Video Poker Super Learning

[backsider]

The paytables seem rather poor but I guess the edge lies in the royal on all the games. Royals look like must hits, or else there'll be lots of empty pockets. I do pretty good on the software trainers on some of those games so I need to learn some adjustments and more games  Hope to see you there!

[Frank Kneeland]

The Royal Flush is just another hand on VP. Try plugging in a 12,000 coin Royal on all those games and see what the return is at those numbers, because with 4% Meter-Rise that's what you'll have to play on nearly every day.Think of it like a Triple-Quadruple on Royal Flushes promotion that never ends.Will you need to weather some large losses between JP. Yes!It's still going to be the best thing in all of LV by a long-shot.~FK   

[backsider]

I don't know..playing for royals that are required much more than on other machines just seems overly risky and seems to take a lot of the fun factor out of it. What if you run into one of those no-royals streaks on these machines, but when you do hit them they're at other casinos that "only" pay 4000 credits? It feels like a lot of good luck is needed on them.

[New2vp]

The Royal Flush is just another hand on VP. Try plugging in a 12,000 coin Royal on all those games and see what the return is at those numbers, because with 4% Meter-Rise that's what you'll have to play on nearly every day.~FK A Conjecture, a Correction, and a Question:Conjecture:  It will be interesting to see how prevalent 12000-coin Royal Flushes are.  Since there are so many machines, there certainly could be one on most days if they get a lot of play since there could be a dozen or more royal flushes every day; the top prize on most days might even be higher than that.  However, with several admittedly simplifying assumptions, I'm calculating that the AVERAGE royal jackpot won will be at least 1000 coins lower than 12000.  It is tough to calculate this precisely without knowing how different people will choose to play and how aggressively they will on average go after the royal; e.g., some may choose to play inferior games because they know them better and might reason that they would be happy with even an 8000-coin jackpot rather than going after the current leader.Correction(?):  Is the EV on the Joker game in Dancer's article correct?  It looks like the EV for a game with the royal paying 4700 rather than the indicated reset value of 4000...unless the joker game's reset actually is 4700 and different from the other 7 games.Question:  When the bank of machines is initialized, will the M start all the royals at 4000, or will they be seeded with some additional dollars to resemble what a more typical distribution of jackpots will look like after some play has occurred?  Just curious as to standard practice there and perhaps at other casinos although I realize that this is a non-standard offering.This idea is a very clever and intriguing concept!  I hope you keep us informed on how the bank is being received by the casino and patrons and if you can share any statistics such as how much play is being received or average jackpot size, that would be enlightening.

[Frank Kneeland]

Conjecture:  It will be interesting to see how prevalent 12000-coin Royal Flushes are.  Since there are so many machines, there certainly could be one on most days if they get a lot of play since there could be a dozen or more royal flushes every day; the top prize on most days might even be higher than that.  However, with several admittedly simplifying assumptions, I'm calculating that the AVERAGE royal jackpot won will be at least 1000 coins lower than 12000.  It is tough to calculate this precisely without knowing how different people will choose to play and how aggressively they will on average go after the royal; e.g., some may choose to play inferior games because they know them better and might reason that they would be happy with even an 8000-coin jackpot rather than going after the current leader.I calculate the average high Royal to be around $10,600. And by that time that hits at around $11,400 the next one in cue will be $10,600. $12,000+ will be common, but not all the time.[QUOTE]Correction(?):  Is the EV on the Joker game in Dancer's article correct?  It looks like the EV for a game with the royal paying 4700 rather than the indicated reset value of 4000...unless the joker game's reset actually is 4700 and different from the other 7 games.[/QUOTE]Good eye. That was indeed the return for 4700 coin. I didn't write the article Bob did.[QUOTE]Question:  When the bank of machines is initialized, will the M start all the royals at 4000, or will they be seeded with some additional dollars to resemble what a more typical distribution of jackpots will look like after some play has occurred?  Just curious as to standard practice there and perhaps at other casinos although I realize that this is a non-standard offering. [/QUOTE]They will be seeding them to jump start the process.[QUOTE]This idea is a very clever and intriguing concept!  I hope you keep us informed on how the bank is being received by the casino and patrons and if you can share any statistics such as how much play is being received or average jackpot size, that would be enlightening.[/QUOTE]I'll be happy to if I know it myself.~FK

[Frank Kneeland]

I don't know..playing for royals that are required much more than on other machines just seems overly risky and seems to take a lot of the fun factor out of it. What if you run into one of those no-royals streaks on these machines, but when you do hit them they're at other casinos that "only" pay 4000 credits? It feels like a lot of good luck is needed on them.Here's an idea then. You cannot hit a 4000 coin Royal if you never play a machine with a 4000 coin payoff.  Do all your play on 12,000 coin Jackpots and all your Royal will be big ones.This will be possible at the M. High Royals are going to be daily events.~FK

[New2vp]

I calculate the average high Royal to be around $10,600. And by that time that hits at around $11,400 the next one in cue will be $10,600. $12,000+ will be common, but not all the time.
Well, good.  Our calculations are close then.  I found the equilibrium point for the highest royal hitting at 10,653.13 coins.  My simplifying assumptions were (1) that the overall average cycle time to build each jackpot would be equal to the sum of the average cycle times of each game, (2) every play would be on the game with the highest royal flush at the time, (3) every play would use computer perfect strategy AS IF the jackpot was ALWAYS at the average value.I know that there will be deviations from my assumptions.  For example, the game with the highest royal is not always the best play.  The computer perfect strategy will change during the quest to take down each royal, the "average" strategy may not be the average of the actual strategies, and people will not always choose the best game when they wager their coins.  The assumptions were set up so that I could find a solution using relatively easy math with the hope that the assumptions were robust enough so that the actual expected results would not vary too much from those derived from the assumptions.And we can be confident that the AVERAGE actual jackpot hit must be related to the AVERAGE actual number of plays per royal because the average meter rise to be added to the reset comes directly from the AVERAGE actual number of plays per royal or average experienced cycle time, if Ican call it that.  If I used the 10,653 figure for each of the 8 games as the average
size of a royal, I was able to balance out the meter rise as equal to
1/2% x 5 coins x [Cycle time for Game 1 + Cycle time for Game 2 + Cycle
time for Game 3 + ... + Cycle time for Game 8].  Each game is played only part of the time, but is accumulating a rising jackpot on each play including that of the other games.  The cycle times are (for each game at 10,653 coins):



16-13
Deuces Wild
37257.06


15-7 Kings or Better Joker Wild
37696.14


6-5 Bonus Poker
31871.05


8-5 Double Double Bonus Poker
31630.92


9-6-4 Double Bonus
31734.78


6-5 Super Double Bonus
32059.91


7-5 Jacks or Better
31902.52


7-5 Bonus Poker Deluxe
31972.98






Total
266125.37

with the last figure off in the 2nd decimal place due to rounding error in the individual entries.266125.37 plays x 0.5% x 5 coins per play = 6653.13 coins added on to the reset value of 4000 coins yields 10653.13 which balances to the average jackpot size in each calculation.I can follow that your value would change during play if you started at 10,600 to 11,400 if you assumed 32,000 plays (32,000 x 0.5% x 5 = 800 coin difference) for each take down WHILE COINS ARE WAGERED ON THAT GAME.  That would mean that you had an average of 6600 meter rise (from 4000 to 10600) accumulated WHILE COINS WERE WAGERED OTHER GAMES.  This, in turn, would imply that you had an average of 264,000 plays (6600 divided by 5 coins divided by 0.5%) while playing the other 7 games.  But that would mean that you averaged 37,714 plays for the OTHER games, but only 32,000 on EACH GAME that you were playing.  I think that may be a bit of an inconsistency in the calculation somewhere that is too big to be explained by rounding.  Your figures would be more consistent and work better if 10,600 is closer to the (rounded) average number at which the royal hits, instead of the average number at which play begins on a particular game.  It seems to me at least that the meter rise during play of OTHER GAMES should on average be 7 times the meter rise during play of THE CURRENT GAME before it hits.Nevertheless, I think this multi-game progressive concept will be very entertaining and I hope that M doesn't decide to pull the games too quickly.  This looks to me like quite an accomplishment on your part. 
Good eye. That was indeed the return for 4700 coin. I didn't write the article Bob did.Sorry, I didn't mean to imply that you made a mistake here; I knew it was Bob's article.  I just wanted to know if my assumption was correct, to see if I should change the cycle time for the Joker game.
They will be seeding them to jump start the process. Excellent!  That makes perfect sense.Thanks for answering so quickly.

[Frank Kneeland]

Who on earth are you? And would you like a job? I rounded down. It's like you copied my math work.And your calculations are in fact slightly better than what I did.Would you mind if I posted your post on vpFREE?I needed to do one like this anyway and it seems silly to redo your work when it was flawless.~FKP.S. I have some dishes that need cleaning as well.

[BOOPSAHOY]

My head exploded after reading that!

[shadowman]

Who on earth are you?  I've come to the conclusion that new2vp is really the next advancement in computer technology after Watson. They've been testing it here to see how well it co-exists with the us mere mortals.