Taking Pai Gow to next level
June 28, 2016 3:00 AM
by Elliot Frome
It is an interesting story that the fortune cookie, considered to be the quintessential dessert for a Chinese meal, is actually an American creation. The same is true of a very popular casino game called Pai Gow Poker.
Pai Gow Poker is a twist on the game of Pai Gow Tiles (which I believe is Chinese). In Pai Gow Tiles, the player and dealer each receive four domino-like tiles and have to split them into two two-tile hands. That’s about where the similarity ends.
Pai Gow Poker was created to attract American audiences because it uses cards and standard poker ranks. It also uses a semi-wild Joker, which can be used as an Ace or to complete a Straight or a Flush. The player and dealer each get seven cards and they must be split into a 5-card hand and a 2-card hand. The only requirement is the 5-card hand must outrank the 2-card hand.
If the player beats the dealer on both hands, he wins. If he loses both, he loses. If they split, it’s a push. The edge for the house is the dealer wins copies (identical hands), and the player pays a 5% commission on winning hands (he is paid 19 to 20 for winners).
Pushes will occur about 42% of the time. This might be considered one of the drawbacks of the game. You can play a significant number of hands and never have any money change hands. The second drawback is the commission. Players don’t like it and it can slow up the game, calculating that 5%. So, someone came up with the idea to make a few changes to Pai Gow Poker and Asia Poker was born (most likely, not in Asia!).
Asia Poker uses the same 53-card deck as Pai Gow Poker. The significant difference is the player and dealer split their cards into three hands – a 4-card hand, a 2-card hand and a 1-card hand. This has a profound impact on the game. First of all, ties are completely gone. The player can win 0, 1, 2 or 3 hands. If he wins 2 or 3, he wins even money. If he wins 0 or 1, he loses his wager. Because a 1-card hand is involved, the number of copies skyrockets and there is no need for a commission. Both of the issues of Pai Gow Poker were resolved.
One of the key components of Asia Poker (and Pai Gow Poker) is the House Way. These are the rules the dealer must set his hand with. They have to be well thought out. Many years ago, I helped Ya Awada develop a game called Mini Pai Gow where the player was dealt six cards and split into a 5-card hand and a 1-card hand.
Because of the high frequency of copies of the 1-card hand, we actually made the House Way overly simplistic and clearly not optimal. Essentially the house advantage with a strong House Way was too high, so we had to “dummy down” the House Way to give back some advantage to the player.
In Pai Gow Poker, the House Way is generally considered to be near perfect. There is no such thing as absolutely perfect as there is always some question on very close hands. Further, it could be a case where if the dealer does A, the player should do B and if the dealer does B, the player should do A. In a game with a 5-card and 2-card hand, these situations are greatly reduced and it is generally regarded that if the player plays the House Way in Pai Gow Poker, he is playing very close to optimal.
With a 3-hand game, I was always less sure about any particular House Way. I had helped develop a game called San Lo several years ago that also dealt six cards and split into three hands (3-card, 2-card and 1-card). In doing that, I learned with a 3-hand game, the House Way tries to strike a balance between the hands. But, the player has the option (to some extent) of sacrificing one hand in order to strengthen the other two. The player’s goal is to win two out of three. Winning all three provides no additional bonus. So sacrificing one hand for the sake of the other two, in essence gives the player some advantage.
I’ve wanted to test my theory out on Asia Poker for the longest time. Thanks to some help from another analyst, David Drapeau, we were finally able to spend the time analyzing the game. In our analysis, we throw out the notion the player should play the House Way all the time and instead look at every possible way he can play his hand (legally) and determine the best way to set his hand.
This process found that about 10% of the time, the player should not set his hand the House Way. By doing this, he can raise the payback of the game about 1.25%-1.5% or about 96.5%. Put this another way, he can cut the house advantage by 25%-33%. We suspect it is a combination of cases of the player sacrificing one hand to increase the value of the other two and areas where the House Way may not have necessarily picked the strongest possible play for the dealer, leaving the player with some opportunities.
In the meantime, know the House Way is good 90% of the time, but take a look at what you wind up with and make sure it makes sense to you as well.
Buy his book Expert Strategy for Three Card Poker now!
Elliot Frome is a second generation gaming analyst and author. His math credits include Ultimate Texas Hold’em, Mississippi Stud, House Money and many other games. His website is www.gambatria.com Email: ElliotFrome@GamingToday.com