Estimated Win Rates
mmal...@ix.netcom.com
The following table is from our new book GAMBLING FOR A LIVING written
by David Sklansky and myself that should be available late this month.
It should make for interesting discussion. In addition, GAMING, POKER, &
LIFE by David Sklansky will also be available later this month. Both
books are produced and published by Two Plus Two Publishing.
Limit Expectations (per hour)
Very Good Selective Very Excellent
Player Good Player Player
$10-$20 $20 $30 $25
$15-$30 $25 $40 $35
$20-$40 $25 $45 $40
$30-$60 $20 $45 $45
$50-$100 $0 $30 $50
$75-$150 -$50 $0 $50
$100-$200 -$100 -$50 $50
$150-$300 -$300 -$100 $0
$300-$600 -$1,000 -$300 -$300
Limit Expectations (per hour)
Selective Excellent World Class Selective
World
Player Player Player
$10-$20 $35 $30 $40
$15-$30 $50 $40 $55
$20-$40 $60 $50 $65
$30-$60 $80 $60 $85
$50-$100 $100 $80 $110
$75-$150 $125 $100 $140
$100-$200 $125 $125 $160
$150-$300 $125 $150 $200
$300-$600 $0 $100 $250
Mason Malmuth
Steve Proctor
In article <334474...@ix.netcom.com>, mmal...@ix.netcom.com wrote...
>
>The following table is from our new book GAMBLING FOR A LIVING written
>by David Sklansky and myself that should be available late this month.
>It should make for interesting discussion. In addition, GAMING, POKER, &
>LIFE by David Sklansky will also be available later this month. Both
>books are produced and published by Two Plus Two Publishing.
>
>Limit Expectations (per hour)
> Very Good Selective Very Excellent
> Player Good Player Player
>$10-$20 $20 $30 $25
>$15-$30 $25 $40 $35
>$20-$40 $25 $45 $40
>$30-$60 $20 $45 $45
>$50-$100 $0 $30 $50
>$75-$150 -$50 $0 $50
>$100-$200 -$100 -$50 $50
>$150-$300 -$300 -$100 $0
>$300-$600 -$1,000 -$300 -$300
>
Uh...I think this may require some explaination. Why should an excellent
players expectation be negative at the higher limit games?
Steve
Stanley Chow
In article <5i2rrt$eij$1...@news.iquest.net>,
Steve Proctor <ste...@iquest.net> wrote:
>In article <334474...@ix.netcom.com>, mmal...@ix.netcom.com wrote...
>>
>>The following table is from our new book GAMBLING FOR A LIVING written
>>by David Sklansky and myself that should be available late this month.
>>
>> Very Good Selective Very Excellent
>> Player Good Player Player
>>$10-$20 $20 $30 $25
>>$15-$30 $25 $40 $35
>>$20-$40 $25 $45 $40
>>$30-$60 $20 $45 $45
>>$50-$100 $0 $30 $50
>>$75-$150 -$50 $0 $50
>>$100-$200 -$100 -$50 $50
>>$150-$300 -$300 -$100 $0
>>$300-$600 -$1,000 -$300 -$300
>>
>[rest of table chopped]
>
>Uh...I think this may require some explaination. Why should an excellent
>players expectation be negative at the higher limit games?
I will make a wild guess that at the higher limits, even excellent
pro's don't have the bankroll to ride out the variance. It is much
more likely to get to the daily "stop-lost" limit so they book a
huge lost.
--
Stanley Chow; sc...@bnr.ca, stanley....@nt.com; (613) 763-2831
Bell Northern Research Ltd., PO Box 3511 Station C, Ottawa, Ontario
Me? Represent other people? Don't make them laugh so hard.
Erik Reuter
I reformatted and re-arranged Malmuth's kindly posted chart. Hopefully
this is a bit more understandable.
Limit Expectations (per hour)
Very Good Excellent World Class
Player Player Player
$10-$20 $20 $25 $30
$15-$30 $25 $35 $40
$20-$40 $25 $40 $50
$30-$60 $20 $45 $60
$50-$100 $0 $50 $80
$75-$150 -$50 $50 $100
$100-$200 -$100 $50 $125
$150-$300 -$300 $0 $150
$300-$600 -$1,000 -$300 $100
Limit Expectations (per hour)
Selective Selective Selective
Very Good Excellent World Class
Player Player Player
$10-$20 $30 $35 $40
$15-$30 $40 $50 $55
$20-$40 $45 $60 $65
$30-$60 $45 $80 $85
$50-$100 $30 $100 $110
$75-$150 $0 $125 $140
$100-$200 -$50 $125 $160
$150-$300 -$100 $125 $200
$300-$600 -$300 $0 $250
--
Erik Reuter, e-re...@uiuc.edu
Tarl Roger Kudrick
mmal...@ix.netcom.com wrote:
: The following table is from our new book GAMBLING FOR A LIVING written
: by David Sklansky and myself that should be available late this month.
: Limit Expectations (per hour)
: Very Good Selective Very Excellent
: Player Good Player Player
Let me be the first to ask the obvious:
What does "selective" mean in this context? How do I know if I'm a
"selective" player or not?
I also wish Mr. Malmuth had posted something about expected
standard deviation, so we could judge how much money to bring on a poker
trip...
--Tarl Roger Kudrick
------------------------------------------------------------------------------
|ta...@access.digex.net
"You get what you settle for." |
Thelma, in "Thelma and Louise" |I don't speak for my company. People
|who visit me can speak for themselves.
------------------------------------------------------------------------------
Greg Wohletz
mmal...@ix.netcom.com wrote:
) The following table is from our new book GAMBLING FOR A LIVING written
) by David Sklansky and myself that should be available late this month.
) It should make for interesting discussion. In addition, GAMING, POKER, &
) LIFE by David Sklansky will also be available later this month. Both
) books are produced and published by Two Plus Two Publishing.
) Limit Expectations (per hour)
) Very Good Selective Very Excellent
) Player Good Player Player
) $10-$20 $20 $30 $25
) $15-$30 $25 $40 $35
) $20-$40 $25 $45 $40
) $30-$60 $20 $45 $45
) $50-$100 $0 $30 $50
) $75-$150 -$50 $0 $50
) $100-$200 -$100 -$50 $50
) $150-$300 -$300 -$100 $0
) $300-$600 -$1,000 -$300 -$300
) Limit Expectations (per hour)
) Selective Excellent World Class Selective
) World
) Player Player Player
) $10-$20 $35 $30 $40
) $15-$30 $50 $40 $55
) $20-$40 $60 $50 $65
) $30-$60 $80 $60 $85
) $50-$100 $100 $80 $110
) $75-$150 $125 $100 $140
) $100-$200 $125 $125 $160
) $150-$300 $125 $150 $200
) $300-$600 $0 $100 $250
) Mason Malmuth
This is potentially interesting, but without some definition of what the
various player types mean and how these seemingly magic numbers were
derived the table is fairly meaningless. Perhaps you would like to
elaborate?
--Greg
hitth...@aol.com
>This is potentially interesting, but without some definition of what the
>various player types mean and how these seemingly magic numbers were
>derived the table is fairly meaningless. Perhaps you would like to
>elaborate?
>
>
I hope they didn't come from the same place
as the last batch of standard deviation #'s.
Best Luck
Ed
Michael Hesse
> >
> > Very Good Selective Very Excellent
> >$10-$20 $20 $30 $25
> >$15-$30 $25 $40 $35
> >$20-$40 $25 $45 $40
> >$30-$60 $20 $45 $45
> >$50-$100 $0 $30 $50
> >$75-$150 -$50 $0 $50
> >$100-$200 -$100 -$50 $50
> >$150-$300 -$300 -$100 $0
> >$300-$600 -$1,000 -$300 -$300
> >
> [rest of table chopped]
>
> Uh...I think this may require some explaination. Why should an excellent
> players expectation be negative at the higher limit games?
>
Because of the way the table is displayed, it may not be clear that
Mason has higher ranks than "Excellent" and his point seems to be that
there are enough world-class players at the top levels that players who
can make decent money in a $20 game cannot compete at the higher levels.
gj bart
sc...@bnr.ca (Stanley Chow) writes:
>I will make a wild guess that at the higher limits, even excellent
>pro's don't have the bankroll to ride out the variance. It is much
>more likely to get to the daily "stop-lost" limit so they book a
>huge lost.
>--
>Stanley Chow; sc...@bnr.ca, stanley....@nt.com; (613) 763-2831
Bankroll limits would not affect the individual's expectation. Most
likely he's just saying that you have to be a fantastic player to win at the
higher limits. Unfortunately, terms like "world class player" really
have little descriptive content, but are bandied about all the time in
poker literature. I think this group could come up with the "nut"
definitions of all the player types described in Mason's post, including
specific skills and their relative importance, if folks are willing.
gj
George M. Rice, Jr.
sc...@bnr.ca (Stanley Chow) writes: > In article <5i2rrt$eij$1...@news.iquest.net>,
> Steve Proctor <ste...@iquest.net> wrote:
> >In article <334474...@ix.netcom.com>, mmal...@ix.netcom.com wrote...
> >>
> >>The following table is from our new book GAMBLING FOR A LIVING written
> >>by David Sklansky and myself that should be available late this month.
> >>
> >> Very Good Selective Very Excellent
> >> Player Good Player Player
> >>$10-$20 $20 $30 $25
> >>$15-$30 $25 $40 $35
> >>$20-$40 $25 $45 $40
> >>$30-$60 $20 $45 $45
> >>$50-$100 $0 $30 $50
> >>$75-$150 -$50 $0 $50
> >>$100-$200 -$100 -$50 $50
> >>$150-$300 -$300 -$100 $0
> >>$300-$600 -$1,000 -$300 -$300
> >>
> >[rest of table chopped]
> >
> >Uh...I think this may require some explaination. Why should an excellent
> >players expectation be negative at the higher limit games?
>
> pro's don't have the bankroll to ride out the variance. It is much
> more likely to get to the daily "stop-lost" limit so they book a
> huge lost.
Or perhaps Mr. Malmuth is rating players in absolute terms. Therefor,
an excellent player will not be able to make money at this level because
the other players at this level are either very good, excellent or
world class!
Robert Copps
> References: <334474...@ix.netcom.com>
> Posted: 4 Apr 1997 20:35:44 GMT
>
> Org. : UNLV
>
> mmal...@ix.netcom.com wrote:
> ) The following table is from our new book GAMBLING FOR A LIVING written
> ) by David Sklansky and myself that should be available late this month.
> ) It should make for interesting discussion. In addition, GAMING, POKER,
> &
> ) LIFE by David Sklansky will also be available later this month. Both
> ) books are produced and published by Two Plus Two Publishing.
>
> [table deleted]
>
> This is potentially interesting, but without some definition of what the
> various player types mean and how these seemingly magic numbers were
> derived the table is fairly meaningless. Perhaps you would like to
> elaborate?
>
This has got to be for tough games.
mmal...@ix.netcom.com
Greg Wohletz wrote:
>
> mmal...@ix.netcom.com wrote:
> ) The following table is from our new book GAMBLING FOR A LIVING written
> ) by David Sklansky and myself that should be available late this month.
> ) It should make for interesting discussion. In addition, GAMING, POKER, &
> ) LIFE by David Sklansky will also be available later this month. Both
> ) books are produced and published by Two Plus Two Publishing.
>
> ) Player Good Player Player
> ) $10-$20 $20 $30 $25
> ) $15-$30 $25 $40 $35
> ) $20-$40 $25 $45 $40
> ) $30-$60 $20 $45 $45
> ) $50-$100 $0 $30 $50
> ) $75-$150 -$50 $0 $50
> ) $100-$200 -$100 -$50 $50
> ) $150-$300 -$300 -$100 $0
> ) $300-$600 -$1,000 -$300 -$300
>
> ) Limit Expectations (per hour)
> ) Selective Excellent World Class Selective
> ) World
> ) Player Player Player
> ) $10-$20 $35 $30 $40
> ) $15-$30 $50 $40 $55
> ) $20-$40 $60 $50 $65
> ) $30-$60 $80 $60 $85
> ) $50-$100 $100 $80 $110
> ) $75-$150 $125 $100 $140
> ) $100-$200 $125 $125 $160
> ) $150-$300 $125 $150 $200
> ) $300-$600 $0 $100 $250
>
> ) Mason Malmuth
>
> various player types mean and how these seemingly magic numbers were
> derived the table is fairly meaningless. Perhaps you would like to
> elaborate?
>
> --Greg
Since there is some confusion as to the meaning of the table I
thought the following might help:
"When you examine the above chart you will notice that we give win
rates for three skill levels. We have also separated them into selective
players and not-so-selective players. When we say a selective player we
mean a player who will only play the better situations. To be more
precise, we are going to call a selective player somebody who only plays
about twenty hours a week, even though he may spend many additional
hours finding those twenty hours worth of better games.
A not-so-selective player isn't nearly as picky. Although he doesn't
just sit in the first game he finds, and does at least try to find the
best game in the room, once he does find that game he will almost always
play."
The reason why some of the win rates at the higher limits are negative
is the fact that at the games at these limits are frequently very tough.
Mason Malmuth
Andy Morton
mmal...@ix.netcom.com wrote:
>
>
> "When you examine the above chart you will notice that we give win
> rates for three skill levels. We have also separated them into selective
[snip]
> The reason why some of the win rates at the higher limits are negative
> is the fact that at the games at these limits are frequently very tough.
>
> Mason Malmuth
Thanks for the table. It's especially interesting for those of us thinking of
playing higher, at least some day.
The table seems to indicate that if you're a 20-40 player, whether or not you're
selective, then the effect of stepping up to higher games will depend on your
skill level: if you're an excellent or world class player, you'll do better to
step up, but if you're only a very good player, you would lower your EV to step up
to a bigger game.
i often wondered why some professionals don't progress higher than 20-40; this may
be the explanation. Having explained what you mean by selective and
non-selective, can you offer some criteria to distinguish very good from excellent
from world class, other than waiting for enough play to distinguish them based on
results?
foldem
>speaking for himself mmal...@ix.netcom.com wrote:
<chart snipped>
> "When you examine the above chart you will notice that we give win
>rates for three skill levels. We have also separated them into selective
>players and not-so-selective players. When we say a selective player we
>mean a player who will only play the better situations. To be more
>precise, we are going to call a selective player somebody who only plays
>about twenty hours a week, even though he may spend many additional
>hours finding those twenty hours worth of better games.
> A not-so-selective player isn't nearly as picky. Although he doesn't
>just sit in the first game he finds, and does at least try to find the
>best game in the room, once he does find that game he will almost always
>play."
>
> The reason why some of the win rates at the higher limits are negative
>is the fact that at the games at these limits are frequently very tough.
>
>Mason Malmuth
This is somewhat interesting, but how selective can you really be at
the very high limits?
Of course, the charts would seem to confirm the belief on rgp that
table selection is a primary component of the successful player.
Paul R. Pudaite
> The following table is from our new book GAMBLING FOR A LIVING written
> by David Sklansky and myself that should be available late this month.
> It should make for interesting discussion. In addition, GAMING, POKER, &
> LIFE by David Sklansky will also be available later this month. Both
> books are produced and published by Two Plus Two Publishing.
Didn't see anyone else ask, so ... I take it these win rates are
also applicable to 7-card stud, not just hold'em? Might they also
be applicable to HORSE?
Also, can you provide some estimated win rates for $1000-$2000 and
$3000-$6000?
Someone asked if it's possible to be selective at the higher limits.
I can answer this one. From time to time, you will find wealthy
people playing in high stakes games, but they're really not that
good, at least for the competition they're up against.
For example, consider a "very good player" entering a typical
$300-$600 game. Sklansky and Malmuth estimate he'll lose $1000
per hour. That "very good player" will double the profit rate
of the "world class players" in the game (from $100 to $200),
raising their expectation close to that of "selective world class
players."
Now that I think of it, I have one more question. How much difference
is there between the win rates of a "world class player" and a
game-theoretically optimal player? Or are "world class players"
supposed to be so close to optimal that that the difference is
negligible?
Paul R. Pudaite
pud...@pipeline.com
Paul R. Pudaite
> The following table is from our new book GAMBLING FOR A LIVING written
> by David Sklansky and myself that should be available late this month.
> It should make for interesting discussion. In addition, GAMING, POKER, &
> LIFE by David Sklansky will also be available later this month. Both
> books are produced and published by Two Plus Two Publishing.
Consider the fate of a world class player who is trying to
assemble statistical evidence to back up his claim. Maintaining
a win rate of $100/hour when taking on all comers at $300-$600,
he needs to reduce the standard error of his mean win rate to
$50/hour to get "two sigma" confidence. Assuming an hourly
standard error of 10 big bets, it will take him 14,400 hours
(over 7 years at full time) to be statistically confident that
he's a world class player!
Paul R. Pudaite
pud...@pipeline.com
Tom Weideman
Paul R. Pudaite wrote:
<snip>
> Now that I think of it, I have one more question. How much difference
> is there between the win rates of a "world class player" and a
> game-theoretically optimal player? Or are "world class players"
> supposed to be so close to optimal that that the difference is
> negligible?
I like this question, but I think it needs a bit of clarification.
Since the topic of game selection came up, I have to question what is meant
by "game theoretically optimal". If you include knowledge you have of
weaknesses of other players in the calculations, then this would be optimal
play >>given what you know<<. I am sure this is what Paul meant. If
instead your calculations assume optimal play from all opponents (or even
just zero knowledge of weaknesses), then this is not optimal strategy, and
I can imagine a "world-class player" outperforming this standard when the
game is particularly weak.
For example, it is a mistake to bluff against a player who calls every time
(or even against a player who calls just a little too often, for that
matter), but someone employing game theory without any knowledge of the
other player would follow their "optimal" bluffing frequency, and get the
same profit as if their opponent were actually playing optimally. This
profit is lower than if the knowledge of the excessive calling of the
opponent was used.
Put another way, if you are a better card reader, tell catcher, and hand
disguiser than everyone else, then you should use these skills and not just
blindly employ the "proper" mixed strategies for various situations.
Outguessing your opponents will show more profit. If they all know you are
able to outguess them, then they should start employing the "proper" mixed
strategies. Of course they don't do this, especially if you have selected
a "good" (weak) game.
I tend to see game theory as a defensive tool, to protect someone from
players with better "information-related" skills. But certainly
information you have about the weaknesses of other players (e.g. they call
a bet on the river with a weak hand more often than is optimal) can be
incorporated into game theoretical calculations (giving additional
constraints to the minimax problems), and this adds greatly to profit.
Maybe Paul can clarify the question even further?
I hope this thread continues awhile longer, as I would love to read more
from Mason and others about how close the best players in the world come to
mathematically-defined perfection.
Tom Weideman
Scott F. Steele
Up here in canada we play cheap poker, usually 3-6-12 or 5-10 anystats for
this????
mel3...@aol.com
In article <pudaite-0604...@ip12.lansing.mi.pub-ip.psi.net>, pud...@pipeline.com (Paul R. Pudaite) writes:
>Consider the fate of a world class player who is trying to
>assemble statistical evidence to back up his claim. Maintaining
>a win rate of $100/hour when taking on all comers at $300-$600,
>he needs to reduce the standard error of his mean win rate to
>$50/hour to get "two sigma" confidence. Assuming an hourly
>standard error of 10 big bets, it will take him 14,400 hours
>(over 7 years at full time) to be statistically confident that
>he's a world class player!
Good point. Even then, he would have no way to know that he was not being selective in picking his opponents. If he were selective, then of course his results might still not lead to the conclusion
that he is world class.
Mel Brown
mel3...@aol.com
>I hope this thread continues awhile longer, as I would love to read more
>from Mason and others about how close the best players in the world come to
>mathematically-defined perfection.
>
>
Interesting, but for most of us the more relevant question is how the _less_ than world-class levels of play are defined. And how is selectivity defined?
Also, as someone has already noted, it might be useful to have some sort of a guide to standard deviations.
Another question: How do these tables vary with the number of players at the table? In particular, at the highest level, games typically feature fewer players at the table. At low to medium levels,
the numbers of players may vary drastically (e.g., as few as 5, or as many as 10). How does that affect expectation/hr?
Mel Brown
FredAces
>by David Sklansky and myself that should be available late this month.
>It should make for interesting discussion. In addition, GAMING, POKER, &
>LIFE by David Sklansky will also be available later this month. Both
>books are produced and published by Two Plus Two Publishing.
>
> Very Good Selective Very Excellent
>$10-$20 $20 $30 $25
>$15-$30 $25 $40 $35
>$20-$40 $25 $45 $40
>$30-$60 $20 $45 $45
>$50-$100 $0 $30 $50
>$75-$150 -$50 $0 $50
>$100-$200 -$100 -$50 $50
>$150-$300 -$300 -$100 $0
>$300-$600 -$1,000 -$300 -$300
>
[rest of table chopped]
I look forward to buying both of these books when they become available.
Since I live in Southern California, I mostly play at the clubs here
(Hollywood Park, Bicycle Club, Crystal Park, Normandie,etc... I don't play
at Commerce due to the insane collection rate increase as of Nov 1) and I
was wondering if these win rates are applicable to my region?
I know that games are *much* tougher in Nevada than in Southern
California. Also, do these win rates reflect collection and tokes? What
is the collection rate assumed to be? Also, in my region, the $30-60
Holdem games are not spread anymore. The $40-80 format has replaced it.
I've never played higher than $20-40 myself, but it seems that the $40-80
is in general looser than $30-60 because of the larger quantity of chips
in the pots. What is the win rate for $40-80 in Southern California?
It's becoming very popular now at this time.
Lfccone
Am I wrong to assume that the $15-$30 game would one of the best to play?
I would rather not risk the additional money with only a small possible
profit potential.Maybe I'm way out on my thinking...but should we look for
a game that gives us a better percentage of profit on our investment...or
should we gamble in a
$75-$150 game.After seeing the table ....I don't think I would want to go
to over $20.00-$40.. no matter how much money I had. I may be wrong,but
the $15-30 and $20-$40 games seem like the best game investments.
Sharpe222
I have a question regarding these win rates. First of all do they include
or discount
rake and or time charges and taking that into account what about the cost
factor
of playing in Atlantic City due to the higher amount of collections.I
kept records
at 5-10 and 10-20 Holdem for about a year and a half from the Taj Mahal
my
5-10 Holdem win rate was $17 per hour while my 10-20 win was $14 per hour.
I have theorized that the time charges combined with the relatively few
hands
dealt per hour contributed to these odd results. While it did seen that I
did get
better cards at the 5-10 table and the quality of opponents was better at
10-20
it wasn't that much of a difference. I played about 1700 hours of 5-10
and 700
of 10-20 also my deviations were far lower at 5-10 than 10-20 despite all
day
sessions my biggest loss at 5-10 is $250 whlie I have had 2 $1000+ dumps
at
10-20 and a 5-6 hundred daollar loss was not uncommon, I would appreciate
comments from all readers via- e-mail or reply's. I'm a young
Proffesional poker
player (played for a living since I was 17 in 1993) and would like some
advice
from players more experianced than me on my win rates and expected win
rates.
Ramsey
In article <19970505064...@ladder01.news.aol.com>, Sharpe222
<shar...@aol.com> writes
>I have a question regarding these win rates.
Your win rate, to mean anything to you, should be after rake & tokes.
This is the way that it is normally quoted in this group but of course
this makes comparisons difficult because the rake/toke differs between
casinos and players.
700 hours/1700 hours are relatively short periods for getting definitive
results. However what really affects your results is not normally the
skill level of the better players at the relative levels. Providing you
are a good player then replacing a good player by an expert player at
your table will hardly impact your win rate. The important factor is
how many donators are there in the game. If the norm in your 5-10 games
is 2 serious donators but it is only 1 in your 10-20 games than this
could easily explain why you are winning more at 5-10.
Also, if your 5-10 games are in general more passive than the 10-20
games then this could explain the smaller variance in the smaller game.
A max loss of $250 does seem small for anything but very passive games,
is your highest loss at any point during the game much higher than this?
The point being that provided the game has a positive ev for you and you
are prepared to play long enough you can 'guarantee' never having a
losing session :)
--
Ramsey
sjri...@sjrindex.demon.co.uk