Best winrates one should hope for?

[d0zer]

I'm curious:

What's the best winrate (BBs/100) you've ever seen (and believed) over a decent sample (20k+ hands)? What is generally average for a half decent player?

At:

2NL
5NL
10NL
25NL
50NL
100NL
200NL
400NL?

...

1000NL?

[wufwugy]

Assuming games are current and static, depends on how good you are. I've seen sick 5ptbb at 3/6.

And 20k hands is nothing.

[mrhappy333]

5ptbb @ 3/6 = how much $$

30 hr? or is my math off? please help.

[wufwugy]

It's ten blinds per hundred hands at three and six blinds cash game fullring no limit hold them.

[BankItDrew]

11 tabling 3/6 @ 5bb/100 = $450/hr

[BankItDrew]

You should really be happy with any sort of positive winrate. Actually wait, it would suck if you are studying the game and are winning pennies.

[wufwugy]

I believe I could make 10$/hour at 2nl playing the most basic basic simple basic strategy. Nobody should be making pennies.

As for a more exact answer to the op, I estimate for fullring

25nl - 8ptbb is killing it
200nl - 5ptbb is killing it
1knl - 3ptbb is killing it

[BankItDrew]

^^^^
after how many hands?

[Miffed22001]

20k hands is a heater, eg you can run super good or sick bad for that time

[BankItDrew]

^^^^^
so how many hands is a good indicator? 100k?

[wufwugy]

Amount of hands cannot quite determine how true winrate is. 5mil or something is probably what it takes to get true winrate, but it's completely impractical. Nobody plays that many hands, and there's so much skill and opponent fluctuation everywhere along the way that it just does not work.

What it takes to evaluate winrate closely, imo, has a lot to do with simply playing and knowing close to how you're running.

Also, most of the time we run normal. Most 20k stretches are close to normal, a much greater degree of 200k stretches are close to normal, and then substantially greater degree of 2mil stretches are close to normal. So we're dealing with likelyhoods of expectation. We can never say a certain sample is enough, but we can say that a certain sample is more likely enough.

My last 250k hands has had one 75k breakeven, and my last 40k has been further below expectation than I can remember, but overall I'm even. I see it being extremely unlikely to run a bit off expectation over 200k. Even 100k b/e are often balanced after 200k.

[d0zer]

My last 250k hands has had one 75k breakeven, and my last 40k has been further below expectation than I can remember, but overall I'm even. I see it being extremely unlikely to run a bit off expectation over 200k. Even 100k b/e are often balanced after 200k.

how long does 250k hands take to rack up?

[wufwugy]

I play 40k/month. That's 8tabling 20 hrs/wk.

[d0zer]

so you went months basically making nothing?

Musta been brutal. I feel in a bitofa rut myself. Losing every set/vs draw I can remember.

So d'u have an uplifting tale of the donkfest days of yore before that?

[wufwugy]

Basically. It can be tricky though since breakevens are often comprised of long downswings/upswings. Rakeback is nice I guess.

When you make your money from sets that can be tough. A slightly better way of understanding how you're running is how AA is doing. Even better is knowing how to analyze your own game. You'll notice that the better a player is, the fewer hands he posts. This seems to be because he has enough fundamental and theoretical understanding to work them through on his own. Evaluating how he's playing/running then becomes largely a part of strategic alterations. I've made a ton of progress by just going over other players' PT stats and figuring why they're different than mine.

High winrates (what everybody should be able to do before they move up) make a bit more money from non-set stuff than you'd think. The higher your winrate, the lower your downswings and shorter your breakevens. Somebody killing 25nl will probably have like 5 bi downswings at absolute worst.

Uplifting tales? I was breakeven on money for my first year and a half (about 250k hands). I didn't beat any games for high rates and got annihilated when I moved up or ran bad (this was as the games were getting worse and worse due to UIGEA and Neteller also). Perseverance has gotten and continues to get me better.

[d0zer]

5mil or something is probably what it takes to get true winrate

But statistical confidence is like a logarithmic function bounded by 100% and your total sample size, in the case of poker is every single hand you've ever played. ...by the time you've hit even 10k, you've got enough of a sample to give some confidence.

I know what you're getting at is "it's not a true measure of skill necessarily". And yeah, that's true...but at the same time, a shark is far more likely to have a higher BB/100 than a donk, especially over more and more hands. 20k is probably enough to determine donkishness, though certainly it's a small enough sample that an ABC TAG could do better than a true shark sometimes.

It's not an absolute measure of skill cause there are so many variables. villain skills and cards to name a few...

In conclusion, I guess my question is kinda silly.

...but even so -- surely FTR players are sharkier than the status quo and we could get some averages / max / mins.

There's gotta be a way to measure the average profitability of online poker in its current state