Welcome To Pai Gow Tiles, Fellow Chinese... Confucius Say: You Cursed For Life

[X] Take 10 hours to learn tile rankings, and JB's advanced strategy with exceptions
[X] Drive to Foxwoods Resort Casino, buy in for $2,000
[X] My very first hand dealt of Pai Gow Tiles: H12 H12 H6 H6
[X] +$23.75

And then...

1/2, 0/9, 2/3, 4/4, 3/3... it never stops.

100 decisions (didn't keep track of pushes), 19 wins. That's right, 81 losses. That's a Z-score of (19 - 49.737)/(4.9999), or -6.15. Negative 6.15 standard deviations away from the mean.

I even banked a few hands too. But let's not push the knife in any further.

Assume that an additional 70 hands are pushes, I only won 11.1% of my hands. LOL. (Wizard says 28.33% is expected.)

[X] -$1,573.75

I couldn't even find a Z-score table to look up the p-value. Anyone with a calculator? I'd imagine it's beyond 1 in 100,000.

All playing decisions based on JB's advanced strategy with exceptions.

Here's a question for Wizard, JB, or any PaiGow fan...

By setting tiles completely in the dark (without looking), is the expected number of wins greater than 19 in 100 decisions? In other words, could a chimpanzee have done better than me while flinging poop at the same time?

I welcome others to comment on atrociously bad runs they have had, in blackjack, craps, PGT, PGP, etc.

Play at MSun the Pork Belly and duck is better.

Suffering sessions of horrible luck has to be the worst thing about gambling, my sympathies.

although the mystery of Pai Gow subsided a bit for me by learning Pai Gow Poker, learning tiles an unfinished personal project.

Reminds me of the last time I played PGP - I played for almost 2 hrs on a 300 buy-in, and only won 3 times. The full table, combined, only won about 10 hands in the same amount of time. Every time you would get 2 pair, the dealer would pull a full house or 3 pair! I managed to cut my losses and only lose $180 of the 300, but it was the worst run I'd ever seen at Pai Gow Poker.

I intend to go to AC in January to tempt the tile Gods, hoping that the $400 I won last time doesn't turn into a monster losing session. We shall see.

-B

Imagine how good its going to feel when it stops!

I suggest playing for $20 or $40 per hand instead of $25. Then you don't have to deal with quarters in your commission.

Holy $&8^ OMFG, I just looked up -6.15 standard deviations on Wikipedia...

(Keep in mind, this is only for -6 sigma... I did a little bit worse, plus I was banking too! WTF!)

Fraction outside confidence interval: 1/506,800,000

Given that we are only dealing with the left tail of the bell curve, the chances of of winning 19 hands (or fewer) out of 100 decisions is...

1 in 253,400,000.

For comparison, the odds of winning the Mega Millions is roughly 1 in 175,000,000, and winning the PowerBall, roughly 1 in 195,000,000.

Man, I need to file a complaint with Foxwoods, or give up the game. FML.

You can get that probability in Excel with the function normsdist(-6.15). It comes to 1 in 2,581,213,130. Ouch!

I never looked into the odds of playing blind. I doubt JB has either. Interesting question though. I'd have to do a random simulation to answer it, and just don't have the time. DorothyGale might be able to answer it, should she stumble upon this thread.

Quote: Acoplander1

By setting tiles completely in the dark (without looking), is the expected number of wins greater than 19 in 100 decisions? In other words, could a chimpanzee have done better than me while flinging poop at the same time?

I welcome others to comment on atrociously bad runs they have had, in blackjack, craps, PGT, PGP, etc.

As someone who has recently been the target of poop-throwing chimpanzees, I can help you out here.

If you select one tile from the group of four, then randomly select a second tile to play with it, there's a 1 out of 3 chance that that is, in fact, the optimal way to set the hand (since the #3 and #4 tiles would then be automatically selected as well). So let's say you had a 49% chance of winning against the house, playing normally. Now let's first make the assumption that every time you set your hand the wrong way, you lost. This would drop your winning perecntage down to (49/3), or 16.3%. So far, the chimp is doing a little worse. But we can't assume that setting the tiles "wrong" will always lose; for one thing, two of the remaining three tiles could be identical (and house way dictates splitting the pair). So we have to assign a winning probability for the random split that results in an error that is greater than zero--say, 25%. Which makes the chimp beating the stuffing out of you.

Next time, bring the chimp.