OT: MATH PEOPLE NEEDED. Slim Pickens, Sheets, etc... help answer! Theo

[The Yugoslavian]

[ QUOTE ]
I think there was a long thread ~3 years ago started by Dali on the subject of how much your ROI is affected by one or two extra good players sitting in your game. Not quite what OP is asking about, but anyone interested in the general topic might want to pull that up from the archives.

[/ QUOTE ]

And a couple years before that there was a bozeman post about it. [img]/images/graemlins/tongue.gif[/img]

Yugoslav

[Shillx]

W/o thinking about it I would assume...

A-F) -1.1%
H) -11.0%
I-J) +8.8%

[j00hndayton]

I think gramps is on the right track in terms of what I am asking. Basically, I want to know how to quantify how much your ROI is affected by one or two extra good players sitting in your game. So I'll def. try to find the post from Dali and Bozeman.

Slim - I left rake out because I thought it would simplify things. Basically, I thought it would be the same as saying the break even players actually loose the rake. If it would be really important when doing these calcs, then by all means add it.
What I wanted to do was keep as many variables as constant as possible so that this question would actually have a mathematical solution. I don't see why there shouldn't be an exact answer if you hold alot of things constant.
Perhaps someone can tell me why this hypothetical doesn't have a precise answer (aside form saying that you cant account for how different the play will be, or who is sitting next to whom, or any other external/psychological factor).

[Slim Pickens]

No worries. You can get a precise answer with a few assumptions I think are very reasonable. The rake is a constant, so it's not that hard to put in. It's just that we like to talk about ROI values after rake but you have to account for all of the money in the end. Just do what I outlined and it will work. That, or just believe Shillx.

[DevinLake]

If you are looking for a quick and dirty solution, just look at it like this.

If you add a 10% ROI and this has equal affect on all players, that means the 9 players have to each eat this 10% evenly.

So, everyone's ROI should drop by 10/9 = 1.111%.

So, the break even players should now be -1.111%.
The -10% player should be -11.111%.
The +10% players should be +8.889%.

Realistically, adding a good player is going to have a larger negative effect on the worse players.

If you assume the 9 players eat that 10% based on their edge in the original game:
- The breakeven players will drop 1.0x%.
- The -10% ROI player will drop 1.1x%.
- The +10% ROI players will drop 0.9x%.

So, 6(1.0x)+1.1x+2(0.9x) = 10%.
x = 1.124%.

Breakeven: ROI = 0 - 1.124% = -1.124%.
Losing player: ROI = -10% - 1.1(1.124%) = -11.236%.
Winning players: ROI = 10% - 0.9(1.124%) = 8.989%

[Sober]

Something interesting that might provide clues here is to look at the head-to-head search on sharkscope. Find players who you have a lot of games played with, and check them out.

I find that, predictably, my roi goes down a couple percent when sitting with some of the better mid-level players (chuckpalms, etc), as does theirs. If you have large enough sample sets with a variety of players, you could probably test these equations out.

Funny, The_Venetian has a -3% roi when sitting with me (over only 55 games, so that means nothing, but still [img]/images/graemlins/grin.gif[/img])

[Slim Pickens]

Devin,
You'll look cooler if you use letters and Greek symbols instead of numbers.

Life Spoiler Alert:
<font color="white">You end up looking a whole lot dorkier, but not really that much cooler.</font>