Highest sustainable BB/100 in 1/2?

[Belok]

In .5/1 i think the highest i've seen or approached was 5BB/100, Would 1/2 be the same? Or closer to 4?. Just curious because I'll be on 1/2 for a long time (The plan is to make 600BBs to get the BR to 1700, cash out 500 and upgrade to 2 monitors, split the remaining 600 between empire and party, and 1/2 it until they're at 1200 each...so a grand total of about 1200BBs profit...ill be chillin for some time)

[CWsports]

I'm also interested in hearing others feeback on this as I'm at 5.94BB/100 after 13K hands and am thinking about moving up and am interested in what rate I would expect to win at $1/$2.

[JDErickson]

any thing above 3 bb/100 hands is usually unsusaintable in the long run over muliple level. There a a few that have done it but for most people 3 bb/100 would be excellent

[Entity]

The other thing to remember is that very few of us will be playing 1/2 over the "long run." 10k hands is a very short run, and you might run 1BB/100 or 5BB/100. I'm nowhere near 10k at 1/2, and am around 3BB/100 right now, but it seems meaningless.

I've decided not to worry about winrates much until I hit 50-60k hands, which is a long time away.

Rob

[btspider]

it varies so drastically and the limit is so low that its entirely meaningless. do you really want to play 100K hands at this limit to see if you can beat it for 4 BB/100 or 4.5 BB/100? these are gateway limits to the 2/4 - 5/10 games that are easily beatable by solid players.

that said, it also varies drastically according to when you play. i don't think there is any way to sustain 5 BB/100 playing a 4-6 BB average pot table during the 1/2 day. on friday/saturday nights, an expert could likely manage it.

be happy with 2 BB/100 +. 3-5 and you may be running well. 0-2 and you may be running bad or playing near your expected winrate. you will just never know.

[Eder]

Just working on switching from NL SnG's to learning limit...at .25/.50 after few k hands I'm around 3 BB/hr...played 1/2 for a k or so...ended around 1BB/hr so moved down...any recomendation/ play more at present level? Move up? I have bankroll for higher and am winning player in up to $20 NL SnG's.

[helpmeout]

The 4bb/100 stat is a myth to make bad players feel good.

I'm sure with good table selection much higher figures are easily sustainable at low limits.

But really, who cares. Why would you want to stay in such a low limit when you have the BR to move up.

Here is an idea, when you beat $1/$2 for 300BB move up to $2/$4 then use your winnings from $2/$4 to make your purchase.

Trying to make 1200BB at one limit is asking for trouble.

[JDErickson]

[ QUOTE ]
I'm sure with good table selection much higher figures are easily sustainable at low limits.


[/ QUOTE ]

You are wrong. Please find me some players with long term results higher than 4.

[helpmeout]

What sort of long term results are we talking about?

Its impossible to have longterm results of +4BB at any low limit because the person would have moved up.

Do you really think that someone playing $15/$30 successfully couldn't beat 1/2 for more than 4BB/100. Thats crazy.

[Ajax410]

Not that my sample size is large enough (or above 4...I think I'm a hair below), but I think that a long-term rate of 4+ BB/100 at 1/2 is absolutely possible. I played around 13k hands at 1/2, and I'm a better poker player now than I was 3 weeks ago, simply by playing better opponents over the course of my 15k hands at 2/4. I am positive that the top players at this site, most of whom play 15/30 or 30/60, would have no trouble CRUSHING 1/2 to the tune of 5-6 BB/100. Some of them are pretty incredible players - it's just the opportunity cost of experimenting is way to high considering how much more they make at the higher limit tables.

Ajax

[JDErickson]

[ QUOTE ]
What sort of long term results are we talking about?

Its impossible to have longterm results of +4BB at any low limit because the person would have moved up.

Do you really think that someone playing $15/$30 successfully couldn't beat 1/2 for more than 4BB/100. Thats crazy.

[/ QUOTE ]

Just as an example. My 1/2 results. 38000 hands 2.99 BB/100.

Of course I still have a lot to learn.

Do these 15/30 players get any better hands playing at the micro limits? Sure micro players are usually pretty bad but 15/30 players are also very bad. Why do you think they would have double their win rate at micro limits?

If a good player was multitabling 4 tables then I think they could come close to 4 bb/100, maybe even a little over. But to assume that these win rates are normal or expected is not reasonable over the long run.

[emonrad87]

I think it is very sustainable for win rates of 5-7BB/100 on 1/2. You just dont see this very often (if at all) because anyone who makes this/would make this will attain their desired # of BB in their bankroll and move up. Then, whenever they say "i regularly beat 1/2 at 6.48BB/100" they will be instantly dismissed as a sample size alert. For example, if said person wanted to make 550BB at 1/2 in order to have a 300BB bankroll at 2/4 and began with a $100 bankroll, it would take him only 8488 hands to do so. This sample size would never be considered significant, even if the trend would have continued - we will never know. So, who knows? I'd say up to 7 BB/100 is attainable. I know i'm not gonna sitck around at 1/2 to find out tho! /images/graemlins/laugh.gif

[JDErickson]

Heres another post about the same thing.

Win Rates (http://forumserver.twoplustwo.com/showflat.php?Cat=&Board=smallholdem&Number=1167437 &Forum=,All_Forums,&Words="win%20rate"&Searchpage=1&Limit=100&Main=1167237&Search=true&w here=bodysub&Name=&daterange=1&newerval=1&newertyp e=y&olderval=&oldertype=&bodyprev=#Post1167437)

[joker122]

[ QUOTE ]
Do you really think that someone playing $15/$30 successfully couldn't beat 1/2 for more than 4BB/100. Thats crazy.

[/ QUOTE ]

I just had the best idea for a quest.

[helpmeout]

eww 38000 hands move up already

$0.5/$1 5,068 +$368.09 7.26BB/100
$1/$2 (6max) 3,770 +$521.25 6.91BB/100
$2/$4 1,107 +$224.62 5.07BB/100

I'm not claiming that I am a good player or that my results are sustainable. But I did feel very comfortable at beating both $0.50/$1 and $1/$2 mainly due to great table selection. I did also run bad at the end of my $1/$2 as well as playing poorly I am also playing poorly at $2/$4 (calling station mode).

The games at 50c/$1 and $1/$2 (6max) are extremely weak if you pick the right tables.

Anyone who has studied hard, has a good deal of nano limit experience as well as pokertracker and good table selection should be able to beat these games easily.

I think people who play one limit for too long will experience worse results over time due to being bored/complacency.

Anyway I know I will get butchered in $2/$4 now cos of this post so I better hit the books.

[Mr. Graff]

[ QUOTE ]
Not that my sample size is large enough (or above 4...I think I'm a hair below), but I think that a long-term rate of 4+ BB/100 at 1/2 is absolutely possible. I played around 13k hands at 1/2, and I'm a better poker player now than I was 3 weeks ago, simply by playing better opponents over the course of my 15k hands at 2/4. I am positive that the top players at this site, most of whom play 15/30 or 30/60, would have no trouble CRUSHING 1/2 to the tune of 5-6 BB/100. Some of them are pretty incredible players - it's just the opportunity cost of experimenting is way to high considering how much more they make at the higher limit tables.

[/ QUOTE ]

They would have to make some major adjustments to their game though. Less fancy stuff and more straight forward valuebetting. Also a lot more running to the hills when people come out raising on a 33T flop - that someone IS holding a 3 much of the time even though he is in mid position.

[jason1990]

[ QUOTE ]
do you really want to play 100K hands at this limit to see if you can beat it for 4 BB/100 or 4.5 BB/100?

[/ QUOTE ]

I just want to point something out for people who might not realize it. The general rule of thumb is that "you need 100K hands to know if you are a winning player." It is not "you need 100K hands to have an accurate estimate of your winrate."

Consider this: suppose you have a SD of 17 BB/100. If you play 100K hands, then your winnings after all those hands will have a SD of about 500 BB. If you end up winning 4000BB and you want a 95% confidence interval on your winrate, then you can only conclude that your winrate is somewhere between 3 and 5 BB/100. If you want a 99.5% confidence interval, then it's even worse. You only know you have a winrate between 2.5 and 5.5 BB/100.

If you end up doing poorly, it's the same story. If you end up winning 2000 BB after 100K hands (with this same SD), then all you can say (with 99.5% confidence) is that your winrate is somewhere between 0.5 and 3.5 BB.

In any of these cases, you would know you are a "winning" player, but you would not have pinned down your winrate with any level of precision.

[bisonbison]

helpmeout, your stats are approximately meaningless.

When the worst player in the world wins at 7BB/100 they also "feel very comfortable at beating both $0.50/$1 and $1/$2 mainly due to great table selection."

Guys, most of us won't end up beating 15/30 for 2BB+/100. Sure, someone who is beating 15/30 for that much can probably sustain 4BB+/100 at 1/2, but you need to understand that that player is much better at poker than you think.

[jason1990]

Okay, the precise claim is that, if you have a SD of 17 BB/100, and you want to know your winrate to within 0.1 BB/100 (with 99.5% confidence), then you need to play about 26 million hands. (This is a follow-up to my previous post.)

Here's the math: If you play n 100-hand sessions, then your average winnings (in BB/100) will have a SD of 17/sqrt{n}. To know your winrate to within 0.1 BB/100 (with 99.5% confidence), you need this to be less then 0.1/3. So we want

17/sqrt{n} < 0.1/3
sqrt{n} > 17*3/0.1 = 510
n > 510^2 = 260100.

So you need to play over 260000 100-hand sessions to achieve this level of precision.

Changing the confidence level will not affect these calculations much. To get away with a lot fewer hands, you need to change the 0.1 to something much bigger.

So, in summary, 100K hands is not nearly enough to know you're beating the game for a specific winrate. It's only enough to know that your winrate is in some acceptable (but wide) range (as in the previous post). So with only 100K hands, you can know that you're beating the game, but you can't know (accurately) by how much.

[Fiddler]

[ QUOTE ]
Okay, the precise claim is that, if you have a SD of 17 BB/100, and you want to know your winrate to within 0.1 BB/100 (with 99.5% confidence), then you need to play about 26 million hands. (This is a follow-up to my previous post.)


[/ QUOTE ]

Now we are talking about a REAL quest! Should be easy to get one of the high-limit players to settle this for us. /images/graemlins/smile.gif

[JDErickson]

[ QUOTE ]
eww 38000 hands move up already

[/ QUOTE ]

Shall I also list my 42000 2/4 hands and my 36000 3/6 hands and my 10000 5/10 hands?

As I thought, your sample size is zip. In fact my sample size is zip.

Congrats on running so well in your short runs at the levels. It won't last.

[Blarg]

[ QUOTE ]
$0.5/$1 5,068 +$368.09 7.26BB/100
$1/$2 (6max) 3,770 +$521.25 6.91BB/100
$2/$4 1,107 +$224.62 5.07BB/100


[/ QUOTE ]

Those numbers are too tiny to mean anything, helpmeout.

Here are two chunks of my 2/4 play on two different sites from the first week of this month to compare:

1248 hands, $521.50, 10.18 BB/100
3746 hands, $1,241.75, 8.29 BB/100

I felt pretty good about those numbers, but they indicated good results more than good play. Good to be happy about, worthless as signposts of anything, really. I also posted some pretty lousy numbers this month, and overall just made average results. Both winning and losing players have all kinds of results in the short term.

Sample size man really needs to come out with his baseball bat and give us a thorough beat-down when we think differently with only some small number of sessions strung together to back us up.

[Quercus]

21000 hands of $1/2, BB/100 of 3.29. I'm improving, but I still make plenty of mistakes.

[Blarg]

Sweet. The nice thing about good runs like that is they give you something to average out the bad runs with. There's a good chance you'll still come out with some really nice numbers over the next 20k hands even if things run pretty sour for a while.

My main hope when it comes to win rates is just to make 2 BB/100 and have it be very sustainable. The sustainability is worth just as much to me as how high the win rate might possibly go.

[helpmeout]

Sure my stats are pretty meaningless and even if I continue on my good run and am playing 5/10 having played another 10k hands you guys still will say sample size too small.

My point is that there are many +EV plays that an average player like myself and others miss out on.

If you take out most of your mistakes and add in a couple of great plays then there is much more potential for gain.

A very good player can probably squeeze a few BB into an already big pot or gain a few BB from stealing a pot.

There are at least a couple of opportunities every 100 hands.