Mafiascum.net

not that spreadsheet, the crazy spreadsheet. which is a different, but much cooler spreadsheet with yet another strange scoring system.

You can do what tennis does and make ratings drop off after a year.

Oh wait a second, CC. You adjusted your total prize pool not by +2500 per person but by the percent increase over 16?

I'm getting different numbers than 92875 using your table for:
Seafoam Islands
Arkham Asylum
Conclave
NAH
EtU
Arkham 2
PSV
2016
LoL
Nexus

In post 100, zoraster wrote:I think that a 10:1 ratio of first to second is totally inappropriate for a rolling determination of supposed skill. Like if you just want a fun "I wonder what our winnings would be if we used survivor" then sure. Keep it the same or very, very similar. But as a replacement or alternative measure to the elo system I think it'll be sorely lacking.

In post 101, hiplop wrote:in survivor often second place is a bad player

yeah this

second place (or third place) is often someone who was bad at the game and therefore got dragged there by someone else for the free win

In post 113, zoraster wrote:Technically xofelf has a lower win rate than TheBadOne and is higher ranked, but that's the difference between 6.25% and 6.67% so almost not worth mentioning

IT IS TO ME!

See, I feel like this list is maybe better.

My whole point is that people think making FTC is enough and I'd love a ranking system that weighted wins harder to try and accommodate for that but I'm not sure how to really do that.

In post 127, zoraster wrote:Oh wait a second, CC. You adjusted your total prize pool not by +2500 per person but by the percent increase over 16?

I'm getting different numbers than 92875 using your table for:
Seafoam Islands
Arkham Asylum
Conclave
NAH
EtU
Arkham 2
PSV
2016
LoL
Nexus

On the ones you mentioned, the total isn't exact because multiple players tied, I believe

In post 128, xRECKONERx wrote:

In post 100, zoraster wrote:I think that a 10:1 ratio of first to second is totally inappropriate for a rolling determination of supposed skill. Like if you just want a fun "I wonder what our winnings would be if we used survivor" then sure. Keep it the same or very, very similar. But as a replacement or alternative measure to the elo system I think it'll be sorely lacking.

In post 101, hiplop wrote:in survivor often second place is a bad player

yeah this

second place (or third place) is often someone who was bad at the game and therefore got dragged there by someone else for the free win

I mean, either we operate under the assumption that places matter and thus are relevant for the sorting of results or we don't. If we don't, then a pure Win % is more appropriate than a misleading $ amount.

Because as it operates above, all you're doing is creating a system that shows Win % for those with at least one win and then something more meaningful for everyone else.

another thing: i think active rankings should be a limited time period vs inactive/lifetime rankings?

p.edit: not sure that's a wrong thing though.

idk

I'm not sure it is either! But then we should have a "Winner's Circle" or something that shows How many people have won and what percent of their games that comprises.

Pretending that money does anything other than show win percentage for those with a win seems dishonest to me.

in this $$$ system, top 5 of all time is:

snakes
betch
reck
cc
jal

in the elo system, top 5 of all time is:

snakes
SiW
STD
BPC
Reck

i kind of feel like the first system is maybe more indicative of survivor history than the latter system?

and given the closeness of most of the $ amounts, one win is enough to catapult someone into the upper echelon, which is what i think is important

meh money does something win % doesnt because like

ive won 3 times
ive played 12
3/12 aka 25% winrate doesnt look great

esp compared to bella winning 2/5
or jess winning 1/2

but when you consider that out of my 9 losses, 4/9 involved making merge it looks better?

i dunno i think $$$ amount is probably more indicative of someone's historical impact at least. like, even in the scenario where someone played 5 games, if they're placing in the bottom every time, that's gonna be overshadowed by one person making jury one time, right?

elo gives you a "current" snapshot of the playerbase
$$$ gives you a historical long-term snapshot

This also gives more weight to someone winning in a 24-person game than someone winning an 18-person game, for example

right, which i think is super important too?

In post 136, xRECKONERx wrote:meh money does something win % doesnt because like

ive won 3 times
ive played 12
3/12 aka 25% winrate doesnt look great

esp compared to bella winning 2/5
or jess winning 1/2

but when you consider that out of my 9 losses, 4/9 involved making merge it looks better?

i dunno i think $$$ amount is probably more indicative of someone's historical impact at least. like, even in the scenario where someone played 5 games, if they're placing in the bottom every time, that's gonna be overshadowed by one person making jury one time, right?

Then list by total number of wins. Your merges and other finishes OTHER than your wins are largely irrelevant to your total $ won.

In post 138, CuddlyCaucasian wrote:This also gives more weight to someone winning in a 24-person game than someone winning an 18-person game, for example

I have zero objection to that.

Then list by total number of wins. Your merges and other finishes OTHER than your wins are largely irrelevant to your total $ won.

i sorta disagree

i dunno i think there's a system in which we say, okay

making FTC, there's x % of total value tied up in that
making jury, there's x % of total value tied up in that
then not making jury has a smaller value than either of the others

y'all take this shit so serious, man

It's a game, just have fun, find some idols, get your allies voted out, go home mid-jury, like the cool kids do!

like if we list by total wins i look like im on par w/ snakes and betch and theyre very clearly above me in a big way

In post 143, Xalxe wrote:y'all take this shit so serious, man

It's a game, just have fun, find some idols, get your allies voted out, go home mid-jury, like the cool kids do!

can we skip past the part where people act like not giving a fuck is cool bc idk seems fruitless

In post 143, Xalxe wrote:find some idols

Well shit, I will never a be a cool kid

In post 142, xRECKONERx wrote:

Then list by total number of wins. Your merges and other finishes OTHER than your wins are largely irrelevant to your total $ won.

i sorta disagree

i dunno i think there's a system in which we say, okay

making FTC, there's x % of total value tied up in that
making jury, there's x % of total value tied up in that
then not making jury has a smaller value than either of the others

Your $ totals are made up 93.6% by your two first place finishes.

yeah but that 7% tho

lol