More realistic injury file

[Matter2003]

Originally Posted by Antonin

Some people in other threads have expressed concern over the number and severity of injuries in OOTPX.

But I'm confused about whether or not this new injury file is a viable replacement for the one currently included in the game. I appreciate the work that went into it, but I do know Satchel did some work on an injury file for OOTP9...

Try it out and judge for yourself. Backup your prior injury file, and then if you don't like it, you can always switch back. Every injury that has been updated/changed has been done so after researching the injury as it pertained to however many baseball players that had it I could find over the past few years, various sources talking about recovery lengths, etc...

[Matter2003]

Originally Posted by PhillieFever

Ah,BOO! I thought I figured it out but I guess not.I also had the number thing happen to me too,but I switched the file,didn't re-import and then it worked fine.Hopefully this dosen't become a major issue.

Hmm...I see the injury always seems to be listed in parenthesis when that happens and it shouldn't be. I am going to look into it...

[PotatoPeeler]

Originally Posted by Matter2003

I thought it was pretty clear from what I wrote that it is still being worked on. I'd say currently it's about 60% or so done...

I missed that for some reason. Sorry.

[Flying Dutchmen]

Originally Posted by Matter2003

In virtually all cases I have found baseball players having injury durations shorter and longer than the suggested time stated either by doctors or in the articles regarding them being injured(most times the articles state about how long the player is expected to be out). I am using modern day recovery times. Half of those injuries requiring complicated surgery(UCL, MCL, ACL, etc)would have been career ending injuries back in the day...

The purpose is to research the injuries listed and find a variety of players that have had the injury, and look at the lengths of time they have been out so there is a more complete profile of the injury. As I said previously, some of the injury min/max times were just ridiculous. Not every baseball player who has forearm inflammation is going to be out 119-120 days with no other option.

Since there seems to be a lot of talking past each other - I think I'd put something injury log said in another thread into this thread to clarify one big complaint that you have:

Originally Posted by injury log

If you think there is a problem with a duration in injuries.txt, name it, and I'll tell you which player was sidelined for that duration in real life. Some duration windows are short to make some diagnoses infrequent, as they are in real life. In past versions of the game, that was the best way to control the frequency of any individual injury - because the game chooses the duration before choosing the diagnosis, if a window was narrow, the injury would not often be used. With the frequency numbers, these narrow windows may be unnecessary, but I don't think anyone yet knows just what the frequency numbers in injuries.txt do.

The funky durations you're seeing is for this reason probably - not because they are "accurate". He was trying to control injury occurrence frequency as near as I can tell. I have seen no evidence that your edited file achieves the same thing but I'd be interested to see it if you have it.

[satchel]

Originally Posted by Matter2003

Not every baseball player who has forearm inflammation is going to be out 119-120 days with no other option.

Not every player who has forearm inflammation is going to be out 119-120 days; but every player, from a certain defined sample, who did have forearm inflammation, was out some specific number of days, which could have been 119-120.

That's why I asked whether this new injury file is meant to take injury durations from the "civilian" world, and transplant them into baseball, or whether it's meant to mirror the injury durations that pro players actually did experience at some certain time in the past.

Originally Posted by Antonin

...I do know Satchel did some work on an injury file for OOTP9...

I did some minor finishing work on a modded injuries.txt file for OOTP9 which Injury Log had authored. We offered that file to Markus, and now it's the injuries database for OOTPX (the game lists Injury Log, Ian, in its credits).

The methodology used in compiling that file is logical and clear. The design of this one is less so.

Originally Posted by Matter2003

...the injury as it pertained to however many baseball players that had it I could find over the past few years, various sources talking about recovery lengths, etc...

That's why I'd like to get a better definition of the parameters of Matter2003's file, such as its date range and its sources (there are some good, comprehensive online lists of player injuries). That way we can compare it with our current injuries database, with the goal of eventually making the best file possible, according to the criteria we set.

[satchel]

Originally Posted by Flying Dutchmen

He was trying to control injury occurrence frequency as near as I can tell.

The principal way to control injury frequency is by using the "Suspension, Injury & Fatigue Options" in "Game Options" to set the "Injury Frequency." Modding the injuries.txt file can affect the frequency of injuries, but it's a roundabout way to do it. What the injuries.tx file can affect is the frequency of particular injuries, relative to the other injuries. The title of this thread calls the new file "More realistic" and I'm curious as to whether it is, or not.

[Matter2003]

Originally Posted by satchel

That's why I'd like to get a better definition of the parameters of Matter2003's file, such as its date range and its sources (there are some good, comprehensive online lists of player injuries). That way we can compare it with our current injuries database, with the goal of eventually making the best file possible, according to the criteria we set.

I would say the ranges are probably the past 5 years or so. Sources are online articles regarding the player injury, fantasy baseball sites that talk about the player injuries like rotoworld, etc, descriptions of their injuries on the DL and following their injuries and the length of time from when it was initially reported to the time when they resumed playing...

For instance, under player news, it might have something like April 25th: Sam Maglie was injured in last night's game, suffering a shoulder injury. Tests need to be ran, but he is expected to be out 4-6 weeks with right shoulder inflammation.

Then 2 weeks later his status is updated and it says: May 8th, Maglie is undergoing treatment for his shoulder inflammation and is progressing nicely according to team sources.

etc,etc...these reports are pretty detailed and basically have dates when the player was injured, notes about his recovery/rehab, and the date when he returned to the lineup...pretty detailed stuff in a lot of cases...

The more cases within that time frame, the better--small number of cases means the injury file likely is not as realistic as it should be, because we are modeling for the minimum and maximum times a player could be injured with this injury---not the min/max time for a specific person having this injury...

I mean look at viral infection as an example---someone really thought it was a good idea to have a min time of 70 and a max time of 74 for a viral infection because only 1 person probably had this documented in baseball? If 1000 cases of a player having a viral infection were found, I am pretty sure someone having it last 70 days would be 1 case out of a thousand or perhaps 1 out of ten thousand...the majority would be a few days...so because this 1 case happened to happen during the timeframe you were looking for, I am to believe that every baseball player who ever suffers from a virus is going to miss 70 days? I am sorry, that is just beyond the realm of believability to me...

[Flying Dutchmen]

Originally Posted by satchel

The principal way to control injury frequency is by using the "Suspension, Injury & Fatigue Options" in "Game Options" to set the "Injury Frequency." Modding the injuries.txt file can affect the frequency of injuries, but it's a roundabout way to do it. What the injuries.tx file can affect is the frequency of particular injuries, relative to the other injuries. The title of this thread calls the new file "More realistic" and I'm curious as to whether it is, or not.

Yeah this was exactly what I meant - sorry the wording was awkward at best. I meant occurrence of particular injuries as compared to others, as you said.

[Matter2003]

Originally Posted by PhillieFever

Ah,BOO! I thought I figured it out but I guess not.I also had the number thing happen to me too,but I switched the file,didn't re-import and then it worked fine.Hopefully this dosen't become a major issue.

This only seems to be a problem in the news stories. The injuries are fine if you look at the injury log report or the under the knife report...

It appears the problem is in the news stories themselves...

[injury log]

Originally Posted by Matter2003

INot every baseball player who has forearm inflammation is going to be out 119-120 days with no other option.

Understand that the duration window is another injury frequency control. If the window is narrow, the game will rarely use the injury, since the game will only rarely decide that an injury should last 119 days. If you widen the duration windows for the rarest injuries, you'll find them occurring more often than is realistic in your game.

[injury log]

Originally Posted by Matter2003

I mean look at viral infection as an example---someone really thought it was a good idea to have a min time of 70 and a max time of 74 for a viral infection because only 1 person probably had this documented in baseball? If 1000 cases of a player having a viral infection were found, I am pretty sure someone having it last 70 days would be 1 case out of a thousand or perhaps 1 out of ten thousand...the majority would be a few days...so because this 1 case happened to happen during the timeframe you were looking for, I am to believe that every baseball player who ever suffers from a virus is going to miss 70 days? I am sorry, that is just beyond the realm of believability to me...

I was trying to figure out what you were talking about - 'viral infection' in injuries.txt has a window of 2-9 days. Instead you mean the 'intestinal virus' diagnosis, which has a 70-74 day window. Note that the game already has several 'stomach virus' and related diagnoses; the severe version of these in OOTP is 'intestinal virus'. Since it's surely very rare that a real life baseball player is sidelined for an extended period with a similar diagnosis, the duration range is narrow. The duration is based on the time Justin Speier missed in '07.

[PhillieFever]

Originally Posted by Matter2003

This only seems to be a problem in the news stories. The injuries are fine if you look at the injury log report or the under the knife report...

It appears the problem is in the news stories themselves...

Have you recreated it then? I'm curious because if you have,then I can report this the Markus to get it squashed ASAP.

[injury log]

And just a few comments on the philosophy behind the file posted in this thread:

-it is not a good idea to identify whether an arm injury is to the right or left arm. A pitcher sustaining an arm injury in-game will almost always suffer that injury to his throwing arm, and the game can't distinguish, by looking at the lines in injuries.txt, which arm that is. Further, pitchers can lose ability because of injury in the game, and that only makes much sense when injury affects their pitching arm;

-separating the 'fractured rib' diagnosis into multiple diagnoses, each of which identifies the specific rib that has been fractured, will make the fractured rib diagnosis appear 24 times more often, all other things left equal. There was only one player sidelined in 2007 with a fractured rib (Jerry Hairston), so it is not a common diagnosis;

-some diagnoses that are used in real life do not appear in the file because they are duplicative. 'Intercostal Strain' is one example; the file already includes overlapping diagnoses that are more often used in real life;

-there's certainly scope to improve the frequency numbers in the file. It's time consuming to do, unfortunately, since so many factors interact to determine how often each injury occurs - the number of different play types on which the injury can happen, the range of durations the injury is permitted to have, and the frequency value in the injury file. At least now, with the frequency value in the file, it's possible to make this aspect better, but I'd be nearly certain it's not all that close at the moment (though much better than in past versions of the game).

-someone complained above about the durations 'hamstring strains' are permitted to have in the game. In real life, for injuries lasting about one month, 'hamstring strain' is *by far* the most common diagnosis in baseball - it's not even close. Phil Hughes, Cody Ross, Endy Chavez and Dave Dellucci were all sidelined for between 50 and 90 days with hamstring strains in 2007, and Rocco Baldelli missed five months, though his circumstances were unique. The max value for one of the two severe hamstring strain diagnoses in the file should be lowered - that's been logged as an issue.

[Matter2003]

Originally Posted by PhillieFever

Have you recreated it then? I'm curious because if you have,then I can report this the Markus to get it squashed ASAP.

Yes...in news stories many times you will see a number listed(appears to be the injury frequency perhaps?) as the injury. However, when you go in and look at the injury log report or the under the knife report, the injuries report normally. Also sometimes on the player history screen, you will see an injury listed in parenthesis when no parenthesis is in the file...for instance something like (strained right triceps) when it is actually strained right triceps.

[Matter2003]

Originally Posted by injury log

And just a few comments on the philosophy behind the file posted in this thread:

-it is not a good idea to identify whether an arm injury is to the right or left arm. A pitcher sustaining an arm injury in-game will almost always suffer that injury to his throwing arm, and the game can't distinguish, by looking at the lines in injuries.txt, which arm that is. Further, pitchers can lose ability because of injury in the game, and that only makes much sense when injury affects their pitching arm;

-separating the 'fractured rib' diagnosis into multiple diagnoses, each of which identifies the specific rib that has been fractured, will make the fractured rib diagnosis appear 24 times more often, all other things left equal. There was only one player sidelined in 2007 with a fractured rib (Jerry Hairston), so it is not a common diagnosis;

-some diagnoses that are used in real life do not appear in the file because they are duplicative. 'Intercostal Strain' is one example; the file already includes overlapping diagnoses that are more often used in real life;

-there's certainly scope to improve the frequency numbers in the file. It's time consuming to do, unfortunately, since so many factors interact to determine how often each injury occurs - the number of different play types on which the injury can happen, the range of durations the injury is permitted to have, and the frequency value in the injury file. At least now, with the frequency value in the file, it's possible to make this aspect better, but I'd be nearly certain it's not all that close at the moment (though much better than in past versions of the game).

-someone complained above about the durations 'hamstring strains' are permitted to have in the game. In real life, for injuries lasting about one month, 'hamstring strain' is *by far* the most common diagnosis in baseball - it's not even close. Phil Hughes, Cody Ross, Endy Chavez and Dave Dellucci were all sidelined for between 50 and 90 days with hamstring strains in 2007, and Rocco Baldelli missed five months, though his circumstances were unique. The max value for one of the two severe hamstring strain diagnoses in the file should be lowered - that's been logged as an issue.

1) Yes, I just realized that as I looked at a pitcher with an arm problem...will have to correct in next release by removing the arm descriptors for certain injuries...

2) I will be further balancing the file by adding many more variations of the more common injuries to rebalance the file...

[PhillieFever]

Originally Posted by injury log

And just a few comments on the philosophy behind the file posted in this thread:

-it is not a good idea to identify whether an arm injury is to the right or left arm. A pitcher sustaining an arm injury in-game will almost always suffer that injury to his throwing arm, and the game can't distinguish, by looking at the lines in injuries.txt, which arm that is. Further, pitchers can lose ability because of injury in the game, and that only makes much sense when injury affects their pitching arm;

-separating the 'fractured rib' diagnosis into multiple diagnoses, each of which identifies the specific rib that has been fractured, will make the fractured rib diagnosis appear 24 times more often, all other things left equal. There was only one player sidelined in 2007 with a fractured rib (Jerry Hairston), so it is not a common diagnosis;

-some diagnoses that are used in real life do not appear in the file because they are duplicative. 'Intercostal Strain' is one example; the file already includes overlapping diagnoses that are more often used in real life;

-there's certainly scope to improve the frequency numbers in the file. It's time consuming to do, unfortunately, since so many factors interact to determine how often each injury occurs - the number of different play types on which the injury can happen, the range of durations the injury is permitted to have, and the frequency value in the injury file. At least now, with the frequency value in the file, it's possible to make this aspect better, but I'd be nearly certain it's not all that close at the moment (though much better than in past versions of the game).

-someone complained above about the durations 'hamstring strains' are permitted to have in the game. In real life, for injuries lasting about one month, 'hamstring strain' is *by far* the most common diagnosis in baseball - it's not even close. Phil Hughes, Cody Ross, Endy Chavez and Dave Dellucci were all sidelined for between 50 and 90 days with hamstring strains in 2007, and Rocco Baldelli missed five months, though his circumstances were unique. The max value for one of the two severe hamstring strain diagnoses in the file should be lowered - that's been logged as an issue.

While I believe the new file can with some work be very good,I have to agree with IL here on all of his points,I've ran a few long term tests and all of these issues took prevelance.

[Matter2003]

Originally Posted by PhillieFever

While I believe the new file can with some work be very good,I have to agree with IL here on all of his points,I've ran a few long term tests and all of these issues took prevelance.

I am working on some frequency balancing right now that has improved things quite a bit in regards to injury occurrence. I can definitively say that my file does produce more injuries(mostly of the DTD and under 7 day variety) which is a good thing because the default file was still not producing enough. On very high settings, my file produced 1597 injuries versus 1291 with the default. Since the realistic modern day injury setting should produce around 1350 or so, this should be pretty on target with that number, so that definitely is a good thing....

[PhillieFever]

That's great news M.I really think this file can be really sweet as long as you keep getting the cooperation that so far has really helped shape this project up.

[hmatthias]

Figured out a small work around to the "number in news story" problem.
If you import the file, then save and restart the program, all is well. Looks like something doesn't get reset when the injury file re-imports.

[Rabbott]

i still think this new file even thou needs some tweaks is way more playable then the default one! it was getting tiring to see everyday another player go out with an injury for 2 months or longer almost every single day yet very little day to day injuries



thanks again keep up the good work