Issue - Injury clusters

[Cleasby]

This version from testing 10+seasons appears to be assigning injury clusters far more than previous versions.

What I mean is injuries to the same positions on the same team versus a more random assignment.

I have the injury setting on normal and to give you a flavour of what I am seeing here is just one seasons snapshot but this seems to be a trend for almost every season I have tested in 22.







I think the method for assigning injuries needs to have a review as this clearly looks more patterned than rng to determine injuries.

[slybelle1]

I tend to agree that the injury system can be improved. I see players with Achille injuries, torn acl, and they are out between 8-16 weeks. Those are typically season ending injuries and sometimes into the next season. Not sure that I have seen any injury length greater than 12-16.

[KW77]

adding on to say this isn't a new issue, happens pretty often in 21 as well.

[brooks_piggott]

I'm not sure what the ask is here. The injuries are not random... they are tied to attributes and other things like tackling or possessing the ball. I can look at extending the time period for some of them to get achilles injuries longer... especially now that we're in an 18 week season instead of 17.

[slybelle1]

I'm not sure what the ask is here. The injuries are not random... they are tied to attributes and other things like tackling or possessing the ball. I can look at extending the time period for some of them to get achilles injuries longer... especially now that we're in an 18 week season instead of 17.

My response is probably different than original poster, but since you asked.

- More realistic injuries (too many fractured skulls, broken skulls, Migraines, etc.)
- More realistic durations like my previous examples. Even adding more weeks might not be enough for example on torn ACL, if player gets hurt in week 1, I don't see many reports in real life suggest the player can come back in 12-16 weeks. It's pretty much a season ending injury with really no option to return in the season.

As for randomness, I was searching for info on how to modify the injuries.xml file and came upon this exchange, which did seem to imply some randomness in their assignment. So although maybe not random when an injury chance occurs, does seem random on how the injury type is determined.


I wouldn't mind making my own injury file, but not sure it really allows for the flexibility needed to make them more realistic. Right now seems you can have min/max severity and whether on or off the field. Don't believe we can control frequency and under what conditions they would come up realistically. Even with the min/max severity, not sure we can directly impact the number of days/weeks as I believe it caps at 16 weeks and even two different injuries with the same numbers still comes up differently within the game,

[zac]

Think post just saying it feels like some sort of code or unplanned element of the injury system that leads to a large % of injuries occurring to the same team or position group on one team. Could be a bias in observation on our end that we notice when injuries happen in bulk but it could be explained by bad luck maybe but sometimes does feel off.

At large, hoping the injury system overall can be improved given how injuries impact football. Maybe tie injury system into medical flags for draft prospects? Would be cool to weigh drafting a elite prospect with questionable injury history in college for example.

Injury risk flags or injury history more visible on player pages would be helpful. Would be good maybe to tie major injuries into occasional large rating drops (e.g., RB never the same after a big knee injury so rating drops 10 pts after coming back from injury).

[brooks_piggott]

Which injury is selected is random, but which player gets injured is driven by other factors. I don't have an injury.xml file per position if that makes sense. But we certainly can update the injury.xml file to have "weights" so that you see the more common ones a bit more. If you have data you've come across let me know... it would save me time having to look it up.

[slybelle1]

Which injury is selected is random, but which player gets injured is driven by other factors. I don't have an injury.xml file per position if that makes sense. But we certainly can update the injury.xml file to have "weights" so that you see the more common ones a bit more. If you have data you've come across let me know... it would save me time having to look it up.

Thanks, that would be helpful. Is there any way to have a parameter as well which would better influence the duration in terms of days/weeks? Unless the min/max already does that, but would require some explanation. Increasing max length would be great too for long term injuries. I have seen injury types have the same min/max value, but ultimately have different range of weeks.

Some possible injury sources

https://www.footballoutsiders.com/stat- ... n-position (I like this one as it shows by position. I know you can't make that change but should give you some ideas on frequency)
https://www.nfl.com/playerhealthandsafe ... njury-data

[KW77]

I'm not sure what the ask is here. The injuries are not random... they are tied to attributes and other things like tackling or possessing the ball. I can look at extending the time period for some of them to get achilles injuries longer... especially now that we're in an 18 week season instead of 17.

the point was that we've observed a weird amount of times where several players at the same position go down at the same time.

are you saying that there's a possibility that the opposing player(s) they are getting matched up with can just turn into a wrecking ball and injure 3 players at a position in the same game?

Understand that it's a really really hard thing to test but it's something we've observed for seasons in PF21, even PF20 if I remember correctly.

[brooks_piggott]

Certain positions get injured more because they get tackled more or are on the field more. And a slew of other factors in the engine. But not because we're targeting positions or giving some super powers to defenders.