Making matrix-6 more balanced

[gigabyteTroubadour]

i absolutely agree w you on 24 but i know like jack shit on how i could support that argument that that should be done

[mhsmith0]

To work it requires

1) mods and listmods who are willing to enforce harsher standards on low quality ICing
2) a group decision to engage in education about what kind of ICing is desired
3) a core of ICs who are willing to bear that load (since otherwise games just never fill - iirc it used to be that nacho and Thor did almost all the ICing for a while, and I don't think a two IC group is at all acceptable or functional)

I think it's the sort of thing that takes an active group decision to really move down that road, not at all sure there an appetite for it tbh.

[Accountant]

(one easy example: slapping on a "hi everyone I'm the IC here's a bit of intro theory and/or links" opening post and then acting like a normal player the rest of the way becomes actively unacceptable conduct for an IC)

i'm guilty of this

[mhsmith0]

Fwiw most ICs are guilty of this.

[Accountant]

I'm trying to improve

[Toomai]

So the spreadsheet link on
viewtopic.php?f=5&t=39739

got pulled for some reason

It did? Um. I'll try to fix that ASAP. For now though here's the current:

Setup	Wins	Losses	Winrate	Games
1 (JK)	31	33	48.4%	64
2 (RB, Cop, Doc)	38	33	53.5%	71
3 (BP, Trk)	28	40	41.2%	68
A (JK, RB, BP)	28	42	40.0%	70
B (Cop)	30	33	47.6%	63
C (Doc, Trk)	30	38	44.1%	68
Total	185	219	45.8%	404

EDIT: Fixed link here. Updating the main stats thread topic now.

[gigabyteTroubadour]

wow town winrates have gone down quite a bit

[mhsmith0]

Have they? I feel like 45% is around what the average has been for a while, maybe I'm wrong? Also setup 2 has created a bit of a cushion above 50% which is intersting.

[Toomai]

I would like to note that there's another factor to winrate that most people probably aren't considering: the winrate of each individual

player

. Winrates per-player are always going to be closer to 50% than winrates per-faction. For example, in our current Matrix6, where town's winrate is 45.8%, each individual player's winrate is 47.7% (because while town wins less often, when it does, more players win in total).

Here are some more numbers to illustrate the concept:

• A town winrate of 68.0% will give us a player winrate of 60.0%.
  
• A town winrate of 60.0% will give us a player winrate of 55.6%.
  
• A town winrate of 50.0% will give us a player winrate of 50.0%.
  
• A town winrate of 40.0% will give us a player winrate of 44.4%.
  
• A town winrate of 32.0% will give us a player winrate of 40.0%.

So while it's still best to strive for 50%, if we care about each

player

having an equal chance of winning and losing, the range of acceptable

faction

winrates may be a bit larger than expected.

(Of course this opens up a can of worms as to "when" each player's expected winrate should be determined: before or after they're assigned to a faction? So it might be best to just ignore this post.)

[mhsmith0]

I think that the key is for the setup not to be horribly unbalanced, and then over the long run winnrates will tend to tie back reasonably well to player ability (I say this having eaten a parade of bad luck losses but lolvariance I guess )

[RadiantCowbells]

I don't see the purpose in tracking individual winrates as town in a game where the opposing faction can simply remove people that they consider threatening.

The better you get the less you influence games the closer your winrate gets to the average.

[Infinity 324]

Getting killed n1 influences the game a lot, then the next strongest player won't get NK'ed and so on.

[RadiantCowbells]

You're assuming there is a next strongest player and that they're town.

My wr in games where I was n1ed is like 30%

[Infinity 324]

There has to be a next strongest townie...

[mastina]

Fwiw most ICs are guilty of this.

It IS possible to swing the other way and be guilty of the opposite extreme though.

[Loopdan]

Bumping this thread. I'd like to see ThinkBig's setups tested.

[mhsmith0]

That's what the micro queue is for. I don't think there is, or should be, any momentum towards actually changing the newbie setup which is well balanced in each of the six setups over a pretty large sample size.

[Loopdan]

I think the BP claiming strategy will only be a blip on the overall balance of Matrix6.

However I'm conflicted on starting every Newbie game with a BP (and then possibly tracker) mass-claim.

[mhsmith0]

I think that the strategy as currently understood is merely a minor help to town win rate, and that it will likely have a detriment to cop and cop/doc/rb setups as it makes fake claiming doc much easier (50-50 to avoid a cc in goon/goon after we learn there's no tracker), but at the same time less believable.

I think people put too much stock into the idea that it somehow breaks the setup, and allow it to be a detriment to scum hunting. I also think that people bitching about or debating the strategy in lieu of scum hunting is as or more detrimental.

[TesXX]

Replace the BP with a Town Las Vegas.

[FireScreamer]

Whether the setup is broken in terms of winrates or not is besides the point. The first mafia experience we are giving new players is a series of claims based on a strategy they are told to just accept. It isn't very engaging, it isnt particularly representative of the game.

Whether its a good thing for winrates or not matrix 6 has been broken in terms of providing the kind of experience I think new players should get.

Just my thoughts as a new player.

[ThinkBig]

Whether the setup is broken in terms of winrates or not is besides the point. The first mafia experience we are giving new players is a series of claims based on a strategy they are told to just accept. It isn't very engaging, it isnt particularly representative of the game.

Whether its a good thing for winrates or not matrix 6 has been broken in terms of providing the kind of experience I think new players should get.

This is exactly how I feel.

[FireScreamer]

Happy to provide a newish players perspective to any playtesting group you set up in micros. Let me know if/when that happens.

[ThinkBig]

Happy to provide a newish players perspective to any playtesting group you set up in micros. Let me know if/when that happens.

Sounds good. Do you want to pre in?

[FireScreamer]

Sure thing.