Crazy Injury Numbers?

[Assos]

Want to talk about playing through injuries, talk to https://twitter.com/oldhossradbourn He does a good job of telling you as it is!

Think about all of the pitchers that are down this year already. Here is the injury list so far on MLB

Fantasy Baseball Player Injury Updates | MLB.com: Fantasy

Game is fairly realistic. Sometimes your iron men get freak injuries.

[Markus Heinsohn]

Originally Posted by pucku33

If you don't want realistic injuries, just edit the file and be done with it.

Or simply lower the injury frequency. Game Menu -> Game Settings -> Players & Facegen tab -> Injury Frequency, left side of screen.

[injury log]

Originally Posted by majesty95

I think I've identified the problem.

I opened the injuries.txt file and there are some major injuries (specifically pitcher injuries) that are given a "4" or "5" or "common" frequency. Many of them (Torn PCL, UCL, labrum, elbow ligament, etc) are pitching related. Would explain why I lost 3 of my top 4 SP and my top P prospect to injuries of 5 months or more in season one. Seems that this may need to be addressed...

That's not how things work. Those frequency numbers in the injury database are all relative. If you had only two arm injuries in there, and one had a '5' frequency, and one had a '1' frequency, that would just mean that when OOTP wants a guy to have an arm injury, it will almost always use the '5' diagnosis. But changing those numbers doesn't change how many injuries you get. If you want to change the number of injuries you have, adjust the injury frequency in Game Setup.

There are ways to adjust the balance of long and short injuries by editing injuries.txt as well, but you'd need to know how the injury system works. I can explain if anyone's really interested in doing that, but only if that's the case, because it would take me a while!

[majesty95]

I think you guys are missing the point of what I am saying. I'm not complaining about the amount of injuries. I am saying that the occurrence of serious injuries, especially to pitchers, is too high.

I have no problem with having injuries. I understand that the DL is used on average about 15 times per year, per team in real life. I'm ok with that. My issue is with the rate at which pitchers are going down in the game for long periods of time

Here's my data:

In 2014, there were 476 instances of players being put on the DL (roughly 15 per team). 30% of all injuries were 60 days plus and 19% were 90 days or more.The game mimics these numbers fairly well.

My issue is in the occurrence of significant elbow and shoulder injuries (specifically to pitchers).

In 2014, 3 of 70 elbow injuries were UCL tears. In my game (just past mid-season) 3 of 23 elbow injuries have been UCL tears. In 2014, 0 of 70 elbow injuries were elbow fractures. In my game, 3 of 23 have been elbow fractures. In 2014, 1 of 70 elbow injuries were bone spurs or bone chips. In my game, 3 of 23 elbow injuries were bone spurs or bone chips. All of these instances are extremely high compared to real life.

The oddity was, however, that 0 of 23 elbow injuries were Tommy John surgeries whereas 20 of the 70 elbow injuries in 2014 were Tommy John. That did balance out the numbers somewhat. However, significant elbow injuries in my game were still about 20% higher than in 2014.

The real issue might be in shoulder injuries though.

In 2014 there were 63 shoulder injuries, 14 of which (22.2%) were 90 days plus. In my game, 10 of 31 shoulder injuries were 90 days or more (32.2%). That's almost 50% higher!

In 2014, 5 of 20 (25%) cases of shoulder inflammation were 90 days or more. In my games, 4 of 5 (80%) were 90 days plus. Small sample size but more than 3x as high! In 2014, 6 of 63 (9.5%) shoulder injuries were "surgeries" (ie possible rotator cuff/labrum injuries). In my game, 7 of 31 (22.6%) shoulder injuries were torn rotator cuffs or torn labrums. That's more than twice as high!

Significant elbow injuries to pitchers were definitely high (about 20%). However, significant shoulder injuries were off the charts. They were nearly 2.5 times as high as real life.

So, I think there is an issue here. Not in the overall amount of injuries or even in the amount of long-term injuries. But certainly in the amount of long-term injuries to pitchers.

[Syd Thrift]

Yeah, I'm sorry but the sample size is downright tiny. I'm not dismissing the idea that OOTP could have an issue; I'm saying that if you want to prove this, you need more than "3 of 22 injuries were really bad".

[majesty95]

Originally Posted by Syd Thrift

Yeah, I'm sorry but the sample size is downright tiny. I'm not dismissing the idea that OOTP could have an issue; I'm saying that if you want to prove this, you need more than "3 of 22 injuries were really bad".

I didn't say "3 of 23 injuries were really bad". I said I spent several hours compiling and comparing data, and even in a small sample size, the significant injuries to pitchers, especially shoulder injuries, are SIGNIFICANTLY higher than real life.

I also said earlier that some of the frequencies in the injury.txt file were higher than real life and that was confirmed by my data.

I'm not accusing anyone of being incompetent or not wanting it to be as realistic as possible. I'm simply saying that based on my research into this that there is an issue with long-term injuries to pitchers. Whether that be because of an inadvertent input error or outdated data I do not know. I just wanted to point it out and hopefully get it corrected. If they choose not then so be it. But I would hope that they would since the best possible simulation of real life is always their goal.

[Doc_Brown]

Look at the date of this article Ivan Nova To Undergo Tommy John Surgery ? MLB Trade Rumors.

When it was released, he was already the 15th pitcher to have TJS in the very early stages of the season. I'm guessing a few more had TJS after this article.

[injury log]

A few things:

• OOTP injury frequency is based on a much larger sample of data than what you've looked at; it's based on a database of more than 1000 injuries. The last time I checked (which was last year if I remember correctly), the frequency of long pitcher injuries was slightly too low, and the frequency of long batter injuries was too high. That should still be true, since I don't think anything was changed, code-wise.

• I don't understand some of the distinctions you're drawing when you subdivide your data. You seem to be regarding 'torn UCL' and 'Tommy John surgery' as different injuries, when they're the same. OOTP often uses different names for the same injury, because that helps to make certain injuries more common, for reasons I could explain but the explanation would be boring.

• within the injury model in OOTP, it's extremely difficult to control precisely the frequency of individual injury types. That's just because of the injury architecture OOTP uses now; it's not designed to do that. Really the best we can hope for is to get the right distribution of short and long injuries, and the right distribution by body part. But when the game wants to give someone a leg injury, it's basically impossible to make sure it picks 'strained hamstring' the exactly correct percentage of the time.

•*OOTP doesn't distinguish among types of arm injury now. A shoulder injury has the same effects as an elbow injury. So if you're finding that the proportion of shoulder injuries is too high relative to the proportion of elbow injuries, that's only a cosmetic difference. It doesn't have any effect on game results. I'd like OOTP to properly account for the differing effects of shoulder and elbow injuries, though we'd need a bit more data on how those types of injury affect player skill (numerical data, not just the anecdotal observation that 'shoulder injuries lower velocity, elbow injuries hurt control'). OOTP does, however, use real life data to determine how, and how often, a pitcher's ability will be negatively affected by a serious arm injury; it just groups elbow and shoulder injuries together.

[majesty95]

Injury, I was not as concerned about elbow injuries after my analysis. I still think they are a little on the high side. ~20% still seems high even on a smaller sample size. However, I was REALLY concerned about the shoulder injuries.

Significant shoulder injuries being nearly 2.5 times real life, even over a half season, is still high. It also allows for situations like I had where 4 or my top organizational pitchers all went down with season ending injuries in the first three months of the season. I don't think that's ever happened in real life.

I understand that it is a statistical random number generator essentially. Or a dice roll if you will. I just think that in the injury.txt file that the occurrence probability of 120 day plus elbow and (specifically shoulder) injuries are too frequent. I think some of those 4's and 5's should probably be 3's which would balance it out more. On the other hand, I think there are some 1's and 2's (5-30 day injuries) that should probably be 3's.

Again, I'm not asking for an overhaul of the system. But the way the injuries.txt file is constructed I think allows for some statistical irregularities that could be avoided with some minor adjustments. If you need lengthy statistical data to make that change I just am not willing to compile it. I'd rather it be done for the sake of everyone else but I'm not the guy to take it any further than I have. I've edited my file to make it more inline with what I believe is realistic and that should suit me as an individual fine in my games (at least I hope it does lol).

[injury log]

Originally Posted by majesty95

I think some of those 4's and 5's should probably be 3's which would balance it out more. On the other hand, I think there are some 1's and 2's (5-30 day injuries) that should probably be 3's.

That's not how the model works. The game decides a player should be injured, then the game decides how long the injury will last. Only then does it look at the injury file. It finds a diagnosis that is permitted to last for the correct duration, and permitted to occur on the correct type of play. If it finds one or more diagnoses, it chooses among them using the frequency numbers. If it finds no suitable diagnosis, the injury does not happen.

So if you lower the frequency numbers for some of your long-duration injuries, you won't get fewer long-duration injuries. You'll just make those specific diagnoses occur less often, relative to the other long-duration diagnoses. To actually reduce the number of long injuries (only) you have to make it so the game just doesn't find a single suitable diagnosis sometimes when it looks at the injury file. So if you make it so no injury ever can last for exactly 255 days, then every time the game specifically wants to injure someone for 255 days, then no injury will happen. Or if you remove all the long-duration 'when Hit By Pitch' injuries from the file, you'll get fewer long injuries, because no one will ever sustain a long injury when hit by a pitch. But changing only the frequency values won't accomplish your objective.

[majesty95]

Ok. That makes sense. So how do we adjust it so that there are fewer 120+ day injuries and fewer shoulder (and to some extent elbow) injuries? Assuming one does not like my data and feels these injuries are too high, how do they go about correcting it?

[majesty95]

Anything? I'd like to modify this if possible. I'm convinced this is out of whack even if others aren't...

[Assos]

Play on a lower injury setting.
I still do not think this is out of whack, I just think you are having bad luck, like the 2014 and 2015 Texas Rangers.

[Syd Thrift]

There are 4 different ways if memory serves that a player can get hurt. If you remove all season ending injuries for one of those injury varieties you've effectively lowered the season ending injury rates by... not really 25% because the injuries occur at different frequencies, but by some amount whose frequency you can test by playing out 20 or 30 seasons. Note that if you leave one item in for season-enders, regardless of its frequency, it will occur every single time the game decides it needs to generate an injury of that severity. Career ending injuries work the same way.

[evanbarth]

Is the "high" setting the closest to realistic number of injuries?

[majesty95]

Originally Posted by evanbarth

Is the "high" setting the closest to realistic number of injuries?

Supposedly. The overall numbers may be right but you get very few short-term or day-to-day injuries and too many (IMO) several week to several month injuries. I've been playing on normal and enjoy it more even though I still don't think there are enough short-term injuries vs long-term...

[ra7c7er]

Personally I don't mind the way the game does injuries. I rather deal with fewer injuries overall and have a statistically higher then real life amount of long term injuries then constantly having to adjust everything every time someone gets a blister because they are out for 10 days.

One thing I've seen is that with early seasons in new games injuries happen more and they tend to be more severe. My guess is that it has to do with the fact that there are no injuries in they system and carry over injuries from prior years. After 22 test leagues I've done so far in OOTP16 I've seen roughly double the sever injuries in the first three seasons then I've seen in seasons after. Not just my team either but overall. And I saw this also with OOTP15.

I always take frequencies of happenings in stats based RNG games with a grain of salt. These games aren't supposed to copy real life. They are based on real life. Even the best soccer stats games (which are the top tier of the stats game world) don't do better then average job of mimicking real life. There is a reason why they most accurate programs used to predict the outcomes of real life sports are made by geniuses, run on super computers, and aren't games.

It's a game it's not real life. adjust the modifiers to get what you want and/or get what you think is an accurate indication of real life. Because regardless what "real life" actually is. Very few here will agree on what it takes in game to achieve that.

[majesty95]

That's a good point. Dealing with minor injuries all of the time could be tedious. So there is a definite positive to the current setup. I guess, as with anything in this game, it all depends on your perspective and personal preference.

[ra7c7er]

Originally Posted by majesty95

That's a good point. Dealing with minor injuries all of the time could be tedious. So there is a definite positive to the current setup. I guess, as with anything in this game, it all depends on your perspective and personal preference.

Exactly

---

BTW just counted the current number of players on the DL or still currently haven't returned from MLB.com it's 177 players (plus or minus a handful I might have missed while scrolling down). Outside of some of the hardcore MLB quickstart guys I don't think anyone wants their game bogged down with that many injuries.

[Thundercrack]

How do you re import the injury file, I've forgotten how to do that