vp simulator

[asteroid]

Hi Phil, sorry to hear you were severely criticized for your alternative strategy. There is no need for me to move the discussion elsewhere because I have little else to say (which I will say next). The strategy I have is mathematically valid and is based on the binomial probability density function. Using the two-pair hand in full pay bonus poker as the example, after 5 hands are played, an inflection point is observed, where the probability of observing at least one 2-pair (one or more "successes" in the binomial pdf) becomes greater than not observing any at all in the sequence of  5 hands played. By the time you get to 47 hands played it becomes extremely unlikely that you will not have observed any two-pair hands (which pay 2 for 1 in full pay bonus poker) even taking into account the [np(1-p)] variance (n=#  of hands played, p=.1299 or 12.99%) assuming optimal discarding/holds. So, if the player observes no two-pairs in 15 hands, they can switch to a flat bet that is a bit larger (how much larger can be optimized via a custom simulation and how many hands can be optimized this way as well) since it becomes more and more likely that a cluster/sequence of hands thereafter (where the cluster under consideration also includes all hands previously played in the sequence) will contain a two-pair hand (or more than one). Once a two-pair is observed after the higher flat bet, the low flat bet level is again adopted by the player. That is all I have to say on this subject. So no need to move the topic. If someone wants to refute the binomial pdf, which has been used to build new casinos and keep the lights on in them, they can go right ahead (I realize each hand is independent from the next, but I am thinking in terms of clusters/aggregates, not individual hands).Cheers, asteroid.------------------------------------------------------------------




# of
hands played
probability of
at least one 2-pair
probability of
no 2-pair


1
0.130
0.870


2
0.243
0.757


3
0.341
0.659


4
0.427
0.573


5
0.501
0.499


6
0.566
0.434


7
0.622
0.378


8
0.671
0.329


9
0.714
0.286


10
0.751
0.249


11
0.784
0.216


12
0.812
0.188


13
0.836
0.164


14
0.857
0.143


15
0.876
0.124


16
0.892
0.108


17
0.906
0.094


18
0.918
0.082


19
0.929
0.071


20
0.938
0.062


21
0.946
0.054


22
0.953
0.047


23
0.959
0.041


24
0.965
0.035


25
0.969
0.031


26
0.973
0.027


27
0.977
0.023


28
0.980
0.020


29
0.982
0.018


30
0.985
0.015


31
0.987
0.013


32
0.988
0.012


33
0.990
0.010


34
0.991
0.009


35
0.992
0.008


36
0.993
0.007


37
0.994
0.006


38
0.995
0.005


39
0.996
0.004


40
0.996
0.004


41
0.997
0.003


42
0.997
0.003


43
0.997
0.003


44
0.998
0.002


45
0.998
0.002


46
0.998
0.002


47
0.999
0.001











































































































































































































































--------------------------------------------------------------------



I developed a similar strategy that works well for me.   Over a two year period, I posted over $18,000 worth of jackpots on this website playing 90% single coin VP.   I was severely criticized because it went against traditional advantage play strategy promoted by professional video poker players.  After a long and nasty battle, I agreed to move the discussion of alternative strategies to the Recreational Forum.   Many forum members are of the opinion that there is only one way to win at VP, play positive games perfectly and wait for math to work in your favor.   Unfortunately, most VP games are not positive and these opportunities are quickly disappearing.  There is a need to discuss alternative strategies when playing negative VP games.   These strategies do not conflict in any way with traditional strategies that have been well proven on positive games.  If you would like to discuss any alternative strategy, please start a thread on the Recreational Forum and I will be happy to respond.

[FloridaPhil]

Traditional thinking is that each hand is totally independent of the other and streaks and repeat patterns are irrelevant in VP.   The fact is small streaks do in fact happen especially in wild card games.  Many people over the years have claimed to be able to capitalize on this phenomenon.  I believe it is possible to use this hypothesis to cut the house edge when playing a negative VP game.  I do not claim my strategy is guaranteed to make you a long term winner, but I do claim it will improve your results over flat betting an overall negative VP game. Over a two year period which was fully documented on this website I played using my strategy and reported the results.  I played a minimum of 6 hours a week plus numerous road trips and vacations where I played for a week at a time. I posted a picture of every win of $1,000 or more.  The vast majority of the time I played single coin quarters.  After two years I posted a long term profit of a little under $2,000.   I did not count comps in my winnings or any outside gambling wins.  My play was mostly single coin traditional deuces wild with 98% odds or less.  Some of my wins were $20 pot shots taken on $1 and $5 machines.  In spite of these results I was severely attacked on this forum by someone who makes money consulting for casinos and selling books and software to allegedly beat these same casinos playing VP.  I never argued against his traditional strategy, I only stated my strategy is better when playing a negative game.  This didn't matter as any alternative to his strategy was immediately met with insults and personal attacks.  He even went so far as to accuse me of taking pictures of other players jackpots and posting them as my wins.  This ended when a poll was taken on this forum.  The choices were to limit discussion to the liking of this person or allow open discussions.  Open discussions won and I agreed to limit my alternative strategy discussions to the Recreational Forum. I am only bringing it up here because your proposed strategy seems similar to mine and I don't want you to suffer the same abuse.

[onemoretry]

(I realize each hand is independent from the next,
Am I the only person who thinks it's inconsistent to say this, and, yet, design a betting strategy that seems to be based on recent results?

[FloridaPhil]

I believe each hand is independent.  I also believe streaks happen.  Anyone who plays deuces wild knows this to be a fact.  The question is whether you believe it is possible to capitalize on them?

[asteroid]

It would be very unusual to observe no two-pair hands in 47 hands of optimal hold full pay bonus poker. If a person increases their bet slightly (say from 10 cents to 12 cents on a 1 cent up to 100 credit vp machine) after observing no two-pair hands after 15 hands (15/47ths of the way through the sequence), they are more likely to hit a two-pair hand with a higher bet than not at some point during the remainder of the 47 hand sequence. I guess no matter how heavily I emphasize a focus on an aggregate number of hands (sequence) rather than a single hand, which of course is (in theory) independent of the previous result, the focus will always remain on a single hand. Basically I am betting on a convergence to 13% over the long-haul. The central limit theorem.If an infinite number of hands are played the proportion will converge to 13% two-pair hands. Thought experiment: consider an urn with 13000 blue balls (two-pair outcome) and 87000 white balls. Mix the contents of the urn thoroughly. Draw a 47 ball sample from this urn and tabulate the percentage of blue balls in the sample. Put the 47 balls back in the urn and mix thoroughly. Repeat this 1000 times so that you have 1000 tabulations. Average the tabulations. The average will be close to 13%. This is analogous to a 47 hand vp sequence where the player keeps repeating the experiment of raising his or her bet slightly and flat-betting at this slightly raised value if no two-pair hands are observed through14 hands. The bet is then decreased back to its base value when the two-pair hits. Repeat over and over. With the 13% convergence over all these 47-hand sequences (47-ball sample draws from the urn) where the betting algorithm is adhered to, the player will eventually converge on higher betting sizes during two-pair clusters over an aggregate number of 47-hand sequences (say 1000 as in the urn example).sorry I won't post anything else on this any more. [QUOTE=asteroid]
(I realize each hand is independent from the next,
Am I the only person who thinks it's inconsistent to say this, and, yet, design a betting strategy that seems to be based on recent results? [/QUOTE]

[onemoretry]

I believe each hand is independent.  I also believe streaks happen.  Anyone who plays deuces wild knows this to be a fact.  The question is whether you believe it is possible to capitalize on them?In order to capitalize on them, you would have to have the ability to predict when, and for how long, they will occur. Does anyone really have that ability?

[onemoretry]

It would be very unusual to observe no two-pair hands in 47 hands of optimal hold full pay bonus poker.
That's quite true (the chance of it is about 0.14%). But, if you have been unlucky enough to have experienced 46 consecutive non two pair results, the likelihood of 47 consecutive ones is 87%.

I gather you do not wish to address the issue of inconsistency in approach, i.e., varying bet size on the basis of prior results in a supposedly random situation?

[asteroid]

This statement in red is false. the likelihood is: 0.129 to the 47th power (with optimal discarding). This number is somewhat smaller than .87. As for addressing the issue you refer to, please see the urn thought experiment or just keep thinking in terms of single hands and disregard it. I broke my promise of not posting on this subject. I won't do so again.[/QUOTE]the likelihood of 47 consecutive ones is 87%.

I gather you do not wish to address the issue of inconsistency in approach, i.e., varying bet size on the basis of prior results in a supposedly random situation?[/QUOTE]

[FloridaPhil]

[quote=onemoretry]In order to capitalize on them, you would have to have the ability to
predict when, and for how long, they will occur. Does anyone really
have that ability?[/quote]Reducing the house take on a negative video poker game does not involve making any predictions. If you seriously want to discuss how this is done, post a tread on the Recreational Forum and I will be happy to discuss this openly and in detail.  If all you want is an endless argument, we have better things to do.

[onemoretry]

This statement in red is false. the likelihood is: 0.129 to the 47th power (with optimal discarding). This number is somewhat smaller than .87. As for addressing the issue you refer to, please see the urn thought experiment or just keep thinking in terms of single hands and disregard it. I broke my promise of not posting on this subject. I won't do so again.the likelihood of 47 consecutive ones is 87%.

You chose to not include the qualifier to that statement, i.e., the likelihood of 47 consecutive non two pair results is 87% if the previous 46 have also been non two pairs.

Per your earlier post, the likelihood of 47 consecutive non two pair hands is not .129 to the 47th power, it is .871 to the 47th power. But that's irrelevant. When you hit the deal button, the machine does not really care if you have had 46 consecutive non two pair hands or if you've hit several of them in a row - it just randomly spits out the cards.