patch list this change hasn't been expounded on (it simply reads "Player development has been further tuned").

Markus has since kindly explained the change, so my question has been answered this time around. Unfortunately, this thread has evolved into an argument between those who side with Markus' view on player development vs. those who side with RonCo's. Had the change been explained up front, this thread would never have been started and the argument may have been averted.

If I may offer one critique (and this reflects just my own opinion), I feel there are times when important information about the workings of the game is left unclear or unexplained entirely. It's important for us, the users, to be mindful of the fact that the developer's time is limited and he can't be posting here all day answering each and every one of our questions. I personally accept that and am glad that so much time is being devoted to development. That being said, I believe it's also important to make sure that key game information gets communicated to the users clearly and in a timely fashion. Users shouldn't have to guess what a fine-tuning of player development means; it should be spelled out (and as it turns out, it only took a sentence to do so).

Good communication has positive benefits for the developer as well as the users. While there's always going to be that vocal minority that complains loudly and bitterly when they don't get their way, I believe the vast majority of the user base is predisposed to give the developer the benefit of the doubt... if they understand where he's coming from! I know I feel much better about Markus' reasoning behind player development after reading his posts in this thread, for instance (even if I don't necessarily agree with all his opinions).

So that'd be my message: place an emphasis on communicating key game changes and addressing questions about fundamental game mechanics clearly and promptly to avoid confusion. Less confusion should equal less arguments and greater understanding of the developer's point of view.

You're right, communication is a good thing! However, if I'd have to document every change that could conflict somehow with some sort of setting that was suggested by a user, then I wouldn't have time coding anymore... One thing can to be taken for granted though, whenever I tweak something and do not get into detail explaining the tweak, then it's definitely not a massive change but rather a fine nuance that was changed.

[Syd Thrift]

Originally Posted by Markus Heinsohn

What you say about OOTP here is all wrong! Do some research! Pitchers in OOTP do NOT have a bell curve for control, they progress like in real life. Same as K's.

I'll take your word for it on this one.

Injuries are the same story... Pitchers in OOTP suffer almost as many injuries than hitters, the injury system was built using data from Baseball Prospectus' Will Carrol, and it matches his data perfectly fine.

But it doesn't match CBS Sportsline's injury data, not even close. Since I don't have access to Will Carroll's work, I have to go with what I do have access to. The more serious injuries are semi-close (although there is no such thing as a Tommy John surgery in OOTP, the kind of injury that causes a pitcher to miss 18 months but often leaves him coming back better than he was prior to the surgery) but overall actual injuries are somewhere around 25% of what they are IRL, and my experience is that they're nowhere near 55/45 hitter/pitcher. Injury_log has done some good studies on this already in another thread so I won't repeat them here. I've looked at leagues myself and found that OOTP *never* goes as deep into its pool of starters as real-life baseball does, and a large part of this I think is due to injury or lack thereof. You can take a year at random, figure out the average number of starts for the top 4 or 5 starters, and OOTP's average top 4 or 5 will *always* be higher. Whether it's the 1st year of the sim or the 25th, whether it's historical players or fictional, no matter what OOTP will *always* give more starts to the top 4-5 than real-life managers do.

But getting back to the pitchers... the game produces guys with 600 decisions and losing records. Certainly you can agree that this is an issue.

[Markus Heinsohn]

Originally Posted by Syd Thrift

But getting back to the pitchers... the game produces guys with 600 decisions and losing records. Certainly you can agree that this is an issue.

I've never seen that in my tests to be honest, and the distribution of starts is pretty realistic too.
EDIT: I just ran a test (16 teams league), and 108 pitchers started more than 5 games in this league. 2008 real NL? 114. Seems pretty close if you ask me. Now, in the NL no pitcher started more than 34 games, but that is pretty unusual, in OOTP the high was 37 (by 2 pitchers), but that is fine since the game models the 1990+ ERA, and a few pitchers have started 37 since then. Usually the high in OOTP is 36.
EDIT2: I did let the league continue for 30 years, and the guy with the most decisions and a losing record was some dude who had a record of 214-215... reminds me of Charlie Hough (216-216). So, I have no idea with which settings you got 300-300 pitchers...

As far as the injuries go, please keep in mind that freely available data is not really accurate, as reoccuring injuries are often counted twice, and also players leaving games and then playing the next day (doesn't happen in OOTP) are counted too. The data from BP was definitely different, and I'd rather trust Will Carroll than freely available injury reports.

[injury log]

Originally Posted by Markus Heinsohn

As far as the injuries go, please keep in mind that freely available data is not really accurate, as reoccuring injuries are often counted twice, and also players leaving games and then playing the next day (doesn't happen in OOTP) are counted too. The data from BP was definitely different, and I'd rather trust Will Carroll than freely available injury reports.

Players do, of course, leave games in OOTP and then come back the next day- when they sustain day-to-day injuries. That's a good thing, and realistic too.

I compiled a spreadsheet of every injury that occurred during the real life 2007 MLB season. I don't really see how my data can be disputed; for each injury I've recorded the date of occurrence and the player name, and you can verify each entry by checking news reports. I was surprised by some of the results. For example, there are about 7 injuries per day in real life MLB (well over 1000 per season), 65% of which are short term (1-14 days) injuries- roughly 850 short term injuries per year. That includes illnesses (which are much more common in real life than in OOTP) and certain off-field issues, like bereavement leave, which is also very common. And nearly 75% of long term injuries (> 3 months) are suffered by pitchers.

My data matches Will Carroll's data for medium and long-term injuries almost exactly, but as has been discussed before, Will Carroll did not provide any data about short term injuries; he just gave a guess. I have compiled that data, and it's not based on guesswork but on real-life injury information.

Injuries are a complex issue, and OOTP does fairly well with them, but there's certainly room for improvement. I'm hoping we can talk about how to make this aspect of the game better for OOTP-10. We had some good ideas in beta this year, and I'm hoping to find some time to write up a proper design document- some simple changes could make quite a big difference, I think.

Markus- I'm happy to send you the spreadsheet, but it's too big (2.1 megs) to send through this forum.

[Markus Heinsohn]

Originally Posted by injury log

Players do, of course, leave games in OOTP and then come back the next day- when they sustain day-to-day injuries. That's a good thing, and realistic too.

I compiled a spreadsheet of every injury that occurred during the real life 2007 MLB season. I don't really see how my data can be disputed; for each injury I've recorded the date of occurrence and the player name, and you can verify each entry by checking news reports. I was surprised by some of the results. For example, there are about 7 injuries per day in real life MLB (well over 1000 per season), 65% of which are short term (1-14 days) injuries- roughly 850 short term injuries per year. That includes illnesses (which are much more common in real life than in OOTP) and certain off-field issues, like bereavement leave, which is also very common. And nearly 75% of long term injuries (> 3 months) are suffered by pitchers.

My data matches Will Carroll's data for medium and long-term injuries almost exactly, but as has been discussed before, Will Carroll did not provide any data about short term injuries; he just gave a guess. I have compiled that data, and it's not based on guesswork but on real-life injury information.

Injuries are a complex issue, and OOTP does fairly well with them, but there's certainly room for improvement. I'm hoping we can talk about how to make this aspect of the game better for OOTP-10. We had some good ideas in beta this year, and I'm hoping to find some time to write up a proper design document- some simple changes could make quite a big difference, I think.

Markus- I'm happy to send you the spreadsheet, but it's too big (2.1 megs) to send through this forum.

I was talking about 0-day injuries basically... Anyway, feel free to email me the spreadsheet! I'll have another look...

[phillosopherp]

Originally Posted by Syd Thrift

For all the talk of homeruns and such, IIRC RonCo's verdict on the hitting aging model in OOTP was that it was okay. It could definitely be better but it's a lot better than it used to be. The real issue was - and, frankly, is - with pitching. Pitchers still have a general bell curve to their quality throughout a career even though control, for example, starts at one level and moves up throughout a player's career on average (whereas stuff, as measured by K/9, starts relatively high, might move up a little bit, but then starts a long, slow descent). Additionally, pitchers are *waaaaaaaaaaaay* too consistent from one year to the next. I think a realistic injury system would help with this, particularly if there were a bunch more day to day injuries (and that were properly doled out; IRL pitchers suffer around half of all injuries whereas in the game it's more like 25%), but something large needs to be done.

And this is the kind of thing you can see just by looking at the leaderboards. You do not see guys go 300-301 over the course of their career because .500 pitchers just don't have that many seasons where they're good enough to help a rotation. They'll have, say, 5 .500 type seasons, a .600 season, and a .400 season that ends their career as a starter. If they're lucky that .400 season comes last and they have a career like Erik Hanson. If they're not, that bad year comes early and nobody remembers them (even to the extent people remember Erik Hanson).

In this case, I think the AI is partially at fault as well. In the game, each team chooses the 5 best starters and puts them in the rotation. Okay, sometimes a manager will have bad information and won't *literally* do this but it will always try to. IRL a manager of a bad team will try out guys who are probably worse than the top 5 but might not be because they're old or had weird careers. My Mariners, for example, probably had 10 guys in the system who could start more effectively than R.A. Dickey but what the heck? It wasn't like giving Carlos Silva a couple extra GS was going to win them the World Series, and there was always that chance that Dickey's big 2007 in the PCL might have meant something.

One effect of the above is that IMO the game gives mediocre old pitchers more opportunities than they get in the actual bigs. To a point, I think the game *should* give guys with big-time primes more chances at starting, but in real life if I'm a GM and can choose between a guy who has a long history of being mediocre and a kid who is probably bad but might have a chance at being average, who am I going to choose? If I'm competing, the known mediocrity of course; then again, if I'm competing I'm probably trying to fill the role with something better than a known mediocrity via trade or whatever. If I'm not competing, I'm taking the kid 100% of the time.

Anyway, I don't think that the game is served well either by people saying they're going to stop playing unless they get exactly what they want, or for that matter people responding to every criticism as though they're the member of the opposite political party or something. It is vital that criticism be heard because in the absence of a decent competitor this is probably the only way Markus understands where the game needs revision; at the same time, you have to word it correctly because you're persuading and not debating.

Now this I can get on board with. Very good post!!!

[Childress]

Originally Posted by Markus Heinsohn

OOTP 9 reflects this...

Really? Now that's impressive!

[phillosopherp]

Originally Posted by injury log

Players do, of course, leave games in OOTP and then come back the next day- when they sustain day-to-day injuries. That's a good thing, and realistic too.

I compiled a spreadsheet of every injury that occurred during the real life 2007 MLB season. I don't really see how my data can be disputed; for each injury I've recorded the date of occurrence and the player name, and you can verify each entry by checking news reports. I was surprised by some of the results. For example, there are about 7 injuries per day in real life MLB (well over 1000 per season), 65% of which are short term (1-14 days) injuries- roughly 850 short term injuries per year. That includes illnesses (which are much more common in real life than in OOTP) and certain off-field issues, like bereavement leave, which is also very common. And nearly 75% of long term injuries (> 3 months) are suffered by pitchers.

My data matches Will Carroll's data for medium and long-term injuries almost exactly, but as has been discussed before, Will Carroll did not provide any data about short term injuries; he just gave a guess. I have compiled that data, and it's not based on guesswork but on real-life injury information.

Injuries are a complex issue, and OOTP does fairly well with them, but there's certainly room for improvement. I'm hoping we can talk about how to make this aspect of the game better for OOTP-10. We had some good ideas in beta this year, and I'm hoping to find some time to write up a proper design document- some simple changes could make quite a big difference, I think.

Markus- I'm happy to send you the spreadsheet, but it's too big (2.1 megs) to send through this forum.

Hate to double post but I didn't keep up as much as I would have liked in this thread. I like this data set, good job. The one thing that I have thought as I play game this year is, the injuries look a lot better but the fact that no one comes down with the common cold or has a parent die is strange. Personally I follow a lot of baseball guys both on and off the field and have noticed that this is actually the most common reason for the old more mid-career guys having days off. I have always wished this was in the game.

I look forward to seeing what next years injury file looks like.

Also Markus, don't let all this get under your skin mate, it seems that a lot of people might be carrying over a lot of this from previous years. Also I would like to point out to the others pointing to RonCo's numbers that Markus did show you the difference. That is an example of perspective getting in the way of a good understanding between folks.

[Markus Heinsohn]

Originally Posted by phillosopherp

Now this I can get on board with. Very good post!!!

Apart from the fact that my data and tests show that this post does not contain info that applies 100% to OOTP 9 See here: Player development in OOTP 9.2.2

[Markus Heinsohn]

Originally Posted by Childress

Really? Now that's impressive!

The whole development algorythm is impressive

[Markus Heinsohn]

Originally Posted by phillosopherp

Also Markus, don't let all this get under your skin mate, it seems that a lot of people might be carrying over a lot of this from previous years. Also I would like to point out to the others pointing to RonCo's numbers that Markus did show you the difference. That is an example of perspective getting in the way of a good understanding between folks.

Yeah, I think this sums up this entire thread quite nicely

[StyxNCa]

Originally Posted by Markus Heinsohn

I plan the following checkbox:
[x] Read the user's mind and perform exactly as he expects, no matter if realistic or not.

LOL

Excellent!! That would solve all problems!!

[DesertFox07]

Originally Posted by 1998 Yankees

I can attest to this as well.

I started the UABL in 1910 and am up to 2001 and thus far, my results have been completely inline with my expectations. I'm not against people like RonCo doing their thing, but I support Markus' hard work to get to this point. Thanks!

[evanbarth]

I don't mean to fan the flames, I have an MLB league in 2007 that is progressing into the future...which development codes should I use or does it even matter?

[Markus Heinsohn]

Originally Posted by evanbarth

I don't mean to fan the flames, I have an MLB league in 2007 that is progressing into the future...which development codes should I use or does it even matter?

You should be fine using the default ones