BB/hr.-- is it a fallacy?

I play part-time. Just finishing my 2nd year. (1100 hrs. total) First year I made about $14. an hr. playing mostly 20-40 HE. (The games dried up where I play.) This year playing 30-60 and 50-100 I'm not doing too good, I have been running bad. It looks like it is going to be hard to get into the plus side this year.(The 50 games that I have played in aren't any tougher than the tight aggressive 20-40 that I was used to) I have to travel to play. Staying a week here then not playing for 2 or three weeks, then going to play for another week in the side games at a tournament. I just got back from Foxwoods.

Talking to some pros there I was told that making a BB/hr. is a fallacy. They claim that to make a BB/hr one has to play where there is always a really good game. They claim that nobody in the 30-60 game at the Bellagio is making a BB/hr. They say that the most you can hope for is a SB/hr. Thoughts, comments....

"Talking to some pros there I was told that making a BB/hr. is a fallacy. They claim that to make a BB/hr one has to play where there is always a really good game. They claim that nobody in the 30-60 game at the Bellagio is making a BB/hr. They say that the most you can hope for is a SB/hr. Thoughts, comments...."


I don't see a fallacy. The concensus here seems to be that 1BB is the maximum goal. Doesn't 'maximum' by default assume consistently good games?


I think that with decent game selection the games around here are beatable for 1BB. Doubtful I approach that figure, but I attribute that entirely to leaks in my game.


Tommy

A BB an hour at $30-$60 is possible but you need to play extremely well and try to do most of your play when the games are good. Very few of the pros are able to achieve this level.

As in poker as in sports, 'pros' are not all of the same caliber.


sam

A BB an hour at $30-$60 is possible but you need to play extremely well and try to do most of your play when the games are good. Very few of the pros are able to achieve this level.


In your opinion Mr. Malmuth, or any of the others here, at what level is a game easiest for a pro to earn one BB an hour? I haven't really thought this through. I've always been under the impression that all levels are somewhat equal for earning power. What I mean to say, if you can consistantly beat $20-$40 for one BB an hour, isn't the next obvious step up to $30-$60 where a player would stay until managing to beat that for one BB an hour? Then onwards to $40-$80 etc? Or for the average limit pro, do the $20-$40 limits kind of cap it all. Thanks for any responses.

Look around the table. Do you see people making calls they shouldn't? Fractionally shouldn't?


Do people lay down to your raise when you're a winner, but you call their raise when you lose a very close one?


Bluffs. Occasionally nobody has anything, and they know you will bluff. Do they chase you more than they should, or drop more than they should?


Do you make any money on your bluffs?


You must sniff out where that BB is coming from. If when you have it they call, and when they have it you call - meaning if the cards dictate who wins - you will lose time.


It is not enough to play well. Your playing well has to be met with errors. Your playing well has to be met with consistent and boring enough play that you can tell what is going on, that it isn't camouflaged by a bunch of big breaks that could have gone either way.


I see games where people play their hearts out, and whoever has the better cards ends up winning each hand.


MY ADVICE: TRY REDUCING THE NUMBER OF HANDS YOU PLAY, SO THAT YOU REDUCE YOUR HOURLY FREQUENCY TO A VERY FEW HANDS LIKE TOP-PAIR-GOOD-KICKER. YOU MAY MISS FLUSHES, BLUFFING OPPORTUNITIES, AND SO ON. YOU MAY HAVE TO RAISE A RAG ONCE AN HOUR AND SHOW IT TO MAINTAIN THE EXPERIMENT. BUT THEN WHAT YOU WILL BE ABLE TO FOCUS ON IS AM I WINNING MONEY WITH THIS SIMPLE SUBSET OF HANDS? (MAYBE ONLY COUNT SITUATIONS WHERE OPPONENTS HAVE THE SAME SORTS OF HANDS OR LOOSE RAGS.) AM I MAKING MORE MONEY WHEN I WIN THAN THEY MAKE WHEN THEY WIN? CAN I ISOLATE AN ERROR THEY ARE MAKING IN THESE SIMPLE SITUATIONS? HOW OFTEN WILL I GET THESE HANDS AN HOUR? HOW OFTEN WILL THEY MAKE THIS ERROR?


Then you might get a better idea what is a fallacy and what is not in your situation.


SM

I'd like to echo what Mason said,


"you need to play extremely well and try to do most of your play when the games are good."


Being a professional means that you pretty much have to play on a daily basis, like it was a job. This means that you'll often be playing when the games aren't good. More importantly, most professionals don't have what it takes to step down in limits to reap the greater profit in the lower limit games -- when the higher limit games are bad.


The peter principle is the curse of ego. Few winning players can stand to step down to 15/30 after beating the 30/60 for a year.


- Andrew

Great advice eLROY. I just printed your response and will go over it from time to time. Thanks!

No. The games tend to get tougher as you move up in limit. The small limit games are the easiest to beat for one big bet an hour even with the relatively larger rake. The reason virtually no one does it is that the players who are capable of doing it have gone to a higher limit.

From my actual stats:


20-40, +$48.32/hr. Better than a big bet per hour. This is on almot 500 hours of play.


30-60, +$41.20/hr. Actually *less* than I make at 20-40, after more than 230 hours of play. Only about 30 hours in the bellagio game, at which I am profitable but not hugely so.

So a big bet per game is possible, given the right game. Also, my variance is lower in 20-40, of course. So it's clearly a better game for me. If only it ran all the time....


- target

Thanks to MM and others for replying. So, if I'm reading this right, except for a very few of the top class pros, most solid players who depend on poker for a living find the $20-$40 tables most profitable.


What about the folks playing at the very high limit tables $100-$200 and up? A gathering of eight sharks waiting for one rich fish to show up?


Apologies for the seemingly obvious questions but there are no cardrooms where I live. Apart from three months in Vegas over 25 years ago, I've never played in a legal card room/casino.


One last question which embarrasses me. How many chips in a casino rack? Up one rack, down three racks, I need a frame of reference here.

n/t

"20-40, +$48.32/hr. Better than a big bet per hour. This is on almot 500 hours of play."


500 hours ain't nothing. You probably don't even know you are a winner yet.


Assuming a 10 big bet SD your looking at a 99% confidence that your earn rate is between -5.35/hour and 102.00/hour. And if the game you are playing in is very aggressive its possible you have a higher SD.

These units of fluctuation only apply in casinos that use $2, $3 $5, $10, and $20 chips, where because of the handy denominations, all games use one primary color chip, and are either 4-and-8-chip games, such as $20-40 with $5 chips, or 3-and-6-chip games, such as $6-12 with (with $2 chips) and $9-18 (with $3 chips).


Winning a rack at $4-8 means winning 100 $1 chips, which is a smaller win, by fluctuation comparison, than winning a rack at $3-6. That discrepency aside, the rack and stack (20 chips)lingo is a handy way to talk. Examples: The jammed up pot had over two racks in it. I went through three racks without seeing a river card.


Another handy unit is the "dealer." It's the newest unit of time on planet earth.


Tommy

This reminds me of a question I posted several weeks ago which got no real replies. If 500 hours isn't enough, how many hours do we need to play to be sure that we are winning? I just calculated my hourly rate today. I've played 450 hours of poker. This total includes online poker as well as live play. It includes when I'm sitting at a low limit table and when I've taken a shot at the $60-120 game at the Bellagio. For my $20-40 Hold 'em play in Chicago (my main game) I've made $28.50 an hour over 272.75 hours. I hope I am a winning player and I'm trying to improve my game. So again, after how many hours can I start to be sure that I'm actually beating this game?

How many hours you need depends on how much variance you have (which has a lot to do with the game you play in). However, 1000 hours seems to be the standard number that a lot of people quote. Remember though that because of the way things converge, 500 hours is not half as reliable as 1000 hours; it is much less so.

I have talked to some players who averaged close to $60 per hour in the Bellagio $30-$60 game but they have only played a few hundred hours in which they based this average. Over several thousand hours of play, it is unlikely that anyone really averages $60 per hour in the Bellagio $30-$60 game. This is why the game has such a high turnover.


I also know of a player who won almost $40,000 in 1000 hours of playing $20-$40 along the Gulf Coast. Over the next 800 hours, he lost $13,000. He dropped from $40 per hour to about $15 per hour. After about 2300 hours of play his long term average is just around $20 per hour (one small bet per hour).


I believe one small bet per hour is all that is possible in the very long run for a good player in games that are $30-$60 and higher. If someone is really averaging much more than this, it is because they are not only a good player but a statistical deviate.

"How many hours you need depends on how much variance you have (which has a lot to do with the game you play in)."


Correct


"However, 1000 hours seems to be the standard number that a lot of people quote."


Without knowing which question those people are trying to answer its hard to call them wrong. But guessing at the question I'm guessing their answer is Incorrect.


"Remember though that because of the way things converge, 500 hours is not half as reliable as 1000 hours; it is much less so."


Incorrect. We are talking about an inverse square here. To cut your SD in half you need to play 4 times the number of hours.

We are talking about an inverse square here. To cut your SD in half you need to play 4 times the number of hours.


I haven't done stats in a while. I thought that SD converged at a rate proportional to the square of the number of hours played, so that after twice as many hours your results are four times as reliable. Are you saying I have it backwards and in fact SD converges as the root of the number of hours played? Can anyone else say which it is?


And I assure you that 1000 hours is the "standard" figure that's thrown around for the length of time for a win rate to have meaninfully converged, at least enough that if you are truly a winning player you should be winning.

I also know of a player who won almost $40,000 in 1000 hours of playing $20-$40 along the Gulf Coast. Over the next 800 hours, he lost $13,000. He dropped from $40 per hour to about $15 per hour. After about 2300 hours of play his long term average is just around $20 per hour (one small bet per hour).


Jim, I have seen you mention this story (or one similar to it) here several times in the past. With all due respect, I don't think that it proves anything. It is more than possible that the reason this player's win rate changed is that he started playing worse (which could happen in a variety of subtle ways and for many reasons) or that the game got a little bit tougher. While it is also possible that his play and the game stayed completely constant, we have no way of knowing that so this story really doesn't demonstrate that win rates take a long time to converge, that it's impossible to win a BB/hr or anything else.

I said:


"We are talking about an inverse square here. To cut your SD in half you need to play 4 times the number of hours."


I meant "inverse square root". Thus why I said you need 4 times the hours.


"I haven't done stats in a while. I thought that SD converged at a rate proportional to the square of the number of hours played, so that after twice as many hours your results are four times as reliable."


I just checked. GTaOT pg. 64:

"Now suppose that this person wants to play more hours to cut his standard deviation in half. Since the overall standard deviation is propportional to the inverse of the square root of the sample size, four time the number of hours are needed."


I think you are confusing that the SD converges as an inverse square root but the absolute deviation is the SD times the square root of the sample size.

But Bobby he is not the only one. I know of many players who have had the same thing happen to them. Furthermore, he was not playing in just one game. He won the money playing in Shreveport, Tunica, Lake Charles, and Las Vegas public card rooms. He lost the money in the same places. He posted his story in depth a few months ago on one of these forums.


The belief that if you play well you must get a good result over a finite period of time is deeply ingrained in many people. But it is more like a religion than anything based on empirical evidence.


If I remember correctly, David Sklansky has stated that in his estimation there are not more than a couple of hundred players across the country that make $50,000 per year or more playing in ring games. One has to only beat a little $15-$30 game for a mere $25 per hour to average $50,000 a year if it is played like a full time job with 40 hours a week for 50 weeks a year. $25 per hour is not even one big bet per hour in $15-$30 and is more like one small bet per hour in $20-$40. I would venture to say that most of the "Sklansky 200" are in California, particularly Southern California. If you look elsewhere in the nation, they are few and far between.