The new user interface is in preview!

Want to check it out? Click here! (If you don't like it, you can still switch back)

The draft is underway!

Click here to go to your war room, or visit the war room item in the draft menu.

League Forums

Main - Community Help Forum

player weights

By Dogboyuk75
4/02/2020 8:23 am
hey guys i understand the principles behind player weights and have read sethericks and others player weights suggestions.is there anywhere this is explained in more detail?? for example why have primary skill at 100 secondary at like 18 and other at 1.why not have all relevant skills at 100??and why use these particular numbers ??from what ive gathered setherick is the man will all things MFN but im just trying to understand the theories/reasoning behind figures .thanks in advance guys.stay safe :)

Re: player weights

By Bexus76
4/02/2020 10:37 am
Because a QB for instance needs to have accuracy. The importance of accuracy and other attributes are not even close to equal. If Acc and FOV are both set at 100, then the ACC rating makes up for 50% of that total.
If ACC and 9 other attributes are all set at 100, then ACC only makes up for 10% of that players overall rating. If you set ten different attributes at a 100 weight, in theory, if ACC was zero and all other attributes were at 100, you would have a completely unusable QB with a 90 OVR rating.

Re: player weights

By Bexus76
4/02/2020 10:41 am
So, if you want a QB's accuracy rating to make up 50% of his overall rating, set Acc to 100, and all other attributes to a total of 100, so around 11 points each for 9 other attributes, or 20 each if you only want to factor in 5 other attributes.

Note: in the current code as I understand it, each position has one defining trait that dominates all others. Usually speed, strength, or in QB's cases, accuracy.
Last edited at 4/02/2020 10:44 am

Re: player weights

By Dogboyuk75
4/02/2020 10:56 am
Ah got ya mate sort of understood main skill versus lesser skills .but didn’t understand lesser skills gotta total 100 as well.appreciate the clarification buddy

Re: player weights

By Bexus76
4/02/2020 11:06 am
They dont HAVE to total 100. I was just using that as an example for IF you wat your QB's accuracy to count for HALF of his overall rating. It's not a bad starting point though. Setherick put a lot of time into figuring out how much each rating is worth and factoring out those weights. I'd start with his, see how it works for you, and tweak it from there.

Re: player weights

By slowtospeak
4/13/2020 7:58 pm
If the secondary ratings were greater, it would make the main ratings less prominent. It's not that the secondary ratings don't count; it's just that the main rating is really, really important.

Re: player weights

By JasperFrost
4/19/2020 10:09 pm
setherick also assumes a parabolic skill distribution which would mean everything not top level is irrelevant as long as the skill is decent

Re: player weights

By slowtospeak
4/22/2020 2:14 pm
Even though the *behavior* is parabolic now, some of the ratings *themselves* are still linear. For example, you need a secondary attribute to be at least 21 in order to make the sortable position lists on the team home page, no matter what anything else is.

Also, there is a hidden "experience" variable which shows up more often if you are not using lots of other attributes. If a returner or long snapper has been terrible for a long time, he might show up on the list even if you don't want him to.

Re: player weights

By JasperFrost
4/22/2020 8:36 pm
I feel like I'm missing the point of your comment

Re: player weights

By setherick
4/23/2020 6:58 am
slowtospeak wrote:
Also, there is a hidden "experience" variable which shows up more often if you are not using lots of other attributes. If a returner or long snapper has been terrible for a long time, he might show up on the list even if you don't want him to.


Where is JDB's confirmation about this? I know that this was something that (I think) Bryson10 was convinced about, but I've never seen a confirmation of it?

There are "experience" attributes that are not hidden. One is positional experience, and the other is play experience. Positional experience is basically ignorable. Players get it for playing in games at a specific position. Your WRs playing RBs will almost max this out, but never be at 100 because they train/practice with the WRs for instance. But for the most part, it doesn't affect game play.

Play experience does affect how your players play, but is mostly limited to specific situations. The worst is how play experience is treated with QBs. It affects passing accuracy as part of a broader algorithm that determines ball placement. It takes running a play 20 or so plays to stop seeing obvious negatives to applied to your passing game.