SOOPOO
SOOPOO
Joined: Aug 8, 2010
  • Threads: 89
  • Posts: 3952
September 25th, 2016 at 9:43:46 AM permalink
Quote: gordonm888

, I have posted a number of math analyses in this forum (including one on the first page of this thread) that no one has commented on.



I appreciate your posts and analysis. I think your analysis of this particular game doesn't go far enough. What we (those not versed in analysis like you are) want is the house edge of the new game. I think that figuring it out will be EXTREMELY difficult. It's easy to figure out what the Ace high push means. How exactly to program in a flush 10/9 versus a king high pai gow with a Q/10..... thats way above my pay grade. I'm hoping it is not above yours!
Paigowdan
Paigowdan
Joined: Apr 28, 2010
  • Threads: 115
  • Posts: 5692
September 25th, 2016 at 9:44:22 AM permalink
Quote: gordonm888

TwoFeathers, you are among the many people in this forum that I do like and respect. You are also the only fellow southeastern-er I am aware of who posts with any regularity on this forum!

Shackelford is industrious, intellectually curious, bright and pays attention to quality in his work. He has achieved a lot and I respect him greatly.

Babs is smart and grounded, she impresses me as a decent human being - sort of a warden amongst the forum's misfits and weirdos.

I am fascinated and very impressed with Romes and Miplet especially. Also, there are a couple of dozen people who only post occasionally in the forum who I wish would post more often - I would love to learn from them.

However, I have posted a number of math analyses in this forum (including one on the first page of this thread) that no one has commented on, and my disappointment in that has probably soured my attitude towards the in-forum math geeks..

Most of all, I admit to having philosophical differences with PaigowDan and (especially) MathExtremist. They seem to have some good qualities but I have become weary of the parochialism, arrogance and condescension that they have directed at me (and others) in their posts. I am unaccustomed to condescension and, as a result, I'm pushing back at Dan with my own condescension. It is what it is.



Very fine, but expect push back when you:
a) Make ad hominen attacks on those you do not know ("Narrow slice of life/ of Frozen TV dinners")
b) Falsely criticize a fine game design effort by a designer that not only made it to the casino floor, but was signed by Shuffle Master and is getting serious gambling action by players at a fine casino.
c) On the issue of Arrogance: posting an impressive NASA and engineering resume (which I felt was to self-aggrandize and to put down others lacking such a fine background) - and without the relevant gaming industry work and creds. Furthermore, if I were so arrogant, I would not have openly credited and praised a competitor's game design work and success. Indeed, I defended the man.
d) It isn't enough to post or know math. One has to know game design, casino operations, player approaches to new games, merits of the mechanisms in games, and what have you.
e) If we disagree with your arguments, do not view it as an ad hominen attack, to which you respond with real ad hominen attacks and condescension. View it in terms of game mechanism discussion, or what a game designer achieved, what players may look for in a game, etc. and keep it to that.
f) do not feel hurt or soured if and when there is a tepid response to a post that contains fine work or insight. Some are blase. For myself, If I see something good in a game that others would love to rip apart, or fault a game's characteristics or argument, I get ripped apart as part of the blood sport of this site, and I'm used to it. I argue points, but I will, however, respond to and rip into personal attacks.
Last edited by: Paigowdan on Sep 25, 2016
Beware of all enterprises that require new clothes - Henry David Thoreau. Like Dealers' uniforms - Dan.
Paradigm
Paradigm
Joined: Feb 24, 2011
  • Threads: 40
  • Posts: 2036
September 25th, 2016 at 10:15:06 AM permalink
Quote: gordonm888

The dealer (or player) will be dealt a 7-card Ace-High Pai-Gow with a frequency of 9.4%. Here is what I calculate for a range of ace-high pai-gow hands with a composition-dependent model that I have.


Hand Win Push Lose EV
A-KQ-J743 0.1446 0.3753 0.4811 -0.340
A-KT-9732 0.1206 0.3177 0.5614 -0.447
A-QJ-T732 0.0911 0.2621 0.6486 -0.562
A-JT-9732 0.0426 0.2053 0.7527 -0.712
A-87-6432 0.000 0.1116 0.8887 -0.889


With an AKQxxxx hand, you are able to put a KQ in front, but your EV is still -0.34. When your Ace-high pai-gow is AQJTxxx, your hand up front drops to a QJ and your EV plummets to about -0.56. And if your 2nd card is a Jack or lower, then God help you.

So, allowing the dealer's hand to automatically push when its an ace-high pai-gow is worth something like 5 - 5.5% in EV. That's an enormous advantage for the dealer that the player needs to overcome by being able to see the dealer's hand and setting his hand optimally.


Getting this thread back on topic ;-), the above indicates that the frequency of the Ace High Pai Gow hand occurs for the dealer just under 10%. I think a less noticed mechanism is in the 6-8% range...players notice a bad event one in 10-11 hands. Once in 14 or so hands, while seeming a small difference, is important in "feel" of a mechanism.

The positive here would be that for some of those Ace High Pai Gow hands, the player may have less than an Ace High Pai Gow and the push mechanism may help them. Gordonm888, could I ask you to calc how often a player would benefit from pushing on the Ace High Dealer Pai Gow because otherwise they would lose?

5-5.5% HE shift to the house seems like a lot to overcome...I think the elimination of the commission on wins goes a long way in mitigating that shift and evidently seeing the dealer's hand gets you the rest of the way and then some to a net 1.6% HE. The 1.6% HE without commission [thanks for clarifying that for me Mips ;-)] mean the game would be better for the player than regular PGP...not sure a casino wants a PGP game for the long term with a smaller edge as the game's speed already requires higher minimum bets to increase dollar volume going through the game mechanism to be acceptably profitable in absolute terms. This is likely the reason Thunder Valley spread the game with the Commission.

On the topic "lack of surprise" or suspense of the dealer's hand being revealed at the end of the hand, the question is whether that excitement for the player can be substituted by watching the dealer hand reveal prior to squeezing their own hand. I would certainly play the game this way, leaving my packet of 7 cards untouched and then watching the winning bar being set based on the reveal.

I would have a tendency to just muck my cards if I couldn't beat the dealer...as I think about that, the game could incorporate that rule (e.g. the player shall muck their loser hands immediately) in an effort to speed up the game...that type of time and motion savings may provide some room in HE necessary to maintain profitability. If the dealer didn't have to reveal and check 50% of the player's hands because they were mucked as losers, that would be a positive for game speed.

Still getting my head around how all of this would come together as a player experience...I wish this game was on trial in Vegas as I would definitely play it while at G2E next week. There are a couple of games on trial per the NGCB website that are complete disasters vs. this game....sigh
Attempting to add value one post at a time
Paigowdan
Paigowdan
Joined: Apr 28, 2010
  • Threads: 115
  • Posts: 5692
September 25th, 2016 at 10:42:02 AM permalink
Quote: Paradigm


Getting this thread back on topic ;-), the above indicates that the frequency of the Ace High Pai Gow hand occurs for the dealer just under 10%. I think a less noticed mechanism is in the 6-8% range...players notice a bad event one in 10-11 hands. Once in 14 or so hands, while seeming a small difference, is important in "feel" of a mechanism.


One of the issues with a higher pai gow hand (Ace-high) is that its effectiveness doesn't correspond to its frequency; the dealer would not have lost all to the players to save the house here with a push, as some still would have pushed in any case. By contrast in Freebet/push-22 in BJ, ALL of the remaining players who would have won now push, so it is more efficient as a house edge mechanism.
Adding a clause "Dealer pushes with Ace-high or Queen-high" would add about 1.3%, for a HE of 2.9% or so, more in line with PGP's house edge, but also slightly more cumbersome, while having both a qualifier and a commission would drag the game down, and just be a poke in the eye to the players. Nothing slows the game down like calcing commissions THEN cutting checks.
Beware of all enterprises that require new clothes - Henry David Thoreau. Like Dealers' uniforms - Dan.
gordonm888
gordonm888
Joined: Feb 18, 2015
  • Threads: 21
  • Posts: 553
September 25th, 2016 at 11:39:01 AM permalink
Quote: Paradigm

Gordonm888, could I ask you to calc how often a player would benefit from pushing on the Ace High Dealer Pai Gow because otherwise they would lose?



Thank you for asking! Here is my calculation::

On 1.04% of hands, the dealer will have an Ace-High Pai Gow (which is defined to include either a natural ace or a joker as the high card), and the player would have a hand that would normally lose to the dealers hand. (A dealer will be dealt an ace-high pai-gow with a frequency of 9.79%, I slightly misquoted that number in an earlier post. )

Now that I've worked out my methodology, I guess it makes sense to calculate how often the player would have had a winning hand against an Ace-High Pai-Gow, only to have it turn into a push. I'll update you on that later.
Last edited by: gordonm888 on Sep 25, 2016
y2d2
y2d2
Joined: Jul 10, 2012
  • Threads: 2
  • Posts: 45
September 26th, 2016 at 6:27:08 AM permalink
Quote: Paradigm


I would have a tendency to just muck my cards if I couldn't beat the dealer...as I think about that, the game could incorporate that rule (e.g. the player shall muck their loser hands immediately) in an effort to speed up the game...that type of time and motion savings may provide some room in HE necessary to maintain profitability. If the dealer didn't have to reveal and check 50% of the player's hands because they were mucked as losers, that would be a positive for game speed.


That's how it's played at TVC, but some players are extemely novice and insistent on setting a losing hand for a dealer to check, even when others at the table are letting them know it's a loss and mucking early is ok.

Every once in a while you will have a player who is SO novice they will set their hand wrong even with dealer card knowledge. Turning pushes into losses, wins into pushes. Saw it the other night.
y2d2
y2d2
Joined: Jul 10, 2012
  • Threads: 2
  • Posts: 45
September 26th, 2016 at 6:39:15 AM permalink
From a player perspective, I absolutely love it vs regular or EZ PGP to suit my tastes. I hate the frustration of things like losing to copies/kickers, not being able to win with a strong hand like QQ/KKAxx because dealer has QJ/AAxxx or QJ/4422x. Loved the 78/AAJJx plays when dealer has KK/6655x.

If it's winnable/pushable, it WILL, instead of MIGHT.
y2d2
y2d2
Joined: Jul 10, 2012
  • Threads: 2
  • Posts: 45
September 26th, 2016 at 6:42:00 AM permalink
By the way, house way for two pairs smaller than 6's is to ALWAYS play two pair behind, regardless of low hand. For example, 73/55442.

66 or better get split, I THINK without ace in low hand. Could be wrong, perhaps things like 6655 gets played with K-high low hand? I forgot what the cutoff is for the pair size with hands like QQ66Axx or JJ66Axx. Pretty sure 66/QQAxx, but not positive about JJ

Besides looking up HE math, knowledge of the house way for dealer hand isn't really useful to the player, as the player gets to see the hand set prior to seeing their own hand.

A slight improvement to game speed might be to allow the players to see their own hand as soon as it's dealt, rather than waiting for the dealer hand to be revealed and set. I'm not sure if this rule is in place to assist in game security of having the dealer's eyes always available to monitor players when the floor or surveillance staff are not.

I'd bet a dealer-ace-high side bet with even a 10-15% house edge would be quite popular with players. I dunno, say, pay 8-to-1 or 7-to-1 on dealer ace high pai gow
ams288
ams288
Joined: Sep 26, 2012
  • Threads: 11
  • Posts: 3023
September 26th, 2016 at 7:13:39 AM permalink
Quote: y2d2

By the way, house way for two pairs smaller than 6's is to ALWAYS play two pair behind, regardless of low hand. For example, 73/55442.



I've always liked this house way. I've been saved many times by this rule. Most MGM properties in Vegas do this (except Luxor and Excalibur). Vegas CET properties split two little pairs when there is no ace present.
The geniuses who thought HRC was running a child sex ring out of the basement of a pizza shop just can't connect the dots on all this trump-Russia stuff....
Paradigm
Paradigm
Joined: Feb 24, 2011
  • Threads: 40
  • Posts: 2036
September 26th, 2016 at 7:22:52 AM permalink
Quote: gordonm888

Thank you for asking! Here is my calculation::

On 1.04% of hands, the dealer will have an Ace-High Pai Gow (which is defined to include either a natural ace or a joker as the high card), and the player would have a hand that would normally lose to the dealers hand. (A dealer will be dealt an ace-high pai-gow with a frequency of 9.79%, I slightly misquoted that number in an earlier post.


Thank You Gordon!! So we are looking at something around 8.75% of hands the player is going to see the dealer A High Pai Gow and not like the outcome. Getting close to 1 in 12 hands which isn't bad.
Attempting to add value one post at a time

  • Jump to: