RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
January 15th, 2015 at 4:40:03 AM permalink
How would you go about learning the strategy for a different game, like Ultimate X or Quick Quads? In the past, I'd practice on Video Poker for Winners...but there doesn't seem to be to practice either game, at least not that I could find. Nor could I figure out a way to insert my own strategy chart.

An idea I had (that I just figured out wouldn't work, I don't think?) is to modify the paytable on Ultimate X so it takes into account the return of the hand as well as the multiplier. Then modify the paytable in VPW so that it generates a strategy that is the same as UX strategy, and practice that way. Unfortunately, I don't know what the return of the paytable would be if you were just using UX strategy but not taking into account multipliers for future hands.

My idea is/was something like this. Take 8/6 BPD for example with 99.7% return. If you are playing regular 8/6 BPD with the regular strategy, the return is 98.5%. Thus, you could change the paytable to be something like:

Jacks or Better: 1 + 2*0.985
Two pair: 1 + 3*0.985
three of a kind: 3 + 4*0.985
straight: 4 + 7*0.985
etc...

But of course, you aren't getting the 98.5% return from the actual payouts, because you wouldn't be using a strategy to maximize those payouts, you're using a strategy to maximize both the multipliers and payouts.

Would my way of generating a paytable to use in VPW work if I could get the proper return % on payouts using UX strategy? In other words, let's say I played regular 8/6 BPD (not UX) but I used UX strategy -- what would my return be? JB?


[spoiler]
As for QQ, I suppose (if possible, haven't checked [don't have VPW on this computer]), if you could create your own paytable with your own desired hands, you could create a (complicated) table with every QQ hand possible, like TTT9A, TTT82, TTT73, TTT64, TTT55, 9998A, 99972, 99963, 99954, etc.....

Or am I just crazy?
[/spoiler



BETTER YET -- is there software (program or online) where I could practice these strategies? And it'd warn me of errors?
Hullabaloo
Hullabaloo
  • Threads: 17
  • Posts: 452
Joined: Nov 30, 2014
January 15th, 2015 at 5:17:50 AM permalink
Quote: RS


BETTER YET -- is there software (program or online) where I could practice these strategies? And it'd warn me of errors?



Have you tried videopoker.com?

Gold members can play some games in training mode. UTX doesn't appear to have it, but QQ does.
JB
Administrator
JB
  • Threads: 334
  • Posts: 2089
Joined: Oct 14, 2009
January 15th, 2015 at 6:49:38 AM permalink
Quote: RS

My idea is/was something like this. Take 8/6 BPD for example with 99.7% return. If you are playing regular 8/6 BPD with the regular strategy, the return is 98.5%. Thus, you could change the paytable to be something like:

Jacks or Better: 1 + 2*0.985
Two pair: 1 + 3*0.985
three of a kind: 3 + 4*0.985
straight: 4 + 7*0.985
etc...


That approach is similar to how I calculated the "optimal single strategy" for Ultimate X games: starting with the base return, then repeatedly adjusting the paytable as you did above, using the base return in the first iteration and the most recently-calculated return in subsequent iterations, until the return and strategy in the last two iterations are identical. Even losing hands have a non-zero payoff, since they still have an implied 1X multiplier on the following round. I don't know of any VP training programs that let you adjust the payout for a losing hand.
tongni
tongni
  • Threads: 1
  • Posts: 203
Joined: Feb 27, 2013
January 15th, 2015 at 7:03:43 AM permalink
Quick quads and ultimate X are probably the only two VP variants that any pro really should know. The strategies are online, just memorize the chart. The strategy for deuces ultimate X is pretty easy to generate as well, plays similar to NSU deuces with holding 4 to a flush with wild and two gap two wild SF draws.
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
January 15th, 2015 at 12:57:22 PM permalink
Quote: JB

That approach is similar to how I calculated the "optimal single strategy" for Ultimate X games: starting with the base return, then repeatedly adjusting the paytable as you did above, using the base return in the first iteration and the most recently-calculated return in subsequent iterations, until the return and strategy in the last two iterations are identical. Even losing hands have a non-zero payoff, since they still have an implied 1X multiplier on the following round. I don't know of any VP training programs that let you adjust the payout for a losing hand.



If losing hands have a 1X multiplier on the following rounds, wouldn't that be the same as everything having a multiplier of 1X less? Like high pair just 1X, 2pair 2x, 3oak 3x, str is 6x etc?

After all, the paytable is just a bunch of ratios. Doesn't matter if a royal pays 800, or 800 million, as long as a high pair pays 1, or 1 million respectively.

How many iterations did it take you, JB? 5? 10? 100? Couple thousand?
JB
Administrator
JB
  • Threads: 334
  • Posts: 2089
Joined: Oct 14, 2009
January 15th, 2015 at 3:21:57 PM permalink
Quote: RS

If losing hands have a 1X multiplier on the following rounds, wouldn't that be the same as everything having a multiplier of 1X less? Like high pair just 1X, 2pair 2x, 3oak 3x, str is 6x etc?


The multipliers are actually half of what they appear to be, since you have to bet 10 coins, but the prizes are still based on 5 coins. So if the 5-coin return of the paytable is 0.99 then the adjusted payoff for each hand is (its 5-coin prize + (its next-hand multiplier * 0.99)) / 2; in this example, a losing hand would be treated as if it pays 0.495 for 1. But of course, after adjusting the paytable and recalculating the strategy/return, the 0.99 figure changes; this was repeated until no more changes were necessary.

Quote: RS

How many iterations did it take you, JB? 5? 10? 100? Couple thousand?


I think all of the games available for UX only needed somewhere between 8 to 15 iterations before they stopped changing.
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
January 15th, 2015 at 6:34:26 PM permalink
Awesome. And I can use something like WOO strategy calculator for this task, yes?
tringlomane
tringlomane
  • Threads: 8
  • Posts: 6281
Joined: Aug 25, 2012
January 16th, 2015 at 12:44:14 AM permalink
Quote: RS

Awesome. And I can use something like WOO strategy calculator for this task, yes?



It will be a bit of a pain as you'll have to adjust the non paying hand category in your Web browser address bar. I wished I would have realized this method sooner. I have always been curious about deuces strategy although it sounds like there aren't too many changes to it according to tongni.

I have a feeling I'm going to be killing some time tomorrow over this.
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
January 16th, 2015 at 3:47:46 AM permalink
Actually, come to think of it, would you have to assign a 1x to losing hands?

Loser: 0
high pair: 1 + return
2 pair: 2 + return*2 (not *3)
3oak: 3 + return*3 (not *4)
etc.

Basically subtract 1 from the multiplier. Because really, everything has a 1x multiplier then gets an added multiplier on top of that (so high pair is normal 1x plus another 1x multiplier for 2x, two pair is 1x plus an extra 2x totalling 3x, etc.).


I see a loser as having 0 value because you gain nothing additional. Then again, maybe my logic is all scrambled up.


PS: I can't figure out how to change the paytable in the URL. When I change the first "d-0" to "d-1", for example, errors arise. I assume the first one (the first d) is used for the losing hands payout. ? Using https://wizardofodds.com/games/video-poker/strategy/a-1-b-74-c-1-d-0-d-1-d-2-d-3-d-4-d-6-d-9-d-25-d-50-d-800/
tringlomane
tringlomane
  • Threads: 8
  • Posts: 6281
Joined: Aug 25, 2012
January 16th, 2015 at 9:29:14 AM permalink
I'm able to do it with Google Chrome. It takes awhile to crunch the payout though.

9/6 JoB where you get half your bet back for "nothing" hands.

https://wizardofodds.com/games/video-poker/strategy/a-1-b-74-c-1-d-0.5-d-1-d-2-d-3-d-4-d-6-d-9-d-25-d-50-d-800/
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
January 16th, 2015 at 1:26:05 PM permalink
Quote: tringlomane

I'm able to do it with Google Chrome. It takes awhile to crunch the payout though.

9/6 JoB where you get half your bet back for "nothing" hands.

https://wizardofodds.com/games/video-poker/strategy/a-1-b-74-c-1-d-0.5-d-1-d-2-d-3-d-4-d-6-d-9-d-25-d-50-d-800/



Ah, it works for some numbers (0.5) but not others (1 or 5). And can't go to 2 or more decimals.
tringlomane
tringlomane
  • Threads: 8
  • Posts: 6281
Joined: Aug 25, 2012
January 16th, 2015 at 3:06:05 PM permalink
Quote: RS

Ah, it works for some numbers (0.5) but not others (1 or 5). And can't go to 2 or more decimals.



Yeah I am having issues too with some games. I got it to work for deuces...hopefully I did it right.

https://wizardofodds.com/games/video-poker/strategy/a-1-b-44-c-1-d-0.24852-d-0.99704-d-1.74555-d-2.74259-d-3.24259-d-3.73962-d-7.48221-d-8.24555-d-13.49407-d-100.99407-d-400.99407/

And it looks like it converges on the multiplier independent return number.

But if I try it for JoB with what I believe is the right numbers, I'm getting a System.OutOfMemoryException though.

https://wizardofodds.com/games/video-poker/strategy/a-1-b-74-c-1-d-0.24846-d-0.9969-d-1.7454-d-2.4938-d-3.7392-d-5.7330-d-6.9815-d-13.2454-d-26.739-d-401.739/

:(
djatc
djatc
  • Threads: 83
  • Posts: 4477
Joined: Jan 15, 2013
January 16th, 2015 at 4:01:55 PM permalink
Somebody will make decent coin if they programed a ux/qq tutor. I love ux but hate the strategy I have to memorize.
"Man Babes" #AxelFabulous
tringlomane
tringlomane
  • Threads: 8
  • Posts: 6281
Joined: Aug 25, 2012
January 17th, 2015 at 12:28:02 AM permalink
Quote: djatc

Somebody will make decent coin if they programed a ux/qq tutor. I love ux but hate the strategy I have to memorize.



Yeah, the strategy is definitely tougher, but people that love 10/7 DB should find it easier...lol

I also got the JoB link I posted to calculate now. Maybe because its 2am here it worked? It doesn't totally match JB's return calucator, but it's within 0.01%. Also the strategy doesn't quite exactly match JB's calculator either...but it's really close. Also the output places the 4 flush line above AA, KK, QQ, JJ, which is misleading. As the only flushes better than AA-JJ have 3 high cards. But the program places the 4 flushes with zero high cards above AA-JJ as well. Because of equity I guess...

However, using the formula for each pay including "nothing"...

0.5*hand_prize + 0.25*hand_multiplier*overall_return_from_JB's_VP_analyzer

Seems to give you a good approximation on strategy with JB's strategy calculator. I'll do more testing tomorrow night. I will have more time late night tomorrow.
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
January 17th, 2015 at 5:36:18 AM permalink
What about the average multiplier, tringlomane? Could that be a reason the strategies don't match up?
tringlomane
tringlomane
  • Threads: 8
  • Posts: 6281
Joined: Aug 25, 2012
January 17th, 2015 at 10:09:45 PM permalink
now I'm getting errors like this...sigh

Error executing child request for cache/a-1-b-57-c-1-d-0.2435-d-0.987-d-1.2305-d-2.474-d-3.948-d-4.935-d-6.922-d-25.7305-d-40.974-d-80.974-d-80.974-d-200.974-d-25.974-d-400.974.aspx.
Description: An unhandled exception occurred during the execution of the current web request. Please review the stack trace for more information about the error and where it originated in the code.

Exception Details: System.Web.HttpException: Error executing child request for cache/a-1-b-57-c-1-d-0.2435-d-0.987-d-1.2305-d-2.474-d-3.948-d-4.935-d-6.922-d-25.7305-d-40.974-d-80.974-d-80.974-d-200.974-d-25.974-d-400.974.aspx.
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
January 17th, 2015 at 11:11:45 PM permalink
JB CAN U PLZ FIXES? :)


PS: I tried messing with VPW, couldn't figure it out. Is it possible to create manual hands, like 3 of a kind with 2 kickers? ie: TTA9 ... Basically manually create all QQ hands.
tringlomane
tringlomane
  • Threads: 8
  • Posts: 6281
Joined: Aug 25, 2012
January 17th, 2015 at 11:40:30 PM permalink
Likely the most common quarter ult x played...8/5 DDB. I wanted to create this first because I might be playing it for 60c a round next week. It's not bad for a 2c unit!

https://wizardofodds.com/games/video-poker/strategy/a-1-b-57-c-1-d-0.2433-d-0.9865-d-1.2298-d-2.473-d-3.946-d-4.933-d-6.919-d-25.73-d-40.973-d-80.973-d-80.973-d-200.974-d-25.973-d-400.973/

As for the errors...sometimes I get them, sometimes I don't. Also this strategy works for 3-play as well, which I would be playing. No changes.


RS: for QQ, just use JB's calculator. You have to select the Quick Quads option instead of 5 Card Draw.
tringlomane
tringlomane
  • Threads: 8
  • Posts: 6281
Joined: Aug 25, 2012
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
January 18th, 2015 at 1:26:23 AM permalink
I want to generate a paytable or something like that so I can plug it into VPW to practice/learn.
tringlomane
tringlomane
  • Threads: 8
  • Posts: 6281
Joined: Aug 25, 2012
January 18th, 2015 at 1:33:13 AM permalink
Quote: RS

I want to generate a paytable or something like that so I can plug it into VPW to practice/learn.



Ugh yeah...that might be a problem. As JB said, he is unaware of any VP program, other than his own, that allows you to put in a non-zero value for "nothing".

First look at 6/5 Bonus...looks the same as 7/5 Bonus except 3 to a Flush with QT is better than QT suited in 6/5 Bonus.

https://wizardofodds.com/games/video-poker/strategy/a-1-b-37-c-1-d-0.2454-d-0.9908-d-1.7363-d-2.482-d-3.963-d-5.2-d-5.945-d-13.236-d-20.982-d-40.982-d-25.982-d-400.982/
RS
RS
  • Threads: 62
  • Posts: 8626
Joined: Feb 11, 2014
January 18th, 2015 at 1:42:25 AM permalink
The problem sort of is you have to reverse engineer the damn paytable to get the correct strategy.

In other words, I think it might just be easier to look at the known strategy, then modify the paytable in a guess-like fashion. ie: "oh a 3 to a flush is a hand to be held, or 4 to inside straight should be held, let's give them more weight". Then with the new strategy, it says to hold XX over YY, but the known strategy is opposite of that, so keep changing the values in the paytable until the strategy matches. The strategy is all that matters. The paytable could return 450% or 2%....as long as the strategy is the same as the known strategy, it works.

It really doesn't matter what the non-paying hand is on the paytable. Perhaps I'll do some work tonight, but, I have a feeling this'll be a tough task.
tringlomane
tringlomane
  • Threads: 8
  • Posts: 6281
Joined: Aug 25, 2012
January 18th, 2015 at 2:07:09 AM permalink
Quote: RS

The problem sort of is you have to reverse engineer the damn paytable to get the correct strategy.

In other words, I think it might just be easier to look at the known strategy, then modify the paytable in a guess-like fashion. ie: "oh a 3 to a flush is a hand to be held, or 4 to inside straight should be held, let's give them more weight". Then with the new strategy, it says to hold XX over YY, but the known strategy is opposite of that, so keep changing the values in the paytable until the strategy matches. The strategy is all that matters. The paytable could return 450% or 2%....as long as the strategy is the same as the known strategy, it works.

It really doesn't matter what the non-paying hand is on the paytable. Perhaps I'll do some work tonight, but, I have a feeling this'll be a tough task.



That's exactly what I'm doing. But you need to have a program that allows you to have a non-zero value for hands worse than jacks or better. And the only program I can think of doing that is JB's strategy calculator...via the web address bar. He defaults to "0" for "garbage" otherwise.
  • Jump to: