## Poll

 I'm fine with the 5 + 20 wager 3 votes (23.07%) I hate the 5 + 20 wager 3 votes (23.07%) Love the RTP at 99% 7 votes (53.84%) Love the x100 dealt royal multiplier 4 votes (30.76%) Love the x50 multiplier 5 votes (38.46%) I would play this in a casino 8 votes (61.53%) Hate the game altogether! No votes (0%) Interesting!!!!! 3 votes (23.07%) Love not paying additional wagers after wins 4 votes (30.76%)

13 members have voted

beachbumbabs
Joined: May 21, 2013
• Posts: 13110
Thanks for this post from:
September 23rd, 2018 at 10:49:40 AM permalink
Here are some specific errors I've seen.

1. Top hand holds 4h6h8h. Bottom hand holds KhKc. Hand should hold that bonus. Instead it promoted the Kh , making less value.

2. Top hand holds KhQhJc. Bottom hand holds KcKh. Bonus should hold, instead replacing Kh with Kc for less value.

3. Top hand holds KhJcTh. Bottom hand holds QhQs. Qh rises, but replaces Kh instead of Th to make the straight, for less value.

4. Top hand contains 4h6h8h. Bottom hand promotes 9h AND REMOVES 8h. 2 errors there, imo, though related. A) not considering straight potential of other cards when removing the 8. B) should remove either of the other cards if it's going to substitute at all, as 986 is equal to 864 in straight or SF potential.
If the House lost every hand, they wouldn't deal the game.
Mission146
Joined: May 15, 2012
• Posts: 11743
September 23rd, 2018 at 11:55:30 AM permalink
Here's one:

7c, 3d, 10s

1.) I forget the exact cards in the winning hand, but inexplicably, it replaced the 3d with a 5c when it could have replaced it with a 6c, thereby giving two outs to the Straight-Flush on the top hand for the following hand. In other words, what ended up being:

7c, 5c, 10s

For the next hand should have been:

7c, 6c, 10s

There's absolutely no reason to favor the five over the six.

2.) I started with:

Ac, 5s, 6c

I forget the exact five cards of the finishing hand, but given the choice between replacing the 5d with a Kc, it instead replaced the Ac with a 4d!!!

Granted, that gives the 4x for the straight, but replacing the 5s with the Kc still would have given a 3x Flush AND would have set up a potential Qc on a following hand for the 50x Royal up top as well as a straight opportunity with any other queen.

3.) I started with:

Qc, 2h, 6h

The finishing five card hand:

Kh, Qs, Kc, Jd, Qd

It replaced the Qc with the Kh for the Flush, but it almost seems like it would be better to replace one of the hearts with Kc to set up a potential 50x Royal Flush up top needing only the Ac. Besides that, any Club would still give a Flush two hands later as well as any Ace/Jack giving a straight and the Jc for a potential SF.

It seems like the auto-chooser only thinks one hand into the future.

What is the 99.1% return predicated upon, by the way? You have to make the Optimal hold based on so many factors which can include:

-Base return of the 9/5 paytable
-Emphasis just on finishing with a WINNING hand (like Multi-Strike) in order to continue the multipliers
-Influence what the Streak Auto-Picker will do with the Three-Card hand, even though it doesn't always do what makes the most sense.*

*One example, though an example where it would do the right thing, is if you had:

Ac, Kc, 3h

And the five-card hand:

Jh, Jc, Qc, 5d, 4h

Obviously, you would want to hold Jh, Jc, Qc because it gives you both a winning hand AND a 50x multiplier going into the next hand. There may be other situations where the correct hold could be influenced both by probability of winning AND getting a particular card to the top...but if the Streak thing isn't going to make the best decision to begin with, then how do you factor that in?

With all due respect, the auto-holder is just really, really, bad. I really think the best thing to do is just let the player pick whatever card he or she wants if you can't get the auto-holder to perfection. Even if you do get it mathematically perfect, you're still going to irritate some people with it holding one card as opposed to the card they would like it to hold.
Vultures can't be choosers.
unJon
Joined: Jul 1, 2018
• Posts: 924
Thanks for this post from:
September 23rd, 2018 at 11:57:52 AM permalink
Quote: beachbumbabs

Here are some specific errors I've seen.

1. Top hand holds 4h6h8h. Bottom hand holds KhKc. Hand should hold that bonus. Instead it promoted the Kh , making less value.

2. Top hand holds KhQhJc. Bottom hand holds KcKh. Bonus should hold, instead replacing Kh with Kc for less value.

3. Top hand holds KhJcTh. Bottom hand holds QhQs. Qh rises, but replaces Kh instead of Th to make the straight, for less value.

4. Top hand contains 4h6h8h. Bottom hand promotes 9h AND REMOVES 8h. 2 errors there, imo, though related. A) not considering straight potential of other cards when removing the 8. B) should remove either of the other cards if it's going to substitute at all, as 986 is equal to 864 in straight or SF potential.

Too hand holds 3sKsAs. Bottom hand promotes 8s for 3s, killing a chance for a straight flush with 2s.
The race is not always to the swift, nor the battle to the strong; but that is the way to bet.
Mission146
Joined: May 15, 2012
• Posts: 11743
September 23rd, 2018 at 12:02:38 PM permalink
Quote: unJon

Too hand holds 3sKsAs. Bottom hand promotes 8s for 3s, killing a chance for a straight flush with 2s.

Here's another:

The bottom hand brings up Qc killing the QK SF/RF draw for the following hand on the top hand. For absolutely no reason whatsoever. If this were to happen in an actual casino, with actual money, no way I ever play this game again.
Vultures can't be choosers.
gordonm888
Joined: Feb 18, 2015
• Posts: 1668
Thanks for this post from:
September 23rd, 2018 at 12:28:14 PM permalink
Quote: unJon

I’d also be interested in knowing if better EV to have three of a kind multiplier or a pair multiplier with and open ended straight flush or royal draw. The program seems to pick the three of a kind every time. My gut says it may be better to have KcQcKd than KcKsKd. Not sure though.

For the bonus multiplier:

I calculate that 3oak is better than pair with open-ended SF draw.

3oak vs pair with a Royal Draw is not as close. 3oak never improves, it always remains 3oak. Pair with a royal draw is tricky - because the cards you need for the royal are either Q,K A -and high cards such as Q,K,A occur more frequently in winning primary hands than low cards -because many winning hands result from drawing to high cards. Also, the royal draw (similar to the oe straight flush draw) will frequently upgrade to either a 3X Flush (or more rarely the 4x Straight) when the royal doesn't hit. Overall, I think a pair with Royal draw is better for the player

Having said that, I have no problem with the existing algorithm that always upgrades the bonus multiplier to the highest possible multiplier for the next hand. That is understandable and transparent to the player. Foregoing an upgrade to an "x5 3oak multiplier" in favor of a "x2 Pair multiplier" in order to preserve the Royal draw aspect of the bonus hand would create the feeling in most players that they got unfairly screwed by the game.
Mission146
Joined: May 15, 2012
• Posts: 11743
Thanks for this post from:
September 23rd, 2018 at 1:51:46 PM permalink
RealizeGaming,

I like you, so this is actually a difficult post to make:

General Observations

1. This basically plays like a particularly top-heavy penny slot machine. I did four different, "Sessions," each with the goal of getting to either \$9,000 or up to \$11,000, after making the first bet, I'd fail to ever even sniff \$10,000 again.

2. The \$5 + \$20 bet is beyond the pale and completely unheard of in terms of what is essentially a single hand Video Poker game.

2a.) I think that some casinos might even agree with this perspective from a jackpot liability standpoint. For one example, consider \$5 denomination Ultimate X. In the event a player makes a \$50 total bet on single-handed, the best possible result is a dealt FH (DDB) followed by any kind of a Royal. In this case, \$100 total is bet and the player wins \$225 + \$240,000 = \$240,225 which is 2,402.25x the total amount of \$100 bet between the two hands.

Even in triple-line \$5 Ultimate X, a dealt Royal following a dealt FH would pay out \$720,000, on a total bet of \$150.

In the case of your game, if the casino offered a \$5 denomination version, the total bet would be \$125 with a possible pay of TWO MILLION DOLLARS+ with a dealt three-card Royal + RF and some other hands possible. That's 16,000x the total amount bet. There aren't many slot machines that even offer payouts that high relative to the bet amount...though there are a few.

In the case of this \$1 denomination game, dealt three card Royal + RF on the main hand is still \$400,000 won on a \$25 total bet. Again, that's compared to \$1 Ultimate X, three-hands, dealt FH + dealt Royal on the following hand for \$144,000 which is \$60 total bet and \$30 per hand bet.

It's just a huge jackpot liability.

2b.) This seems like a really top-heavy game from the player perspective, too.

3.) It doesn't seem like the bonus offers any great benefit relative to the bet amount. My three best results during some 400+ hands were:

---\$5 win + Quad Jacks with a 2x for \$500 = \$505 (barely 20x total bet)

---I won Thirteen Hands in a row and ended up with \$455

\$5 + \$15 + \$15 + \$35 + \$35 + \$35 + \$35 + \$35 + \$35 + \$105 + \$35 + \$35 + \$35

Which is only 18.2x the amount bet. Granted, that's better than the 3x I would have otherwise won on this particular string of hands, but I was down by so much at this point that it was barely helpful. It honestly didn't even really inspire any confidence in me for future hands.

---\$195 total win, but I forget exactly how.

4.) I caught a FH without anything up top and then whiffed on the next hand. It's not fun to not even get 2-FOR-1 on a Full House.

5.) I don't see anyway you make everyone happy with the three-card hand other than letting people pick their own. However, as you pointed out, that's time-consuming, so the casino might not like it.

SUGGESTIONS:

1.) Simplify the game dramatically by allowing for multiple hands, but with the same Three-Card hand per initial bet. In other words, a pair would just stick and be 2x as long as the player kept winning base hands. One exception could be that the three cards will completely redraw if they are not a winning three-card hand. This change would also enable you to drop the bet to something other than \$5 + \$20. I'm not going to do the math on this one, but there might be a way to do it with a \$5 + \$5 bet if the multiplier hand cannot improve.

Another alternative would be to keep the same three-card hand, but allow the game to draw new three-card hands for the same, "Play," from a fresh deck and to replace the current three-card hand if the new one is better.

OR:

2.) I know you won't like this, but you can treat every game as one play rather than theoretically unlimited. If you do this, you could simply have the potential for HUGE multipliers:

Pair: 5x
Flush: 10x
Straight: 15x
Trips: 25x
Straight Flush: 50x
Mini Royal: 100x

(0.169412*5) + (0.049593*10) + (0.032579*15) + (0.002353*25) + (.001991 * 50) + (0.000181*100) = 2.00815 (Average Multiplier)

-You still have the jackpot liability issue to worry about, even more so, actually. However, the player will get a multiplier on about 25.6109% of all hands. More than that, any winning five-card hand with a multiplier is profitable for the player.

The player would be making a ten-credit bet, which is 2x the normal amount, and would have an average multiplier of 2.00815. Therefore, the return of a game would be changed as follows:

(Base Return * Average Multiplier)/2 = New Return

In the case of 9/6 DDB:

(.9898*2.00815)/2 = 0.993833435

The game would also come with zero changes to optimal strategy because there is no, "Streak," component and the player cannot influence the three-card hand.
Vultures can't be choosers.
GWAE
Joined: Sep 20, 2013
• Posts: 8590
Thanks for this post from:
September 23rd, 2018 at 2:07:59 PM permalink
Quote: gordonm888

For the bonus multiplier:

I calculate that 3oak is better than pair with open-ended SF draw.

3oak vs pair with a Royal Draw is not as close. 3oak never improves, it always remains 3oak. Pair with a royal draw is tricky - because the cards you need for the royal are either Q,K A -and high cards such as Q,K,A occur more frequently in winning primary hands than low cards -because many winning hands result from drawing to high cards. Also, the royal draw (similar to the oe straight flush draw) will frequently upgrade to either a 3X Flush (or more rarely the 4x Straight) when the royal doesn't hit. Overall, I think a pair with Royal draw is better for the player

Having said that, I have no problem with the existing algorithm that always upgrades the bonus multiplier to the highest possible multiplier for the next hand. That is understandable and transparent to the player. Foregoing an upgrade to an "x5 3oak multiplier" in favor of a "x2 Pair multiplier" in order to preserve the Royal draw aspect of the bonus hand would create the feeling in most players that they got unfairly screwed by the game.

I agree with your end part. say you had a pair with open ended royal flush. Next had would make a 3oak but it doesnt hold it. Your next 5 card hand is a royal. Now you are getting 2x instead of 3x. That would annoy people. I personally would keep the open ended royal flush for an outside shot of a 50k royal. I think the only way to satisfy everyone would be to either
1. Allow them to make their own decision as to what card to hold, but that would really slow the game down.
or
2. Make it all random. Maybe on the deal it could randomly highlight 1 position on the 5 card hand and 1 position on the 3 card hand and that card moved up there every hand. Even if it took a royal and made it a losing 3 card multiplier. I think it would add some craziness to the game and would probably allow them to up the multipliers some or make the bonus bet less expensive which I would be in favor of
Expect the worst and you will never be disappointed. I AM NOT PART OF GWAE RADIO SHOW
RealizeGaming
Joined: Aug 1, 2013
• Posts: 483
September 23rd, 2018 at 2:56:09 PM permalink
Quote: beachbumbabs

Here are some specific errors I've seen.

1. Top hand holds 4h6h8h. Bottom hand holds KhKc. Hand should hold that bonus. Instead it promoted the Kh , making less value.

2. Top hand holds KhQhJc. Bottom hand holds KcKh. Bonus should hold, instead replacing Kh with Kc for less value.

3. Top hand holds KhJcTh. Bottom hand holds QhQs. Qh rises, but replaces Kh instead of Th to make the straight, for less value.

4. Top hand contains 4h6h8h. Bottom hand promotes 9h AND REMOVES 8h. 2 errors there, imo, though related. A) not considering straight potential of other cards when removing the 8. B) should remove either of the other cards if it's going to substitute at all, as 986 is equal to 864 in straight or SF potential.

Thanks babs.

1. I agree with what you are saying and we will look into it. I'm positive it is making this move because it is programed to work to upgrade the existing string hand to a higher value multiplier.

2. Another example where I see what your saying. The hand is keeping the straight intact but taking a step backward on the string hand because it is creating a gap in the potential straight flush. It is better to have the KhQhJc instead of replacing the Kh for the Kc. I think this was a hold over from our first build where placing a card into the hand added a plus one to the multiplier. Either way, I agree with you.

3. Agree.

4. Agree
RealizeGaming
Joined: Aug 1, 2013
• Posts: 483
Thanks for this post from:
September 23rd, 2018 at 3:11:54 PM permalink
Quote: Mission146

Here's one:

7c, 3d, 10s

1.) I forget the exact cards in the winning hand, but inexplicably, it replaced the 3d with a 5c when it could have replaced it with a 6c, thereby giving two outs to the Straight-Flush on the top hand for the following hand. In other words, what ended up being:

7c, 5c, 10s

For the next hand should have been:

7c, 6c, 10s

There's absolutely no reason to favor the five over the six.

2.) I started with:

Ac, 5s, 6c

I forget the exact five cards of the finishing hand, but given the choice between replacing the 5d with a Kc, it instead replaced the Ac with a 4d!!!

Granted, that gives the 4x for the straight, but replacing the 5s with the Kc still would have given a 3x Flush AND would have set up a potential Qc on a following hand for the 50x Royal up top as well as a straight opportunity with any other queen.

3.) I started with:

Qc, 2h, 6h

The finishing five card hand:

Kh, Qs, Kc, Jd, Qd

It replaced the Qc with the Kh for the Flush, but it almost seems like it would be better to replace one of the hearts with Kc to set up a potential 50x Royal Flush up top needing only the Ac. Besides that, any Club would still give a Flush two hands later as well as any Ace/Jack giving a straight and the Jc for a potential SF.

It seems like the auto-chooser only thinks one hand into the future.

What is the 99.1% return predicated upon, by the way? You have to make the Optimal hold based on so many factors which can include:

-Base return of the 9/5 paytable
-Emphasis just on finishing with a WINNING hand (like Multi-Strike) in order to continue the multipliers
-Influence what the Streak Auto-Picker will do with the Three-Card hand, even though it doesn't always do what makes the most sense.*

*One example, though an example where it would do the right thing, is if you had:

Ac, Kc, 3h

And the five-card hand:

Jh, Jc, Qc, 5d, 4h

Obviously, you would want to hold Jh, Jc, Qc because it gives you both a winning hand AND a 50x multiplier going into the next hand. There may be other situations where the correct hold could be influenced both by probability of winning AND getting a particular card to the top...but if the Streak thing isn't going to make the best decision to begin with, then how do you factor that in?

With all due respect, the auto-holder is just really, really, bad. I really think the best thing to do is just let the player pick whatever card he or she wants if you can't get the auto-holder to perfection. Even if you do get it mathematically perfect, you're still going to irritate some people with it holding one card as opposed to the card they would like it to hold.

You make a great point Mission146. The autopicker only thinks one pick into the future because the game requires you win in order to keep playing with the current built up string hand. The 99.1% return is based off of 1 billion simulations of the game using the DDB strategy and the programmed logic used by the autopicker. Click on the "I" button in the demo to see some more information about the return and if you have any other questions about it, please post them and we will try to get them answered for you.

I don't think the autopicker is that bad based on the game, although I do agree with the majority of player that some of the logic pertaining to flushes and straight flushes could be improved. I do like the idea of letting the player pick the cards and maybe that would be very doable, but I think players would only play it for real money if they had some experience with it. I don't think they would just be walking through the casino and take a chance on it because it would come across as too complicated for a new player.
RealizeGaming
Joined: Aug 1, 2013
• Posts: 483
Thanks for this post from:
September 23rd, 2018 at 3:32:19 PM permalink
Quote: Mission146

RealizeGaming,

I like you, so this is actually a difficult post to make:

General Observations

1. This basically plays like a particularly top-heavy penny slot machine. I did four different, "Sessions," each with the goal of getting to either \$9,000 or up to \$11,000, after making the first bet, I'd fail to ever even sniff \$10,000 again.

2. The \$5 + \$20 bet is beyond the pale and completely unheard of in terms of what is essentially a single hand Video Poker game.

2a.) I think that some casinos might even agree with this perspective from a jackpot liability standpoint. For one example, consider \$5 denomination Ultimate X. In the event a player makes a \$50 total bet on single-handed, the best possible result is a dealt FH (DDB) followed by any kind of a Royal. In this case, \$100 total is bet and the player wins \$225 + \$240,000 = \$240,225 which is 2,402.25x the total amount of \$100 bet between the two hands.

Even in triple-line \$5 Ultimate X, a dealt Royal following a dealt FH would pay out \$720,000, on a total bet of \$150.

In the case of your game, if the casino offered a \$5 denomination version, the total bet would be \$125 with a possible pay of TWO MILLION DOLLARS+ with a dealt three-card Royal + RF and some other hands possible. That's 16,000x the total amount bet. There aren't many slot machines that even offer payouts that high relative to the bet amount...though there are a few.

In the case of this \$1 denomination game, dealt three card Royal + RF on the main hand is still \$400,000 won on a \$25 total bet. Again, that's compared to \$1 Ultimate X, three-hands, dealt FH + dealt Royal on the following hand for \$144,000 which is \$60 total bet and \$30 per hand bet.

It's just a huge jackpot liability.

2b.) This seems like a really top-heavy game from the player perspective, too.

3.) It doesn't seem like the bonus offers any great benefit relative to the bet amount. My three best results during some 400+ hands were:

---\$5 win + Quad Jacks with a 2x for \$500 = \$505 (barely 20x total bet)

---I won Thirteen Hands in a row and ended up with \$455

\$5 + \$15 + \$15 + \$35 + \$35 + \$35 + \$35 + \$35 + \$35 + \$105 + \$35 + \$35 + \$35

Which is only 18.2x the amount bet. Granted, that's better than the 3x I would have otherwise won on this particular string of hands, but I was down by so much at this point that it was barely helpful. It honestly didn't even really inspire any confidence in me for future hands.

---\$195 total win, but I forget exactly how.

4.) I caught a FH without anything up top and then whiffed on the next hand. It's not fun to not even get 2-FOR-1 on a Full House.

5.) I don't see anyway you make everyone happy with the three-card hand other than letting people pick their own. However, as you pointed out, that's time-consuming, so the casino might not like it.

SUGGESTIONS:

1.) Simplify the game dramatically by allowing for multiple hands, but with the same Three-Card hand per initial bet. In other words, a pair would just stick and be 2x as long as the player kept winning base hands. One exception could be that the three cards will completely redraw if they are not a winning three-card hand. This change would also enable you to drop the bet to something other than \$5 + \$20. I'm not going to do the math on this one, but there might be a way to do it with a \$5 + \$5 bet if the multiplier hand cannot improve.

Another alternative would be to keep the same three-card hand, but allow the game to draw new three-card hands for the same, "Play," from a fresh deck and to replace the current three-card hand if the new one is better.

OR:

2.) I know you won't like this, but you can treat every game as one play rather than theoretically unlimited. If you do this, you could simply have the potential for HUGE multipliers:

Pair: 5x
Flush: 10x
Straight: 15x
Trips: 25x
Straight Flush: 50x
Mini Royal: 100x

(0.169412*5) + (0.049593*10) + (0.032579*15) + (0.002353*25) + (.001991 * 50) + (0.000181*100) = 2.00815 (Average Multiplier)

-You still have the jackpot liability issue to worry about, even more so, actually. However, the player will get a multiplier on about 25.6109% of all hands. More than that, any winning five-card hand with a multiplier is profitable for the player.

The player would be making a ten-credit bet, which is 2x the normal amount, and would have an average multiplier of 2.00815. Therefore, the return of a game would be changed as follows:

(Base Return * Average Multiplier)/2 = New Return

In the case of 9/6 DDB:

(.9898*2.00815)/2 = 0.993833435

The game would also come with zero changes to optimal strategy because there is no, "Streak," component and the player cannot influence the three-card hand.

Mission148, no need to feel bad about posting your thoughts. I seriously have no problems with anyone's thoughts and feedback and I encourage people to post them because it can only help me in the long run.

I understand what you are saying about the jackpot liability. It may be unrealistic to expect a casino to be willing to payout such a high amount. The current math model we use for the game allows the game to be very high risk-high reward, but it could be adjusted very easily to make it a "safer" game for the casino. We had one version of the game where the pair didn't reward a multiplier and the multipliers we used were much lower. We also had a very low multiplier for the dealt royal and the build up royal. These arrangements helped us to get the game down to a much smaller starting wager and lowered the maximum jackpot. My point is we have a starting point that can be adjusted in a number of ways to meet any desired RTP. We've had a number of meetings with companies who told us they are not concerned with the math being perfect as long as we have some basis for what we are doing. The companies told us they have a number of dedicated mathematicians who can do all that work. I agree because we have a number of different models we can use with the game.