Ok so me and my friend are going to make a Fantasy system for Valorant and we have a question. How do we make it work with things like playoffs and teams being eliminated?
The way infamy does it is just that the points stack the further the player goes into the tournament which imo is a really bad system because a bad player can be carried deep into a tournament and a good player can go out early despite being better than them.
My solution would be to make a system that gives points based on average performance instead of their total performance, but grants more points per win that way winning is rewarded and stats arent watered down by playing too many games,
if you're interested I would love to help with this sort of thing it sounds fun
I like the average idea. We came up with the idea that points will be given like this.
Normal stuff like
Kill + 5
Death - 5
Assist + 2.5
Blah blah normal point scoring basically.
Then we add their average all season.
Example
Week 5 Demon 1: (Current Performance Points) + (Average Performance all Szn) = Total Points Week 5.
This way players who don't make playoffs far can still contribute. It would just be Average + Average for their points if they didn't make playoffs.
Thats a good start, I think its important to factor in first kills/deaths and kast. I think its very important to give a slight bonus for going further into tournaments just because there more to the game than just individual stats
maybe +2 for first kill/ -2 for first death.
For KAST I feel like the best way to do it is to do like:
0-65%: -2 points
65-70%: 0 points
75-80%: 1 points
80-85% 2 points
85%-100%: 3 points
But honestly I feel like you could just use the VLR rating with like a slight multiplier for every match played, making a unique scoring system is risky because nothing works if its flawed, dont get me wrong the VLR rating system isnt perfect but its a good system, I feel like its just the best way to do it
its very hard to accurately measure the impact of a player since different playstyles will have different results in stats. A good way a measuring it might be combining a lot of the stats described in this video for example. https://www.youtube.com/watch?v=FrLtJSpaSLI