Pai gow poker strategy

Has anyone here read Stanford Wong's OPTIMAL STRATEGY FOR PAI GOW POKER
There is a difference between his two-pair strategy and the strategy put forth by the Wizard.
In some extremely low two pair combinations Wong will play a jack high front hand!
I am not sure of the math need to analyze the two strategies but I am curious which strategy the pros would prefer.

There is a modified version of Wong's strategy here: http://www.edcollins.com/paigow/
And the two-pair strategy of the Wizard is here: https://wizardofodds.com/games/pai-gow-poker/appendix/2/

Basically, the modified Wong strategy says that with two pair totaling 7 or less you can keep them together with a queen high front hand.
With two pair totaling 8 thru 11, keep them together with a king high front,
With two pair totaling 12 thru 16, keep them together with an ace high front,
With any two pair totaling 17 or more split the pairs.

The Wizard strategy says with two pair totaling 9 or less, keep together with a king high front
With two pair totaling 10-15 keep together with an ace high front,
Split two pair totaling 16 or more.

I have been playing the modified Wong strategy and I have only had one hand when the two hand strategy went against me,
But I would have lost the hand either way...
(I played a pair of fours and threes in the back with a QT in front. the dealer had paired sevens and twos in back and QQ on top)

Please let me know what you guys think of the two pair strategy

you can go here and add up the "power ratings" of the hands.

I haven't done it, but that'll show you the difference in value between the systems.

It's always interesting to see my website quoted in forums. :)

As I hope I made clear in my article, my strategy is meant to be "easy" to remember. It is not perfect. It is not exact.

Wong's strategy can be difficult to remember for most people. To use it perfectly would require memorizing a lot of different hands/situations. To some, they can do that, and enjoy doing that.

But my website article and "method" helps to make Wong's criteria a bit easier to remember and implement for the average Joe who enjoys the game. (At least that was the goal!)

The drawback, is, of course, a cost of some accuracy. Exactly how much accuracy we are talking about would probably require a computer simulation, simulating and comparing both strategies over a few hundred thousand hands. My gut feeling the difference is very, very low. Too low to be worth mentioning. You won't play enough hands in your lifetime to see a difference. And you are probably aware, playing "incorrectly" can often turn out to be a winning play for that particular hand.

For some paigow poker hands, I'm guessing the difference in how you play is miniscule. Sort of similar to hitting or standing on a 16 against a dealer's 10 in blackjack. Mathematically, hitting (as I recall) is correct, but the difference is so small, if you stood instead, it wouldn't make much of a difference. Again, you won't play enough hands to see any difference.

Ed Collins

Mr. Collins, I really appreciate the work you did on making Wong's strategy more accessible to my faltering memory circuits! Thank you!

I realize that even playing two hands at once for 2 four-hour sessions a week, I only play about 14,000 hands a year.
Considering there are over 35 million two-pair hands, I know that I am unlikely to come up against questionable play too frequently!

I would also like to tip my cap to Ed Collins for a job well done.
But I do have one question... is 55/44 really a Q-3 or better, or is it a typo.

I have already answered my own interpretation in the "Table Games" header.

Basically
1.) if the High Pair is KK or AA always split
2.) if the Low Pair is 10-10 or better always split
3.) if the High Pair is 77 to QQ A-3 or better plays to front, otherwise Split
4,) if the High Pair is 66 or less K-3 or better plays to front, otherwise Split

In theory: QQ or JJ high pair plays Ace-Face to front else split *if you are Banker*.

Thanks in advance.

Quote: 98Clubs

I would also like to tip my cap to Ed Collins for a job well done.
But I do have one question... is 55/44 really a Q-3 or better, or is it a typo.

Thanks in advance.

Oh my gosh. I just double-checked Wong's book. That IS a typo. Wong DOES show 55/44 requires a K-3, not a Q-3.

I'm surprised no one has discovered this typo and mentioned it until now. Fortunately, my modified strategy does say a King is needed. (If the total is 8 thru 11, a king is needed in your two-card hand if you wish to keep the two pair together.)

Thanks for letting me know. I've already fixed my page.

Ed

Quote: 98Clubs

IIn theory: QQ or JJ high pair plays Ace-Face to front else split *if you are Banker*.

Yes! In fact, that ace-face would be an even-stronger AK with QQJJ, as any other other ace-face you'd have a full house. This, along with frequent copying as banker, forces the rule to work VERY well.
A good basic rule is "split no two pairs with AK top"

Have you guys seen the Pai Gow poker Strategy calculator here?

My estimations (which use a simple program I wrote years ago but only estimate flushes in the interest of time) are
(5s 4s QJ split two-pair)
4C 4D QC JC 8C 5D 5S 82.687882 25.248232 53.446126
QC JC 8C 5D 5S 4C 4D 24.302737 75.540751 49.462782
(5s 4s K32 play K3)
KC 3C 5S 5D 4D 4C 2C 30.926870 75.183029 52.473656
KC 2C 5S 5D 4D 4C 3C 30.820194 75.183029 52.422323
4C 4D KC 5S 5D 3C 2C 79.161904 20.975063 49.653377
(4s 3s QJ split two-pair)
3S 3C QC JC 5D 4C 4D 79.688272 20.516044 49.693436
QC JC 5D 4D 4C 3C 3S 24.110640 74.523759 48.867996
(3s 2s Q85 play Q8 not Q7 - the link above flips at QT-Q9 )
QC JC 5D 3D 3C 2C 2S 23.950662 73.852046 48.459153
2S 2C QC JC 5D 3C 3D 76.798955 16.072168 46.126980
QC TC 5D 3D 3C 2C 2S 20.619579 74.436212 47.144185
2S 2C QC TC 5D 3C 3D 76.452167 16.151327 45.993046
QC 9C 5D 3D 3C 2C 2S 18.984659 74.403507 46.340951
2S 2C QC 9C 5D 3C 3D 76.452550 16.148044 45.991657
QC 8C 5D 3D 3C 2C 2S 18.248857 74.423599 45.996692
2S 2C QC 8C 5D 3C 3D 76.454318 16.153631 45.995220
2S 2C QC 7C 5D 3C 3D 76.446015 16.136646 45.982935
QC 7C 5D 3D 3C 2C 2S 17.923208 74.359824 45.808324
So it does look as if consider playing good Q-high with 3s/2s 4s/2s.

Obviously one can never play K2 (since the cards left are K32) and it takes one of the Kings away making it slightly less likely the dealer makes a King-high.

Yes I have... it seems to be oriented more towards the Vegas rules than the E. Coast (Trump) rules. Useful and informative.

@paigowdan: Whats your take on 77-high two pairs... Ace or King up front before split? Seems rather close either way. My own rules sided with Ace = no split.

Quote: 98Clubs

Yes I have... it seems to be oriented more towards the Vegas rules than the E. Coast (Trump) rules. Useful and informative.

@paigowdan: Whats your take on 77-high two pairs... Ace or King up front before split? Seems rather close either way. My own rules sided with Ace = no split.

7's and 2's or 3's together with a K-x top.
7's and 4's or better: KQ or KJ top, especially when banking.
To push or beat an opponent's KQ/AAxxx hand is especially satisfying...
two pairs with a King up is not a bad hand....

Thanks for that... seemed like a 'tweener.