|
||
|
||
|
Server Time: 11/20/2008 9:52:18 AM PACIFIC |
 A Starting Hand Points-Count System for Hold 'Em, URF, 5. Sep 2003 02:05 | ||
| View ( Message | Thread ) | Return to Thread List | |
| Reading noiseboy's post below ("An easy way to think about starting hands"), I thought maybe it would be a good time to put my recently deviced hold 'em starting hands points-count system out for public scrutiny. First of all: this is just a rudimentary system, a first draft, so please judge it by its potential more than what it might be right now. I am very certain that there are big errors and omissions in it, but that is where you all come in: let's improve the system together. But remember: it is supposed to be a system based on analysis, not just intuition and rules of thumb. The system is similar to what is in general use in the game of bridge. I have not seen any similar system for poker, but maybe there is, and I have missed it. OK, let's get straight to it and present the system first, and then analyze it. -------------------- As far as I know, there are four ways in which a Texas hold 'em pocket can have value: high card value, flush draw value, straight draw value, and set draw value. In my system in its current form, these four dimensions of value are given points in the following manner: HIGH CARD POINTS: You count points for high cards differently depending upon the type of game that you are in. If it is a "flop based" game, i.e. a game where people fold a lot on the flop, and the rest of the hand is a contest of made hands and solid draws and bluffing, i.e. a tight game, high cards have higher values. If it's a "river based" game, i.e. lots of players stay in on shaky holdings and shady draws, i.e. a loose game, high cards have lower values. In between these extremes, we have "turn based" games, which are, well, inbetween! Here are the points given: A: 9 9 9 K: 7 6 6 Q: 5 4 4 J: 4 3 2 T: 3 2 1 9: 2 1 1 8: 1 1 7: 1 AA: 52 52 52 KK: 40 37 34 QQ: 31 25 21 JJ: 22 17 12 TT: 16 10 7 99: 11 6 4 88: 7 4 2 77: 4 2 1 66: 2 1 55: 1 These are rounded off values to make it easier to count in your head. With a little more precision we have: A: 9 9 9 K: 7 6.5 6 Q: 5 4.5 3.5 J: 4 3 2 T: 3 2 1 9: 2 1 0.5 8: 1 0.5 7: 0.5 6: 0.5 So, in a tight game, AJ gets 9+4=13 points. In a loose game AJ is worth 9+2=11 points. The high card value of JJ in a tight game is 22 points, but in a loose game only 12 points. FLUSH DRAW POINTS: The value of having two cards of the same suit in your pockets varies with the number of opponents and how much you have to call. The basic value for an unraised pot with four or more callers is 7 points. With no callers, you still get 2 points. Inbetween you get inbetween points, e.g. (simple to count): 0-1 callers: 2 points 2-3 callers: 5 points 4- callers: 7 points If the pot is raised, you divide the points by two. If you are in the small blind and the pot is unraised, you multiply the points by two. STRAIGHT DRAW POINTS: The value of having two cards with which you can flop an open ended draw behaves much like the points that you get for having a suited pocket, but with an addition for in how many ways that you can flop a draw. If you have four or more opponents, and can flop a draw in: 3 ways, you get 6 points 2 ways, you get 4 points 1 way, you get 2 points And then, it varies with the number of opponents, just as in the case of flush draw points: 0-1 callers: 1/3 times the above points 2-3 callers: 2/3 times the above points 4- callers: the above points Remember, these are values rounded off to make it easier to count in your head. It could be made more exact. SET DRAW POINTS: With a pocket pair, you can flop a set. The value of this possibility behaves as the other two draws, but is much stronger. You get: 0-1 callers: 6 points 2-3 callers: 12 points 4- callers: 18 points Once again, you could have a finer division for the number of callers, and more exact rounding off. -------------------- OK, that's it! Now let's discuss it a bit further. WEAK POINTS There are certainly weak points in the system as it is right now. E.g., it doesn't consider the fact that some flush draws are stronger than others. 32s is valued as strongly as Axs (flush draw wise). This could of course be corrected, I am certain. If it would make any practical difference, I am not sure of, because I have not investigated it yet. Also, I am not sure if the high card points really should just be added together, or if there should maybe be some additional points count for kicker quality. But on the other hand, the kicker of one card is the other card, and you do count both... Well, I just don't know. I haven't analyzed the question enough yet. But it doesn't seem quite right that A2 gets the same value as A8 in a loose game, does it? Oh well, as I said, it's only a rudimentary system for the moment. And then I use the high card value of an ace as the fixed point of the system (i.e., the value of an ace is always the same). Maybe it also should decrease in value in loose games? But then, how much, and why that much? Lots of questions yet to figure out. STRONG POINTS The strong points of this system is that it takes the situation in consideration, and that it is based almost entirely on calculated numbers based on probablility. So you have the relative values of different hands, based on probability, but then you also have to use your judgement on how to apply these points in different types of games and against different opponents. It's not a rigid system. Example: AK vs. T9s In a tight game AK in early position with noone in would be worth 16 points, and T9s would be worth about 9 points. But in late position in a loose game, with five callers, AK would be worth 15 points, and T9s would also be worth 15 points, so they are worth the same in that situation, and in the earlier situation AK was almost twice as good as T9s. Now, are these numbers really really correct? Well, I'd say that they are not, because there are factors that are not included in the system. But if they are important factors, they can most probably be included in some way, to make a better system. Also, the factors that ARE in the system, might not be correctly incorporated. But that should be possible to fix. USING IT So, how do you use a system like this? Well, a simple example would be to say that you should have 13 points to open the pot in early position, 11 points to open in middle position, and 9 points to open in late position. And when you are up against a single opponent, you should have a better points count than what you think that he would open with in his position. And so on. So you see that there are lots of room for judgement and adaptation, but still something firm to lean on for evaluating the strength of your hand in the situation at hand. You're even free to make assumptions on how many players will enter the pot, and the likelihood of it being raised, when you are in early position, and base your count on that. If you are an advanced and experienced player, of course you don't need all this, but I believe that something like this system would be much better for a beginner than the "hand groups" and tables that are widely used today. And I'd love to hear what y'all think about it (I think...), and how it can be improved. CALCULATING IT I won't go into much detail on how the system is calcualated, but it's nothing advanced, and should be quite easy to understand. The high card values are based on the probability that you will have top pair on the flop/turn/river. The draw values are based on the probability of flopping a draw, or a set. I have guesstimated that losing with your made high-card hand and hitting your draw "equals out," so that the probability of flopping them is a good enough measure of their value. So this could most probably be improved. The value of a set draw is multipied by the inverse of the probability of getting there with a flush draw from the flop. The fact that it is not as strong a hand is more than made up for by the fact that you also have an even stronger draw to a full house, and the fact that it is more hidden... The number of opponents and how that affects the value of the draws are also guesstimates, based on what seems to be widely agreed among experts. Four callers, it seems they agree on, give drawing hands full value. Maybe I have misunderstood this. Anyway, this is not based on any calculations. The fact that you get 2 points for a flush and a 3-way straight draw even with no opponents, is also based on hints from experts. A suited pocket seems to raise the hands value to the equivalent offsuit pocket with the lower card raised two steps (e.g. ATs roughly equals the value of AQ). This would mean roughly two points in this system. It's not based on any calculations of my own. OK, that's it, tell me your thought and comments and ideas. -URF | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, johnph77, 5. Sep 2003 03:06 | ||
| View ( Message | Thread ) | Return to Thread List | |
| Interesting concept. Very similar to the point count system used in Contract Bridge. Will await the results of future development. Thanks. | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, dan, 5. Sep 2003 03:51 | ||
| View ( Message | Thread ) | Return to Thread List | |
| You might want to check out an article in Card Player magazine by Michael Cappelletti, August 3rd issue??? I believe he talks about a point system strategy for Omaha Hi/Lo based on bridge's point counting. Don't quote me though...I am not positive all this info is correct. | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, URF, 5. Sep 2003 11:24 | ||
| View ( Message | Thread ) | Return to Thread List | |
| It seems to be correct. Thanks for the info. It makes sense too - Cappelletti is also a bridge player, is he not? Too bad he doesn't specify the system. Probably he want to sell it with his book. Anyone know if he's got a system for hold 'em too? -URF | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, Big Frank, 5. Sep 2003 05:41 | ||
| View ( Message | Thread ) | Return to Thread List | |
| I recall reading an article on-line on a starting a hand strategy using a 30 point system where points were awarded to various 2 card combinations. I did a search on Yahoo with the words "texas hold em point system" and I believe the article was the first in the list but I'm at work and our firewall won't allow me to get to it. Can someone check this out and let us know. Thanks | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, URF, 5. Sep 2003 11:50 | ||
| View ( Message | Thread ) | Return to Thread List | |
| I found a point system (searching on Google) by Edward Hutchison (http://www.homestead.com/erh2/HEM.html), but in my view he's got the relative points WAAAY wrong... The differences between the high card values are much greater than in his system. Scaled down to the same scale as mine, he would give A=9p, K=7.9p, Q=7.3p, J=6.7p, T=6.2p, 9=5.1p, down to 2=1.1p. So you can see that there is quite some difference between my system and his. But maybe his system has some strong points where mine is weak. I haven't really analyzed it. And he also doesn't give any clues to how he's come to the values he's giving. It does seem, though, that it's based on all-in equity, which mine isn't. -URF | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, URF, 5. Sep 2003 12:04 | ||
| View ( Message | Thread ) | Return to Thread List | |
| Also, his system isn't adaptive to the number of opponents, or the type of game, and he also gives rigid rules on how many points you should have, whereas I think that you should have more room for variation in this, based on your read of the game and players and situation. -URF | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, Phish, 5. Sep 2003 07:54 | ||
| View ( Message | Thread ) | Return to Thread List | |
| With all due respect, while your system may be very useful for programing a computer to play poker, I wonder how useful it really is for the beginning player. It is simply too complicated. Someone trying to adopt it will spend so much effort on the pre-flop play that he may well neglect his post-flop play, which is where the money is really won or lost. My advice/system for the beginner is this: find a small loose game (don't play in tight games as you probably won't beat it if you're a real novice). Pre-flop: play any pair, but dump it if you don't hit a set, or if one overcard appears and there is any action. Unsuited cards, play only if both are higher than a ten. Suited, play if both are higher than an eight. Continue only if you hit top pair or better, a flush draw or open-ended straight draw. If you have a made hand as opposed to a draw, play aggressively. Very simple and imperfect, but a sound basis for a real beginner, in my opinion. | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, noiseboy, 5. Sep 2003 08:59 | ||
| View ( Message | Thread ) | Return to Thread List | |
| That's pretty similar to what I tell a rank beginner before they hit the tables. I would add one qualification that the two-big offsuit card hands other than AK and AQ should be played from middle pos onwards when there is no early pos raiser. This gets them out of the trouble of playing something like AJ or AT from EP and getting raised by AK, flopping an A and paying off the whole way. Other than that, the any pair theory is certainly good for beginners in loose games. Actually, what you describe is pretty similar to a very simple version of what I'm getting at, although I'm focusing on them understanding why they play those hands. | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, URF, 5. Sep 2003 09:37 | ||
| View ( Message | Thread ) | Return to Thread List | |
| I don't think that it's really that complicated, especially not if one simplifies it by e.g. always using the "turn based" high card points. One can start out with a simple version, and make it more complicated as one progress, and according to ones likings. But you do point out a good point in that it is always a balance between simplicity and power. One other "simplification" one can make is when it comes to pocket pairs. These are really not that hard to learn a few rules of thumb for when to play and not. You really don't have to remember that JJ is 17+6 or something like that. You can have other simpler rules similar to the ones given in books today for the high and the low pairs, and maybe use the counting for middle pairs, or something like that. But even if you do not use the points counting in actual play, I still think that it could be a great tool for getting a grip on things when thinking about the game when you are not playing. To understand e.g. how much more powerful AA really is than JJ or TT. Or how much less powerful JJ is in a loose game than in a tight game, high-card wise. Any way, without using points for pocket pairs, basically all one has to do is to add two numbers, check how many opponents one has, and if one has a 0-, 1- or 2-gap, and add any draw values. I think that this can be pretty quickly and easily learned. Not any harder to learn than the tables and hand groups that are taught today. But of course, if you are going to put a total newbie at a table to play right now, a simpler system might have to be used. But I am thinking more in terms of the serious newbie that wants to study and learn the game. Trying to understand it, to get a grip on it. I certainly don't think it's too complicated for such a person. It's really not more complicated than the system used to teach newbies to play bridge. Or for that matter is used by most every bridge player, as far as I know, even though I'm no advanced bridge player. -URF | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, noiseboy, 5. Sep 2003 09:04 | ||
| View ( Message | Thread ) | Return to Thread List | |
| One thing I would say is that drawing hands gain value the more callers you have, so you need a way to adjust by making drawing hands worth very few points, but multiplying them by a coefficient for the number of callers??? There would also be a resiliency factor that you would have to consider. For example, small pairs can stand raises more than suited connectors, even though both like lots of callers. The only problem with your system is that the closer it gets to being perfect, the more complicated it will become. | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, URF, 5. Sep 2003 10:16 | ||
| View ( Message | Thread ) | Return to Thread List | |
| "One thing I would say is that drawing hands gain value the more callers you have, so you need a way to adjust by making drawing hands worth very few points, but multiplying them by a coefficient for the number of callers???" Well, that is done, in a simplified way. One gets e.g. 2 points for a suited pocket even with no callers, and 7 points with four or more callers, and inbetween points for inbetween number of callers. Of course one might change it to extend to even more points with even more callers, if that would be better. I'm open to suggestions. Or am I misunderstanding what you are saying? "There would also be a resiliency factor that you would have to consider. For example, small pairs can stand raises more than suited connectors, even though both like lots of callers." Once again, I do think that I have such factors in the system, even though they might need tuning and/or additions. Small pocket pairs stand raises much better than small suited connectors, because they have higher drawing value points. If the pot is raised, drawing values are cut in half, which for 55 with three callers means that it goes down from 12 to 9 (because if a fourth player raises, you then have four "callers"), but 65s goes down from 9 to 6.5. Of course, these exact values does not seem quite right to me - I guess the set draw value dimension might need a bit more points, or maybe it would be enough to extend the increase of drawing values to more than four callers, so that e.g. 55 would get about 11 points in a raised pot with five "callers". Hm. Still seems a bit low though, doesn't it? As I said, there's room for improvement. "The only problem with your system is that the closer it gets to being perfect, the more complicated it will become." Yes, that's true - one has to balance power against simplicity of use. But, well, that's always the case in everything, right? It's certainly true about the systems used in bridge, but these are still used, and successfully so, but advanced players do think above and beyond them. And also, you don't really need perfection, all you need is good enough to separate the playable from the unplayable, somewhere among the borderline cases. I think that this should be possible in a reasonable way. But any system such as this should never be sold as a perfect truth. But it need to be close enough to the truth. -URF | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, vanagon40, 5. Sep 2003 11:46 | ||
| View ( Message | Thread ) | Return to Thread List | |
| Ed Hutchison has a similar system http://erh2.homestead.com/hem.html | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, URF, 5. Sep 2003 11:54 | ||
| View ( Message | Thread ) | Return to Thread List | |
| What a coincident - I just posted on it above in reply to Big Frank. -URF | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, Big Frank, 5. Sep 2003 11:55 | ||
| View ( Message | Thread ) | Return to Thread List | |
| This is the system I mentioned above (but couldn't access the site due to firewall restrictions). I had read it months ago and thought it had merrit. I was in the process of comparing the number system to other starting hand rankings and got sidtracked and never got back to it. | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, URF, 5. Sep 2003 12:10 | ||
| View ( Message | Thread ) | Return to Thread List | |
| If you do take the time to make a comparison, it would be interesting to know how you think that my system in its current form holds up against his, and the other rankings. Of course I haven't given that much guidelines as to what number of points to have to play in different positions and situations, but one can compare the point counts of those hands getting on this and that side of the dividing lines of other system, and see if there is a similar dividing line in the points count, and so on, and thereby if it all seems to make reasonable sense. -URF | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, gary ford, 5. Sep 2003 21:00 | ||
| View ( Message | Thread ) | Return to Thread List | |
| As the debate rages on, it appears that most of you are from the new generation of " math guys" who have taken up poker and are trying to quantify it. It would be intriguing to put 5 of you at a table with 5 of the " rounders " types, who practiced deception and the ability to read their opponents. I have no idea how it would turn out. But for those of you asking a bout point count systems, an article called " a new guide for starting hands in hold-em poker '. ypou should be able to google it or try poker pages.com. Most of the questions that have been raised here are addressed in that article. Fuel the boom-- Gary Ford | ||
| Return to Thread List | ||
| Re: A Starting Hand Points-Count System for Hold 'Em, URF, 5. Sep 2003 22:25 | ||
| View ( Message | Thread ) | Return to Thread List | |
| I recognize the title of that paper, and if it's the one I think it is, it certainly is no points count system, it's a starting hands list ordered into groups, mimicing Sklansky's hand groups, but disagreeing with it. And as far as I know, this paper is crap based mostly on all-in equity. If I remember correctly. As to the "you math guys versus us real poker players" question, I find this silly every time it comes up. It's like some people think that poker works by some kind of magic or something. It of course does not. It can be quantified. It's just that there is so much hidden information that need to be revealed, and the process of revealing this information involves using knowledge of human psychology and all kinds of such hard-to-touch stuff, and this is so complex and ever changing that it might seem like something not following the laws of nature, but of course it does. But since it is so complex and ever changing, it's nothing that you can give a finite set of specific instructions to someone regarding in a book, and then he or she knows everything about this subject forever. Of course not. But guidelines can be given for how to train yourself to be better in this subject. So what you need to do is to separate the "information extraction" part of poker from the rest of poker knowledge. If you do that, there is nothing to stop you from quantifying poker once and for all. Of course, "information extraction" is a very large part of what makes a good poker player good. So a "math guy" can be more knowledgeable in poker theory than a "rounder" (as you put it), and still lose, because he sucks in the area of "information extraction". There's also a big difference in knowing the theory of something, and being able to practically apply that theory. So the "math guy" might even be worse in theory application also. Even if he knows much more indepth theory outside the table. Now, should you scorn theory just because the theoreticians might not play as good as non-theoreticians? Well, not in my book. One just have to recognize that to be a good player, one has to be able to apply the theory, and to extract information to apply it on. That's my view. -URF | ||
| Return to Thread List | ||
| POKER FORUM HOME | POKER FORUM | LINK TO US | ARCHIVE | ONLINE POKER | Copyright 2002, United Poker Forum |
|
Getting Started |
UPF Tournaments |
Poker News, Views, Rules |
Poker Strategy & Psychology |
Money and Bankroll Poker Bonuses & Promotions | World Series of Poker (WSOP) | Play Online Poker | Poker Odds & Statistics | Tournament Poker | Poker Books, Videos & Learning Tools Looking for a Poker Game | Poker Bad Beats | Not Quite Poker | Quizzes and Polls | Forum Suggestions & Bugs |
|
|
|
|
Interesting Links: Online Poker | Free Poker Games | United Poker Network |
|