Elo changes with luck rating

[ChiefPointThief]

Again, that luck rating purely depends on the developer's whim.

Again, I understand this and it shouldn't be this way. This thread was linked from the same topic on the main discussion. In that thread and several other luck related threads I suggested that there should be a better system for determining ratings like putting it in the players hands.

Also, simple adjustments cannot solve this whole 'ELO is flawed' debate.

Maybe it won't "solve" it but I don't see how it makes it worse.

There are tons of games with variants and different setups which change the impact of luck factors.

There are also tons of games that don't have settings that drastically change luck factors. After the players determine the luck factor for the game the dev can then apply the proper elo distribution for that game based on the luck factor.

In the worse case scenario if a base game has a luck factor of 2 and adding settings can increase that to 3 you would use the lower luck factor setting. This method is still better than treating all games like they have a luck factor of 0.

Issues like should different player counts have a separate elo is a separate debate (games like papayoo that are more controlled w/ 3p compared to 7p making the luck factor lower).

In short, the luck rating of metadata is not reliable as you expect.
And this will cause more needless confusion, like ELO 'updates' introduced in the last year.

This is nothing like the bga elo 'updates'. Those changes didn't consider legitimate gaming and were put in place to discourage ppl from playing against themselves.

[Ceaseless]

I did an excel to see better the potential changes: https://docs.google.com/spreadsheets/d/ ... id=0#gid=0 (it's not an exact replication, but should be close)

With k-factor 20 you got Alice at 159,97elo, Bob at 166,26elo, and Cathy at 273,77elo.
If you put a k-factor at 30 (so like putting luck-factor at 1.5), you got Alice at 126,14elo, Bob at 164,97elo, and Cathy at 308,89elo.
And if you put k-factor at 13,3 (so like putting luck-factor at 0.67), you got Alice at 179,65elo, Bob at 172,29elo, and Cathy at 248,05elo.

And so on my third example, the bad streak of Alice was less detrimental at the end, they even goes above Bob in elo.

A3=B3=C3=200,0
FG columns are 1,2 five times, then 2,1 twice, then 1,3 thrice, then 3,1 six times.

I never did end up posting a reply to this. I looked at the chart at the time and typed some stuff out the day I read it, saved it for later, and then never went and made a post. A few months later and here we are. Figured I'd at least mention that yes, I did look at this chart at the time.

Took me a bit to work out the sheet. I'm not much of a computer person. Had to work out a few things to make sense of it at first, but after I got used to that it was a nice chart to mess with.
- The comma is a decimal point
- Alice, Bob, and Cathy are players. Under Winner/Looser however, instead of names, Alice is 1, Bob is 2, and Cathy is 3.
- The rows below Alice, Bob, and Cathy are their ratings starting at 200, going through changes per row due to a game's result, only two of them (named in winner/looser) will play a game and exchange rating points each row. The player that didn't play would keep the same rating.
- The elo winner/elo looser columns name the active players by their ratings, but exist for the % chance of victory to calculate their chances with its formula (a tenth of the rating gap from winner to loser is added to 50% then multiplied by the k-factor.)
- You have an unused k-factor column with a highlighted number to the right representing the k-factor.

Watching the way the win order could do funny things to ratings (and even change the ranks of the players) was nifty. I do think I underestimated the impact some of it could have, though I'm not sure by how much.