I've played 100k hands in 100nl zone on Bovada at a 2bb winrate. For those who don't play on Bovada, zone poker is like zoom on Stars EXCEPT the players are 100% anonymous so you have no idea who you're playing against hand to hand.

The first 50k hands I played I was winning at 8bbs and have been losing heavily over the past 50k hands. Now, I'm not panicking due to a downswing. When I was winning at 8bbs I was thinking its easily beatable (naive to come to that conclusion after 50k hands I know). I'm genuinely wondering if it's possible to beat a pool of anonymous players for a decent winrate. I'd consider 5bb's decent. FWIW I think weekends are beatable for obvious reasons, so if someone were to play only on weekends, they might say yes it's beatable for a lot.

Here's why I am starting to question:
1. http://puu.sh/mXnMr/6c6c0d8feb.jpg (and small blind: http://puu.sh/mXoPi/36b439e150.png)

That is a picture of a 40k hand sample from each position in this player pool. The average stats, to me, look generally unexploitable, or maybe minorly exploitable, but since you have no idea which player is which, you can never actually adjust your play. There's no stat that jumps out at me from any of these positions where I'm like, "yeah ok that's what I can attack."

(actually as I type this I see that their fold to turn cbet and river cbet seem very low, can anyone confirm that? However, even with this, it makes it hard for us to barrel our draws since we won't be able to bluff them off the river very often when we miss?)

The only thing I can think of to exploit the pool is to value bet very thinly.. thoughts?

(stats are:

VPIP/PFR/RFI
3bet/fold to 3bet
cbet flop/turn/river
fold to cbet flop/turn/river)

2. Kind of goes with what I just said above, but I'm finding it hard to theorize how we can make money when we can't adjust our game to account for leaks or player tendencies.

For example, I have filtered for big blind vs BU 3bet, and it's about 8% overall. The problem is you don't know if a particular player has a polarized 3bet range or not, whereas if you had screen names and knew a player's stat from that position you could accurately make that assumption.

It seems "wrong" to assume that most guys will have a polarized 3bet range from the big blind because sometimes I will have to make calldowns in 3bet pots that I would NEVER make against a nit. There are guys in this pool who only 3bet premiums, but we can't adjust for that because we don't know who they are, so we make lighter calldowns to find we're just getting value towned by a nit. This makes me feel like making turn calls w/ 2nd pair or a weak top pair in 3bet pots is possibly unprofitable or breakeven because there's x% chance this guy is a total nit and has QQ+ every time.

I'm starting to rant now, but does anyone have thoughts on the beatability of this type of game?