|
|
Ok, going to note down anything I find interesting as I read through the section of the book. I've read it a couple of times in the past and skimmed it earlier this week, so this evening I'll read through the text and make notes in this post as I think of and come upon them. Once through that I'll go back to bigreds notes and try to write something specifically addressing those topics.
Introduction is worth reflecting on. While the big decisions are of obvious import (and indeed the book will go on to discuss in most details the decisions that may affect your whole stack - i.e. the big ones) the point is well made that it's getting all the small decisions right that make the difference between a professional and an amateur. Approaching poker with a professional mindset means understanding the situations and decisions and being committed to getting the small ones as well as the big ones right.
It occurs to me, reading the index of topics covered (REM process, planning hands around commitment) that the REM process is really something very basic that needs to be firmly ingrained as all advanced topics build on it. And I'm weak - not just in REM as a whole - I can do the equity and maximise things pretty well, but ranges that everything is based on I still don't do well enough and at my current level of skill any amount of time I spend on that is well spent and more is always beneficial. It's super crucial, but it is also mainly experience, and it's actually not the focus of the book. The tools that the book tries to teach almost presupposes that I can establish and mentally operate with reasonable hand ranges, and when I'm bad at that I'll get poor results trying to apply any more advanced techniques. This is not to say that it's bad to work on the E and M of REM or that I'll get nothing out of planning hands around commitment and thinking about SPRs etc - what it means is that I won't truly leverage that lesson until my ability to work with hand ranges has greatly improved. The positive side is that any improvements I make with regard to hand ranges will have a positive knock-on in everything else that I can and do, as it will all come on a much firmer foundation.
The Basics introduction kind of shoots itself in the foot imo. It suggests that knowing your opponent is popularly held to be important, but tends to be overrated, and people tend to forget that you need to be able to apply that knowledge of your opponent. Nowadays I think the topics with odds and outs are generally well understood, but knowing your opponent is not as well understood - or not as subtly and truly grokked as is beneficial.
One of the most fundamental poker skills is accurately estimating your odds
Bears repeating. Very few things are 100%, but that just makes it all the more important to estimate things as accurately as you can. Spoon made a classic post half a year ago offering free poker lessons with the message in it: Go read stuff by ISF until you stumble across something that confuses you. At this point - do not skip to something you understand, but instead put a couple of hours aside and work on that which confuses you and increase your understanding of it. Maybe start a discussion on the topic if you need help. We need to take that same attitude to estimation - the more something confuddles us and we feel lost, the more important and beneficial it is to take the time to try to sort it out. Only in that way can we improve our estimation skills, and estimation skills are absolutely crucial.
I note that the Outs section goes into this. 6 is also debatable it says in the example on page 11. The basic argument is that sometimes 6 makes you a flush while making someone else a full house, and other times it will make you a flush and noone a full house. Sometimes you'll hit 6 and people will fold to your obvious flush, other times someone will hit 2 pair or trips with the 6 and put more money in than they would otherwise. The amounts you win or lose as a result of 6 maybe being and maybe not being an out can vary quite a lot and because of that we may conclude that there is 90% that the 6 is a clean out, but when it's not clean it's so costly for us that we count the 6 as only 40% an out because that's what on average it's worth for us. The important thing here is not to get stuck. Or distracted if you will. The diversion I started on here I could (and would, personally) go on for pages about to try to quantify more accurately, but it's one of those areas where the important thing is not to get it right, but to get it approximately right and then keep moving. If I decide it's 50% and then move on, I have a number I can calculate with and I can come up with answers. Then as my experience builds and as this sideline becomes more obvious to me and I find ways in which I feel I can calculate it with a bigger degree of certainty I might be able to put together a calculation that tells me that I would expect it to be between 50% and 75% depending on the opponent, and I'd know which opponent tendencies make it go in one direction or another and I'll have an almost instinctive understanding of what the value of this out should be in any situation. But that's something that comes after experience - for initial learning it's important to make a decision on an approximation, work with it and move on to learn more of the complete picture, then these fiddly bits can be revisited later on when they are more easy to grapple with.
You have to be able to mentally accept that something you recognize as an approximation is literally good enough - or in fact optimal for learning as accepting it lets you put it from your mind and focus fully on the application of the principle, whereas dwelling on the accuracy of it, as I would tend to do, could be described as tilt.
Discounted outs presents another simplistic example of the same thing, but I think it's important to understand why a gross approximation shouldn't nag the perfectionist in me but rather be accepted as a tool. Also it's worth noting that a 9 out flush in the example is discounted down to 5 outs. As I mentioned in my example - not because you will only have the best hand half the time that a flush out comes, but because the times you don't have the best hand you will lose a big pot, and the times you do have the best hand you probably won't win a big pot. These are fantastic examples of how you can take all kinds of factors into account and express them as outs for a quick calculation.
It's estimation and meant to get us into the right ballpark - something much more fundamental and important is the hand range (which is not the subject of this section) because even small changes to a hand range or a players tendencies can greatly influence EV calculations. In the view of that whether 9 outs are 36% equity or 35% equity is almost completely immaterial.
There's one implied odds method I stumbled on and which I've posted a couple of times but which I've never found in literature which I think is worth learning. Since the book here delves into implied odds and I don't have much to say about that section as presented I'll say my piece here.
It's strange, because it's obvious. Basically we are betting to win the following
1) The current pot
2) The opponents stack
3) The average estimated payout - decided by implied odds.
Commonly we find that we relate the bet size to the current pot for your pot odds and the bet size to the opponents stack to find stack odds. But for some reason people tend to get all backwards when they discuss implied odds and go into obscure odds notations.
Here's what I think (using flop as an example). Pot is $X, bet is $Y, I think 10 cards (adjusted outs) will make me the winner on the next street. So my odds to win are 37 to 10 - or 3.7 to 1. To be profitable I need implied payout odds better than that. In other words, I need to win ($Y * 3.7) - $X on average on later streets for the call on this street to be profitable. It's a dead simple calculation when facing a bet and I can't understand it's not more widely used or documented.
Pot: $30 (pre-bet).
You are facing a bet of $20 (2/3 PSB).
You look at your absolute and relative hand strength, decide you are behind but you have 5 adjusted effective outs to win.
42/5=8.4.
($30+$20) * 8.4 = $420.
You must win $420 on average for calling to be correct for implied odds. $50 of these are already in the pot - for the purpose of this calculation the $20 you'll be putting in the pot do not count, but if you do call you need to be mindful that the pot on the next street will start out as $70 and will affect the bet sizes on that street. Consider how many streets of betting are still remaining and the tendencies of your opponents, but first do this: Consider the effective stack sizes. If the effective stack sizes are less than $370 after the call this is never a good call for implied odds.
Pot: $40 (pre-bet)
Bet: $25
Outs: 9
38/9=4.22
$25*4.22 = $105.5
Needed to extract on later streets (average): $105.5-$65 = $40.5
Starting pot on the next street: $90
Effective stacks? (Hopefully above $40.5)
The only piece of information here that you are not being spoonfed is the number of outs you have, and that is where estimation comes in and why estimation is so important. The rest is really really trivial. Do the calculation a couple of times, try doing it at the table a couple of times and I think you'll find it useful.
Traditional methods has you calculating your odds of winning, comparing them to your pot odds and determining you need "a bit more than what's in the pot" - how is that helpful? This method tells you exactly how much money (on average) you need to extract on later streets for calling on this street to be profitable for implied odds - this is a number you can easily relate to the pot size on later streets and what's behind in your stacks. Keep in mind that it's on average. Sometimes you'll hit your hand, bet it and get only folds.
Going to use the implied odds example from page 18 but calculating it using my method.
Pot: $32 (including bet and call)
Bet: $10
Odds to hit 5-1
Amount to extract later: (5 * $10 = $50) - $32 = $18
Need to extract average $18 later, pot will be $42 going into next street, so less than a 1/2 PSB, pot is multiway, drawing to the nuts, draw disguised etc.
Now what I like to do here is 'bank' the $18. As I play out the hand I remember that I'm already $18 behind, so I plan my play of the later streets around this fact - particularly if I hit, I don't think he'll call a big bet, I bet only $10 and he calls: I've just made me calling the flop for implied odds the wrong decision.
I may end up in a situation where I have one amount banked on the flop and another amount banked on the turn and on the river I need to make good on both those amounts added up or the two implied odds calls together will be -EV.
Ok, continuing with the hand.
Pot: $82 (including bet)
Bet: $40
Odds to hit: 2.6-1
Amount to extract later: ($40 * 2.6 = $104 ) - $82 = $22
Pot will be $122 as we go to the river, and we need to extract the $18 from the flop call and $22 from the turn call on average for the two implied odds calls to be 0 EV. Preferably we want to extract much more. But since this is a mere third of the pot size our chance of getting that money out of the hand is pretty good.
So yeah, the book goes on to say that you will win on average $80 and gee that means you need certain odds etc etc and golly we find that you are then profitable because of implied odds! But the books way of looking at it requires you to first estimate this average that you will win. My approach gives you a target on which you can do sanity checks: If the number is insignificant compared to the pot size it's pretty much a given that you'll get it. If the number is bigger than the money behind it's a given that it will never be profitable etc. I think it's more effective and simpler to do at the table. And I think it's important to keep the banked amounts in mind also - the book will later have an example where on a given street the first decision (a bet) is profitable, the second decision (a call is profitable) but the whole street (the amount of money going in with this hand against that opponent hand range) is NOT profitable. I'm using similar logic here over multiple streets.
The sections on bet sizes and stacks sizes are pretty obvious. I wouldn't go too far to recommend min bets, and I think linking stack sizes to amount of big bets to get all in is profound and useful, but it's really just laying the groundwork for the rest of the book so not really something to go into detail about here.
Ok, time to come up with a response to bigreds notes:
1) I went into some detail about discounting outs above, and I think for a beginner it's perhaps counterproductive to dwell too much on it. A beginner needs to find a reasonable rough estimate (which isn't halving the outs every time) for the situation and then move on it. It's very situation dependent, and sometimes outs can be discounted all the way down to 5% or less.
2) Well, future streets are in the future. That means they're unknown. Also the bets on those streets are based on the pot size, and the pot only grows bigger street by street. So it's logically less important what your equity is on this street and logically more important what your equity is on the later streets and any considerations of implied and reverse implied odds just helps us identify situations where we need to lay down the best hand and invest money in a hand that if it improves rates to give us a lot of profit.
3) Betting the pot makes it less profitable to bluff - you risk more money on a hand that doesn't rate to be a favourite and when called on it you pay the price. The advantage is that you get more money out of your value bets. However, because you bluff less often people will give you more credit for made hands and will call you less often also (which, yes sets up more bluffs in a never-ending cycle). With bigger bets it's easier to make a mistake when it comes to anticipating the opponents reaction.
3) (?) Stack sizes are taken into consideration in every decision on every hand. I may limp, raise or fold hands pre-flop and indeed on every other street based on the stack sizes of the people in the hand already and the people left to act.
4)
a) With deeper stacks I'll be more inclined to slowplay the nuts rather than just bet bet bet.
b) With deeper stacks I'll be more inclined to bet my big draws fast as the implied odds are often good, and getting an opponent pot stuck or married to a hand can make it easier to extract value when I hit
c) With deeper stacks I'll be especially aware of money behind when bluffing - the money behind can be a bigger incentive to fold than the size of the bluff, so I may occasionally be able to use more optimal (small) bluff sizes as long as I can convey the threat of two or three more big bets if called.
d) With deeper stacks I'll be more inclined to play thin value hands slowly and more inclined to throw them away on earlier streets if faced with aggression.
5) No, stack limit hold 'em is a much better name because it more exactly describes the way the game plays. But I guess this is just a US thing where they also name a game football that is played mostly with the hands. I like the sarcastic "unlimited hold 'em" because it exposes the fallacy in the "no limit" part in a different way. But hey, I also think limit hold 'em is misnamed. That's more fixed-bet hold 'em than anything to do with limits.
|
PNL ~ Week 1 ~ Part 1: The Basics (pp. 1-43)
Reply With Quote
Originally Posted by ISF
and the board is 
I'll usually only discount the 
and say I have 8 outs instead of 9. I usually don't even really consider my A as an out, but maybe I should think about that more often and add it in.
, K
(5 players)
