Raising after many loose limpers
Abdul Jalib
your fair share of pots. For example, if there are 5 limpers in
the pot, you're on the button, and you expect the blinds to call
with any two cards, then if you will win more than 1/8 of the time,
you should generally raise.
To get a better idea for the threshold for raising versus *loose*
limpers, here are the hands that win 1/3, 1/4, 1/5, 1/6, 1/7, and
1/8 of the time against 7 opponents with *random* hands in a *no*
*fold'em* game:
No Fold'em Hand Rankings for 7 Opponents with Random Hands
Wins AA AKs KQs QJs JTs T9s 98s 87s 76s 65s 54s AK KQ QJ JT T9 98 87
==== == === === === === === === === === === === == == == == == == ==
1/3+ AA ... ... ... ... ... ... ... ... ... ... .. .. .. .. .. .. ..
1/4+ JJ AQs ... ... ... ... ... ... ... ... ... .. .. .. .. .. .. ..
1/5+ 99 ATs KTs QTs JTs ... ... ... ... ... ... AQ KQ .. .. .. .. ..
1/6+ 66 A2s K5s Q8s J8s T7s 98s 87s ... ... ... AT KT QT JT .. .. ..
1/7+ 55 K2s Q4s J6s T6s 96s 86s 75s 64s 54s A3 K8 Q9 J9 T9 .. ..
1/8+ 22 Q2s J2s T2s 93s 83s 73s 63s 53s A2 K5 Q7 J8 T8 97 87
Each hand listed also includes all the hands between it and the column
heading. For example, K5s means K5s through KQs. K5s is on the 1/6+
line, so that means it wins 1/6 of the time or better (but less than 1/5
of the time.)
Now, it's not true that you could profitably raise with the hands on the
1/8 win rate line, even though they include hands with win rates close
to 1/7. Nor could you profitably raise with most of the hands on the
1/7 win rate line. Your raise requirements must be tighter for the
following reasons:
* Your opponents are not playing any two random cards, only close,
though this is mostly offset by the fact that they would have raised
with their best hands rather than limping.
* There is a risk of a reraise behind from a superior hand left to act.
* The game is not actually no fold'em, and so hands that will often
be folding before the river will win less often in reality than in
no fold'em (e.g., 22.)
So where should the line be drawn? Full blown Turbo simulations
(*not* no fold'em) of raising on the button after 5 loose limpers
(and loose players in the blinds) suggest minimums of:
66 A2s K5s Q5s JTs T9s 98s 87s 76s 65s ... A9 KT QJ .. .. .. ..
This is very similar to the 1/6+ line...
1/6+ 66 A2s K5s Q8s J8s T7s 98s 87s ... ... ... AT KT QT JT .. .. ..
...but with more emphasis on suited connectors. A gapped hand like T7s
has less chance of flopping a draw than a hand like T9s, and so should be
more inclined to see the flop cheaply upon which it will often fold.
Suited connectors are very likely to flop something that causes them to
stay in for the turn, and once on the turn they are likely to see the
river, and once on the river they may have won. So, you actually can
raise with very weak suited connectors like 76s and sometimes even 65s.
Raising with them has the additional benefit of building the pot and
tying others in for your draw.
After limpers, common wisdom is to not raise after limpers with AQ, never
mind KJ or KT. This thinking is generally wrong. Big offsuit hands like
AQ and KJ should not be afraid to raise preflop after loose limpers, as
such hands are indeed very likely to be best at the moment and to win,
even if they are likely to get tortured late in the hand and not win very
much on average. AQ wins 1/5 of the time in 8-way no fold'em, probably
considerably more than that in real games since AQ is extremely likely
to take a card off with overcards on the flop, whereas your opponents
with J7 are likely to fold on the flop. So you are getting the difference
in equity of 1/8 and at least 1/5 from each player when you raise. It
amounts to at least 60% of a small bet, probably closer to a full small
bet, which is a big chunk of your hourly win rate.
However, if your raise would cause all 7 opponents to see the turn,
whereas only 3 would see the turn otherwise, this is one case where
you might not wish to raise with AQ. The reason is that you can't
usually call on the flop with overcards versus 7+ opponents. (For one
thing, you can't count on making a pair on the turn and then having the
board pair low on the river and winning with aces up or queens up, because
your opponents might already have paired every card on the board.) So,
this is one strange case where by raising AQ might so harm its chances
of winning that it would turn itself into an implied odds hand that
would prefer to have not raised. Few games would fall into this
category, however, and normally you should raise preflop after limpers
if you feel you have the biggest cards.
IMO. Comments and criticisms are welcome.
--
Abdul
E4429G
>*fold'em* game:
Abdul:
I have had questions in my mind about all
so called random hand analysis. I am sure that your calculations are correct if
people did in fact play random hands, however, I believe that any calculations
that are based on random hands are inherently wrong in the REAL WORLD of poker
playing. The reason is that some of the hands - the famous 7 2 off, for
example- are not played ( except for BB) and certainly not for a raise. By
including many many hands that are Never played in the calculation the
win/loss ratios are distorted.
I realize that it is hard if not impossible to run simulations with out the
random hands. Never the less, I would like to think that it would be possible
to create some sort of analysis of various hands with out all the 72 off , 93
off and similar hands. I think that such a simulation will give a more
accurate results.
BTW, I enjoy your postings very much. From reading your comments I get the
impression that you play against top notch players that know what they are
doing.
I hope that one day you will play 5-10 or 10-20, and see what it is like to go
against players that don't seem to respond to anything you do or to the board.
Regards,
Robert
GaryCarson
never played. I've played with lots of players that will
habitually call preflop raises in a multi-way flop with
72off when they'd fold 83s. I've seen such players in 2/4
games and in 20/40 games. These are the same players who
will also play every 96 they get. They'll also play 10 4.
If you tell them the real Brunson hand was 10 2 then
they'll start playing that too, just in case you're right.
The reason they do is for the dramatic effect for when they
get lucky. Not everyone plays poker for the same reasons
you do. Some of them seem to play primarily because they
sometimes get an oppurtunity to giggle when they get lucky.
Gary Carson
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
Abdul Jalib
> I am sure that your calculations are correct if
> people did in fact play random hands, however, I believe
> that any calculations that are based on random hands are
> inherently wrong in the REAL WORLD of poker playing.
> The reason is that some of the hands - the famous 7 2 off,
> for example- are not played ( except for BB) and certainly
>not for a raise. By including many many hands that are Never
> played in the calculation the win/loss ratios are distorted.
In addition to the no fold'em results, I presented Turbo results,
which are full simulations of playing out the hands with betting.
Most people disregard Turbo simulations as well, but if not
no fold'em simulations and no Turbo simulations then what?
Peoples' intuitions stink at high odds low probability situations,
which is what we're talking about.
> I realize that it is hard if not impossible to run simulations with
> out the random hands. Never the less, I would like to think that it
> would be possible to create some sort of analysis of various hands
> with out all the 72 off , 93 off and similar hands. I think that
> such a simulation will give a more accurate results.
Already done in my post. In the Turbo sim, not only do the limpers
not have 72o, but they don't have AA either. (I would have to
check to see the actual range that they could have.) However, that's
not realistic either, as 3-6 players do often play 72o, as Gary pointed
out.
> BTW, I enjoy your postings very much. From reading your comments I get the
> impression that you play against top notch players that know what they are
> doing. I hope that one day you will play 5-10 or 10-20, and see what it
> is like to go against players that don't seem to respond to anything you
> do or to the board.
I started at 1-2 in California. I occasionally play low to remind
myself, though unfortunately even the 1-4-8-8's in Vegas are relatively
tight. However, I do plan to make a journey to L.A. soon and remind
myself what California 3-6 was like. A year ago I played 4-8 when stuck
in Sacramento for one night.
Anyway, the whole point of my analysis was to try to capture
what to do in these games after players limp in with almost any
two cards. I wouldn't use no fold'em simulations for tight game
situations, but I think they have some value for ultra loose game
situations.
--
Abdul
GaryCarson
good baseline to start applying some judgement too. I can
certainly see a benfit to using no-fold'em baselines. Using
TTH sim results tends to help that judgement be a little
better informed than it would be otherwise.
The things is, unless you know some precise stuff about the
holdings of specific opponents, nothing is going to give
you the one true answer. Whether you use TTH sims, table
experience, analysis of some simple models, some
combination of all three, or some patented brand of logic
available only to those who made 800 on their math SAT's,
you're still going to have to apply some judgement at the
table. The trick is to prepared and make sure your
judgement is informed as you can make it.
ADB Iceman
Abdul Jalib wrote:
>
> I've said that you should raise preflop when you will win more than
> your fair share of pots. For example, if there are 5 limpers in
> the pot, you're on the button, and you expect the blinds to call
> with any two cards, then if you will win more than 1/8 of the time,
> you should generally raise.
>
> To get a better idea for the threshold for raising versus *loose*
> limpers, here are the hands that win 1/3, 1/4, 1/5, 1/6, 1/7, and
> 1/8 of the time against 7 opponents with *random* hands in a *no*
> *fold'em* game:
I think the posting is excellent to start the process of understanding.
Poker is strange in that there is no rules that "ay you must always" or
"you should never". Even in this case of defining a complete environment
of loose passive, there are some considerations.
One thought not here in a simulation is when one does hit with a 2-2
against A-A and maybe K-J raised.. it pays off in implied odds. See my
BARGE trip report. A-A and a A-Ko, and A-Xs along with 3 other players
participate pre-flop capped. I brought in the second raise UG after
calling with 2-2 in a loose pasive game. When the flop came 10-6-2 the
A-A read my raise for A-10s and not 2-2. The A-X folded the turn but the
A-K called until the river. You know the A-A attacked with vigor on the
turn when a '3' showed. There is nothing threatening in his eyes since
HE KNOWS I would have raised 10-10 UG intially and I would not play any
other holding but paint. I had to have A-10 suited! I changed gears and
the payoff was there!!
Simulations just do not demonstrate this kind of emotion and play. In
the simualtion The 2-2 does not raise after a call, and does not call
the cap 7 ways. The simulation does not mess with the person holding A-A
mind with previous advertisments of only paint UG.
Just a thought.. The posting is again excellent for disscussion. Most
begining players should stick to these guidlines like this until a
better understanding is gained. Just like David Sklanski's hand classes.
7-7 is a class V hand but it is not the same hand as A-Xs which is also
class V hand. But ranking 7-7 as class V and generally saying class V
hands should not be played UG or earlyis a good start.
ADB Iceman.
GaryCarson
.>>>>>>>>> Simulations just do not demonstrate this
not raise after a call, and does not call the cap 7 ways.
The simulation does not mess with the person holding A-A
mind with previous advertisments of only paint UG.
>>>>>>>>
Actually, I have done TTH simes where the hero limped-
reraised with 22 and cold called 4 from the button.
I used to have some homemade profiles of some of the
regulars in a real wild/crazy game I used to play in.
Against that lineup, cold-calling 4 with any pair was right
and, depending on the paritucalar lineup and position of
the players, it was sometimes right to limp re-raise with
22.
Some of those players would never be convicned that AA
wasn't the guarentted nuts and their playing profiles did
reflect that.
You can look at a lot of non-standard things with TTH.
Lee Daniel Crocker
> for one night.
"Stuck"? Some of us actually live here by choice,
you know. :) Games there change a lot, so next time
you're in town, drop me an email and I'll tell you
where the good games are.
--
Lee Daniel Crocker <l...@piclab.com>
ASA DEAN C
Thanks for the numbers. Did you do the odds yourself or are you
referencing? If referencing, where can I find this?
Thanks again,
Asa
tonyj...@home.com
hours of flames worthwhile.
Tony
m...@acm.org
> In addition to the no fold'em results, I presented Turbo results,
> which are full simulations of playing out the hands with betting.
> Most people disregard Turbo simulations as well, but if not
> no fold'em simulations and no Turbo simulations then what?
> Peoples' intuitions stink at high odds low probability situations,
> which is what we're talking about.
Personally I hope it is "Most people" that disregard tools like Turbo
simulations. Their loss.
In any case I hope you realize how many people are encouraged and
enlightened by your shareing of information.
many thankx,
mph
Izmet Fekali
> Thanks for a great post Abdul. A post like this makes reading thru 4
> hours of flames worthwhile.
Oh so true... but Abdul's been busy on 2+2 lately, as he enjoys teasing
that S&M duo with is presence. Apparently he enjoys posting in
HTML format also.
I might get flamed for directing people over to 2+2 site, but I never had
a problem with recommending good content. Take a look at this:
http://www.twoplustwo.com/cgi-bin/xforum.pl?read=27729
I guess that probation period on 2+2 is soon to run out, Abdul ;-)
--
Izmet Fekali
Burek Experts Ltd.
Catering the World since 1389!
GaryCarson
words. Obscene pictures were not part of the terms.
Gary Carson
GaryCarson
really liked the part where David says that betting the
best hand, when there almost no chance that a better hand
would fold but a good chance a worst hand would call, is
bluffing.
It appears that David got so flustered in trying to defend
Masons passive check/calling that he just got tounge-tied.
Pretty funny stuff. I even broke my vow of non-posting.
Roger Kirkham
Wisdom" folder...
A question from the aquarium; If our hero is holding top pair/weak kicker on
the flop, and an over-card falls, and one of the (many) opponents bets, does
our hero fold?
Let me try and think up an example from your tables:
Okay Our 'Abdulian' hero holds King, 10 suited. Flop comes something like 10,
8, 2, mixed suits, our hero (presumably) bets and three or four of his seven
opponents fold. The remaining players call. Then (damn it) the turn card is an
ace, and immediately one of the opponents bets...
Does our hero fold, call or raise? I know 'it depends' but I wonder what (on
average) our simulated hero does in this situation?
I suppose the general question is how well 'our hero' plays after the flop,
having bet or raised with the hands you describe?
Thanks,
- roGER
ADB Iceman
GaryCarson wrote:
>
> iceman said
> .>>>>>>>>> Simulations just do not demonstrate this
> kind of emotion and play. In the simualtion The 2-2 does
> not raise after a call, and does not call the cap 7 ways.
> The simulation does not mess with the person holding A-A
> mind with previous advertisments of only paint UG.
> >>>>>>>>
>
> Actually, I have done TTH simes where the hero limped-
> reraised with 22 and cold called 4 from the button.
>
I believe I see your point. You have always interesting posts that I
enjoy reading. Thanks for your time and interactions!!
The actual situation can be simulated I agree. What I believe can not be
simulated, given todays technologies, is change in emotion. Both a
single player's emotion and the table as an entity. Unless the players
have frontal lobotomies or they are really really good -- they *MIGHT*
react differently the next time the cards are dealt. In the example I
provided, A good local checked and called me with top pair top kicker no
threat board (I could have the trips) rather than raise or bet all the
way to the river after I played the 2-2 unexpectedly UG. The number 6
seat, was inexperienced and went from "painted rock" and emulating the
locals, to playing "loosely goosey" after he saw me take in this huge
pot playing 2-2 UG. The entire table changed emotionally. The local
rocks grumbling was overtaken by joyous laughter and a feeling "I have
come to gamble".
The A-A guy thought he had me read and now he is confused. This really
affected his decision making. I do not believe TTH has a "HAL 9000"
module that goes on tilt or changes behavior suddenly because the mood
of the table changes. This is where I believe experience comes in. If a
beginning poker player sees the simulation but can not connect the
intangibles they will not understand why this play worked for this
table, this time, for this reason. Nor will they understand about making
a loose play and giving some action for future investments.
I do believe you can change TTH to go from painted rock to loosey goosey
but the user must make that change. The program does not have a module
to say X% of the time in this situation the following changes will occur
to the emotional aspect of decision making...
Comments??
GaryCarson
conditions. And, not TTH won't simulate changes in game
conditions.
You can simulate different game conditions and experiment
with how things change under different conditions.
But, detecting a change in game condition is very much an
art I think. Lot's of things can cause a change in the
mood of the table, or game condition or whatever we call it
-- some of them external to the table. I've got a seperate
chapter in my book on just this one thing of the kinds of
things that can trigger a dramatic change in table
conditions. I'm not talking about putting someone on tilt,
I'm talking about the entire dynamic of the table.
That book btw, is not going to be out in December. I don't
know when it's going to be out. Carol (the publisher) is
going thru some bankruptcy type reorg and all books
currently in production are postponed indefintely. There
is another publisher who's interested in buying the book,
and I've talked with a lawyer about trying to get the
rights ceded back to me, but it doesn't look real promising
for the short term. But, I do still own serial rights and
internet rights, so I'm going to be doing some other form
of distribution. I'm just not going to be doing anything
with it for a few weeks.
ALT
primarily because they sometimes get an oppurtunity to
giggle when they get lucky.
This is one of the best and most accurate lines I have read
in a long time.
G. M. Cooper
1. Everyone but you and the turn bettor folds (you're heads up).
2. Even if the bettor is bluffing, he will bet the river.
It turns out here that if he doesn't bluff the river, you
*always* have a borderline call.
3. You implied with 8 preflop players there is no raise (?!).
There are 8 people to see the flop (4 big bets) + 4 see the turn
(2 big bets) + 1 turn bet for a total of 7 big bets when you must
decide whether to call. If the bettor is not bluffing, you have
7 outs to make your hand (2 tens, 3 kings) or 1/6.4 cards left
(since we are assuming he has Ax, we know 7 cards and don't know 45).
If we require 7:1 pot odds to make up for times we improve and still
lose (say to A8; AK is unlikely if we believe he would have raised
preflop)
then we have a borderline call for 1 bet (EV = 0).
But if we have decided to go all the way and pay off a river bet, we
must add in a bluffing percentage X for him.
X we win (7 current + 1 river bet) = 8*X
(1-X) we win 8 bets 1/7 of the time we hit and lose 2 bets 6/7 when we
miss.
EV = 8*X+(1-X)*(8*1/7-2*6/7) >= 0 when X >= 1/15
So if the bettor bluffs > 1/15 of the time in this spot, you can
mathematically call.
Gary
P.S. Don't argue with Barbara Yoon about whether KT has a weak kicker
;-)
G. M. Cooper
"G. M. Cooper" wrote:
>
> 7 outs to make your hand (2 tens, 3 kings) or 1/6.4 cards left
^
5 oops. You had better fold. Besides, is a bluff likely with
four players? In the loose games I'm familiar with, overcards
stick around on a ragged flop and you'll be shown an Ace 90%+
of the time.
Gary
Abdul Jalib
> Thanks for another great post, destined to be immediately filed in my
> "Abdul's Wisdom" folder...
>
> A question from the aquarium; If our hero is holding top pair/weak kicker on
> the flop, and an over-card falls, and one of the (many) opponents bets, does
> our hero fold?
>
> Let me try and think up an example from your tables:
>
> Okay Our 'Abdulian' hero holds King, 10 suited. Flop comes something like 10,
> 8, 2, mixed suits, our hero (presumably) bets and three or four of his seven
> opponents fold. The remaining players call. Then (damn it) the turn card is
> an ace, and immediately one of the opponents bets...
>
> Does our hero fold, call or raise? I know 'it depends' but I wonder what (on
> average) our simulated hero does in this situation?
My inclination would be to raise and then check it down on the river
if I do not improve. If there were 8 players going into the turn and
several were left to act, I would probably fold. If I'm heads up with
a rock, then I'd have to think about whether I had odds to call for
my potential 5 outs to suck out on an ace.
If you're asking what a Turbo player would do, I'm not sure, but Turbo
plays very aggressively and fairly tenaciously, so it might raise, but
I'm guessing it would just call since it didn't have an overcard kicker.
It might even fold if it were not last to act. In general, my feeling
is that Turbo plays fairly reasonably heads-up, but I wouldn't
necessarily trust it for loose multiway pots, especially not the
Advisor_T player, though it's better than a no fold'em simulation.
--
Abdul
Heldar
"I might get flamed for directing people over to 2+2 site, but I never had
a problem with recommending good content. Take a look at this:
http://www.twoplustwo.com/cgi-bin/xforum.pl?read=27729 "
What a wonderful thread that was! Humorous yet wise.
But I'm still waiting for Abdul to post a picture of his wife!
Roger Kirkham
> plays very aggressively and fairly tenaciously, so it might raise, but
> I'm guessing it would just call since it didn't have an overcard kicker.
> It might even fold if it were not last to act. In general, my feeling
> is that Turbo plays fairly reasonably heads-up, but I wouldn't
> necessarily trust it for loose multiway pots, especially not the
> Advisor_T player, though it's better than a no fold'em simulation.
Thanks, I was asking what the Turbo simulated player is 'likely' to do in such a
situation, in order to get some idea of how accurate the simulation is.
Thanks again,
- roGER
Robert Sherwood
stuckER....
the Bay 101 kid.....:-)