|
|
#1
06-29-2005, 01:57 AM
|
|||
|
|||
King Yao\'s DIPO overcomplication
Yao's DIPO complicates a simple multiplication trick developed by Andy Morton and published several years ago by Abdul Jalib. To Yao's credit, he does state that Jalib created a "similar method" to DIPO.
Basically, for DIPO, Yao says to come up with an estimated pot size ratio (ie implied odds) to the bet you are facing, and multiply your outs to come up with the "good number." Then compare the good number to the bad number - ie the number of non outs, which is unseen cards - outs. If the good number is higher, you can call. Example: 8 outs on the river, pot is laying you 5:1, but you think you will make 1 more bet if you hit, so the estimated pot size ratio is 6:1. 8 x 6 = 48 - the Good Number. The bad number is 46-8, or 38 non outs. Since 48 > 38, you can call. The Morton/Jalib method is simpler: Figure out how many unseen cards (either 47 on flop or 46 on turn). Multiply the number of effective outs by the pot odds ratio +1 (+2 for implied odds), and compare it to the number of unseen cards. Example: 8 outs on the turn, pot is laying 5:1, but you think you will win one more bet if you hit. 8 x (5+1+1) = 56 > 46 unseen cards, so you can call. Bottom line, you don't need to figure out the bad number Jalib's way. You simply add 1 to the actual pot odds ratio (plus another 1 for implied odds) before multiplying by the outs, and it does the same thing as DIPO. 
|
|
#4
06-29-2005, 10:44 AM
|
|||
|
|||
|
Re: King Yao\'s DIPO overcomplication
I think you have a point. For some people (including you), Abdul's method is easier. But for others, memorization is easier (as chson posted).
So there is a scale of: simplication - complication But the other dimension of the scale is: understanding - memorization What chson posted was clearly memorization. He, along with others who have noted this to me, has memorized a table regarding how big the pot odds is needed given the number of outs they have. I don't doubt that some people can memorize these charts and have a full understanding of what they mean - however I also believe its easy to fall into a trap of memorizing these charts and lose the understanding of "why". Back to Abdul's method - adding one to the expected pot size, and then mulitplying it by the number of outs - is correct, but I don't think its intuitive. I think its easy for the practitioner to memorize it, but forget exactly why it should be done that way. Many people can move the algebra equation around to figure out DIPO and Abdul's method are the same, but many cannot. I think Abdul's method is "simpler" on the simplication - complication scale, but also it requires more "memorization" and less "understanding" on the memorization - understanding scale. Every method has its own place on the scale. Its a matter of how the user thinks and what he is comfortable with that decides which method he should use and where he should be on that scale. As I have posted before, I need to write an Appendix on this for my next edition. Here are some pot odds calculation methods and where I think they are on the scale. 10 being most extreme to the right, and 1 being most extreme to the left. A 10 on the simplication - complication scale is very complicated. A 1 on the memorization - understanding scale is only memorization. (of course people can still disagree with me on these scale numbers ) EV: S-C = 10 / M-U = 10 DIPO: S-C = 6 / M-U = 8 Abdul: S-C = 4 / M-U = 5 Tables: S-C = 1 / M-U = 2 To me, there is an obvious correlation between the S-C and the M-U scales. In order to simplify, you'll have to memorize. This means giving up some understanding (note - I don't mean that the person that memorizes necessarily doesn't understand - its that it is easy for the memorizer to not understand - or to forget the reasons "why" - because the memorizer doesn't need to understand). I also think a third dimension is useful to think about - and that's how to adjust when circumstances are different. This includes thinking about pot odds situations on the Flop (rather than the Turn), and what happens when there is 2 bets to you (or a chance of 2 bets if you are not closing the action). The correct application of DIPO does a good job in both instances (multiply the Bad Number by the number of total bets: 2 if there is a bet and a raise; 1.5 if there is a bet, and you think there's a 50% chance of a raise behind you). Its not easy in the beginning, but once one gets used to it, its not difficult. I think the flow of understanding is good in using DIPO - from the normal Turn situation of only 1 bet to when there are more than 1 bet. Using Abdul's method, the equation would become: [Outs x (EPS +2 )] / 2 (I think I converted the algebra correctly, correct me if I'm wrong). To me, that is not only exponentially complicated but also is now 100% memorization, because one has to memorize this formula. There may be another way to simplify it to make it less complicated, but it would still include 100% memorization of the formula, which again, can lead to non-understanding. The tables that some people memorize for pot odds may work too, but it gets more complicated. Say someone has bet in front of you - you think you have 5 clean outs, and the expected pot size is 8. If you close the action, your memorized table tells you that you can call. But you think there is a 50% chance that someone in back of you will raise if you call. If you are comparing a Pot Size of 8 to 5 outs as a clear call, then how do you adjust for the possibility of a raise behind you? Its not as easy now. Let's backtrack and say there is a bet and a raise in front of you. Then you can say the Pot Size to Bet Size is 4 (Expected Pot Size of 8, but you gotta put in 2 bets, so 8 / 2 = 4). Now you can compare 4 to 5, and its not a call. But if there is a 50% of a raise behind you, then half the time the Pot Odds is 8, the other half the Pot Odds is 4....for Pot odds of 6. Maybe there's an easier way to do it, but since I don't use the memorization tables, I can't come up with an easier way right now. Even using these tables, you need to do some mental gymnastics at the table to figure things out - all that with the possibility of not truly understanding what you are doing. I think this is a good start for my appendix. I will need to clean it up and probably make corrections that others will point out to me shortly [img]/images/graemlins/smile.gif[/img]
|
|
#5
06-29-2005, 11:12 AM
|
|||
|
|||
|
Re: King Yao\'s DIPO overcomplication
[ QUOTE ]
Yao's DIPO complicates a simple multiplication trick developed by Andy Morton and published several years ago by Abdul Jalib. The Morton/Jalib method is simpler [/ QUOTE ] I disagree. The Morton/Abdul 'method' and the Yao 'method' are one and the same. They are both simply a reworking of the EV calculation. The reason that the calculation for the Morton/Abdul approach APPEARS easier, is that it is assumed that the bet size is equal to 1. In fact in the document "Sucking Out" by Abdul he doesn't consider the case where you have to call more that one bet. (Presumably because he wanted to keep things simple.) To use Abdul's approach when facing more than one bet, you have to change the given formula. When facing a single bet, you calculate Outs x (EPS + 1) and compare it to 46 (on the turn) or 47 (on the flop). However, when facing (say) two bets, you have to calculate Outs x (EPS + 2) and compare it to 47 x 2. If you compare the working out required for either approach, there's hardly any difference. Here's an example. Let's assume the following (using an example from Yao's book on p.89): Outs = 4 EPS = 11.5 big bets on the turn Bet = 1 Yao's method: 4 x 11.5 = 46 46 - 4 = 42 46 > 42, therefore CALL. Morton/Abdul method: 11.5 + 1 = 12.5 4 x 12.5 = 50 50 > 47, therefore CALL. They both require one multiplication, one addition/subtraction and one comparison. Not much difference here. Let's now do this again assuming you're now facing two bets: Outs = 4 EPS = 11.5 Bet = 2 Yao's approach: 4 * 11.5 = 46 46 - 4 = 42 2 * 42 = 84 46 < 84, therefore FOLD. Morton/Abdul's approach: 11.5 + 2 = 13.5 4 x 13.5 = 54 2 x 46 = 92 54 < 92, therefore FOLD. Again, each approach requires two multiplications, one addition/subtraction and one comparison. I don't think you can argue that one approach is less work than the other, IMO. (Unless you find subtraction significantly more difficult than addition).
|
|
#6
06-29-2005, 12:15 PM
|
|||
|
|||
|
Re: King Yao\'s DIPO overcomplication
[ QUOTE ]
[ QUOTE ] Yao's DIPO complicates a simple multiplication trick developed by Andy Morton and published several years ago by Abdul Jalib. The Morton/Jalib method is simpler [/ QUOTE ] I disagree. The Morton/Abdul 'method' and the Yao 'method' are one and the same. They are both simply a reworking of the EV calculation. The reason that the calculation for the Morton/Abdul approach APPEARS easier, is that it is assumed that the bet size is equal to 1. In fact in the document "Sucking Out" by Abdul he doesn't consider the case where you have to call more that one bet. (Presumably because he wanted to keep things simple.) To use Abdul's approach when facing more than one bet, you have to change the given formula. When facing a single bet, you calculate Outs x (EPS + 1) and compare it to 46 (on the turn) or 47 (on the flop). However, when facing (say) two bets, you have to calculate Outs x (EPS + 2) and compare it to 47 x 2. If you compare the working out required for either approach, there's hardly any difference. Here's an example. Let's assume the following (using an example from Yao's book on p.89): Outs = 4 EPS = 11.5 big bets on the turn Bet = 1 Yao's method: 4 x 11.5 = 46 46 - 4 = 42 46 > 42, therefore CALL. Morton/Abdul method: 11.5 + 1 = 12.5 4 x 12.5 = 50 50 > 47, therefore CALL. They both require one multiplication, one addition/subtraction and one comparison. Not much difference here. Let's now do this again assuming you're now facing two bets: Outs = 4 EPS = 11.5 Bet = 2 Yao's approach: 4 * 11.5 = 46 46 - 4 = 42 2 * 42 = 84 46 < 84, therefore FOLD. Morton/Abdul's approach: 11.5 + 2 = 13.5 4 x 13.5 = 54 2 x 46 = 92 54 < 92, therefore FOLD. Again, each approach requires two multiplications, one addition/subtraction and one comparison. I don't think you can argue that one approach is less work than the other, IMO. (Unless you find subtraction significantly more difficult than addition). [/ QUOTE ] You miss the point. In the Morton/Abdul method, you are always comparing to unseen cards, a fixed number. In the Yao way, you are always comparing to a variable number, non-outs. By the way, when facing 2 bets, you don't multiply unseen cards by two in the Morton method. You simply reduce pot odds by half before adding 1 and multiplying by outs.
|
|
#7
06-29-2005, 12:35 PM
|
|||
|
|||
|
Re: King Yao\'s DIPO overcomplication
[ QUOTE ]
Back to Abdul's method - adding one to the expected pot size, and then mulitplying it by the number of outs - is correct, but I don't think its intuitive. I think its easy for the practitioner to memorize it, but forget exactly why it should be done that way. Many people can move the algebra equation around to figure out DIPO and Abdul's method are the same, but many cannot. I think Abdul's method is "simpler" on the simplication - complication scale, but also it requires more "memorization" and less "understanding" on the memorization - understanding scale. [/ QUOTE ] Good point. I'll admit that adding 1 to the pot odds ratio is not intuitive until you realize that you cannot compare odds and chances. Pot odds to card odds = pot chances to card chances ( i.e. outs/total unseen cards). The addition of 1 to the pot odds converts it into a pot chance, which you multiply by outs and compare to unseen cards (because your card chances are outs divided by unseen cards). The beauty of it is that the number of unseen cards is a constant number depending on the stage of the hand (flop is always 47, turn is always 46), whereas the number of non outs (ie bad number) varies from hand to hand. This is the root of my original point that the Morton/Abdul method is simpler. [ QUOTE ] Using Abdul's method, the equation would become: [Outs x (EPS +2 )] / 2 (I think I converted the algebra correctly, correct me if I'm wrong). To me, that is not only exponentially complicated but also is now 100% memorization, because one has to memorize this formula. There may be another way to simplify it to make it less complicated, but it would still include 100% memorization of the formula, which again, can lead to non-understanding. [/ QUOTE ] Not if you understand you are comparing pot chances to card chances. Figure out (actual or implied) pot odds (expressed as a ratio to 1) no matter what the bet is to you, add 1 to make it a pot chance, multiply by effective outs, and compare to unseen cards. [ QUOTE ] I also think a third dimension is useful to think about - and that's how to adjust when circumstances are different. This includes thinking about pot odds situations on the Flop (rather than the Turn), and what happens when there is 2 bets to you (or a chance of 2 bets if you are not closing the action). The correct application of DIPO does a good job in both instances (multiply the Bad Number by the number of total bets: 2 if there is a bet and a raise; 1.5 if there is a bet, and you think there's a 50% chance of a raise behind you). [/ QUOTE ] If there is a 50% chance of a raise behind, and the pot is 8 bets and 1 bet to call, then proper estimated actual (not implied) pot odds are (assuming the original better will call 1 extra bet and that the player behind will fold if he does not raise) .50 (8:1) .50 (11:2) = 6.75:1 Multiplying your bad number (ie non outs) by 1.5 if there is a 50% chance of a raise behind is incorrect. You will fold too often. If there is a chance of a raise behind, I simply assume that chance cancels out my implied odds, and go on actual pot odds for the sake of convenience.
|
|
#8
06-29-2005, 12:57 PM
|
|||
|
|||
|
Re: King Yao\'s DIPO overcomplication
[ QUOTE ]
If there is a 50% chance of a raise behind, and the pot is 8 bets and 1 bet to call, then proper estimated actual (not implied) pot odds are (assuming the original better will call 1 extra bet and that the player behind will fold if he does not raise) .50 (8:1) .50 (11:2) = 6.75:1 Multiplying your bad number (ie non outs) by 1.5 if there is a 50% chance of a raise behind is incorrect. You will fold too often. [/ QUOTE ] You are right that multiplying the Bad number by 1.5 gives the technically incorrect answer. Thanks for pointing that out. This will sound like a cop-out but...: when you do get raised behind, your outs are likely to be lower than you originally estimated. So although the formula of multiplying the Bad Number by 1.5 (when there is a 50% chance you get raised behind) is not techinically correct, it is practically useful. When in a situation where the DIPO answer gives a technically incorrect answer, its likely the original assumptions were correct to begin with: your outs are likely to be lower. For example, if you have a straight or flush draw, it doesn't matter if the 1.5 gives you an incorrect numerical answer, DIPO using the 1.5 still gives you a correct comparative answer, you still are on the correct side of calling (exception: when the pot is tiny). But when you do not have a draw to the nuts, a raise behind you likely means you have fewer outs than you previously expected. So it is practically useful to be more conservative. So technically I agree with you. But I think it is practical. Of course, a devil's advocate will argue that this is now lowering "understanding" which is against the whole idea of DIPO to begin with. I still think the bottomline is that it is a give and take between exactly what the user is looking for in these different dimensions. People have different preferences.
|
|
#9
06-29-2005, 01:23 PM
|
|||
|
|||
|
Re: King Yao\'s DIPO overcomplication
[ QUOTE ]
You miss the point. In the Morton/Abdul method, you are always comparing to unseen cards, a fixed number. In the Yao way, you are always comparing to a variable number, non-outs. [/ QUOTE ] No I didn't. I was assuming that the arithmetic required was sufficiently trivial not to matter. You have a point though, that this may not be true for some people. [ QUOTE ] By the way, when facing 2 bets, you don't multiply unseen cards by two in the Morton method. You simply reduce pot odds by half before adding 1 and multiplying by outs. [/ QUOTE ] You can use either method. 'Reducing pot odds by half' introduces a division, rather than a multiplication. Most people find multiplication significantly easier than division, particularly if the bet size was larger, say three bets. In the example given, most people would find it rather difficult to divide 11.5 by 3.
|
|
#10
06-29-2005, 01:47 PM
|
|||
|
|||
|
Re: King Yao\'s DIPO overcomplication
[ QUOTE ]
[ QUOTE ] You miss the point. In the Morton/Abdul method, you are always comparing to unseen cards, a fixed number. In the Yao way, you are always comparing to a variable number, non-outs. [/ QUOTE ] No I didn't. I was assuming that the arithmetic required was sufficiently trivial not to matter. You have a point though, that this may not be true for some people. [ QUOTE ] By the way, when facing 2 bets, you don't multiply unseen cards by two in the Morton method. You simply reduce pot odds by half before adding 1 and multiplying by outs. [/ QUOTE ] You can use either method. 'Reducing pot odds by half' introduces a division, rather than a multiplication. Most people find multiplication significantly easier than division, particularly if the bet size was larger, say three bets. In the example given, most people would find it rather difficult to divide 11.5 by 3. [/ QUOTE ] I always convert pot odds to an expression of x to 1. Most players do. If some don't, they should. In fact, like most players, I always halve pot odds as the bet doubles, so halving pot odds facing a raise is natural. As for the math of DIPO v. Abdul/Morton, they are the same. DIPO uses odds. Abdul/Morton uses chances.
|
 |
|
|
Linear Mode
