Caribbean Stud Poker - FAQ

Thanks for the compliment. You are off by a factor of five because the dealer can have any one of 5 cards face up. In other words order does matter with the dealer's hand, since the first card is dealt face up. The correct derivation of the total combinations is combin(52,5)*47*combin(46,4) = 19,933,230,517,200.

If you play long enough, the only results you will see is that you will lose all your money. Don't bring to the table more money than you care to lose in that sitting.

Second, Caribbean Stud Poker has a house edge of 5.22%, so if you're trying to stay in the game as long as possible, I would avoid that one. The best table games are craps and blackjack, when played properly.

According to my probabilities in poker there are 1,296,420 ways out of 2,598,960 to form a pair or more. I also indicate at the bottom of that page that there are 167,280 ways to form an ace/king. So there are 1,463,700 ways to qualify, or a 56.32% chance.

By playing blind you are bucking a house edge of 16.607%. If you used my three rules of thumb in my section on Caribbean stud you would lower the house edge to 5.225%.

Good question. I checked out four casinos using Microgaming, Starnet, Cryptologic, and BossMedia software. Starnet uses the conventional rules. Cryptologic and BossMedia each pay 200 to 1 on a royal flush as opposed to 100 to 1. Microgaming has the following paytable.

Note that Microgaming pays even money only on a two pair, but is more generous on everything higher. The following table displays the house edge for each kind of software assuming optimal strategy. Note that Starnet calls the game Cyberstud Poker and the rest call it Caribbean Poker.

To answer your first question, yes, the house advantage can be reduced by sneaking peaks at other cards. For example if you have an ace/king hand it should make you more inclined to raise if you see another player matching the dealer’s up card. In the book Finding the Edge Peter Griffin and John M. Gwynn Jr. address the question of player collusion in Caribbean stud poker. Assuming perfect knowledge of all other cards, and having perfect knowledge of how this information affects the odds, their paper states the player would have a 2.3% edge in a seven-player game. In a six-player game, the house would have an edge of 0.4%.

Somebody else asked this is a past column. The book Finding the Edge presents a paper titled 'An Analysis of Caribbean Stud Poker' by Peter Griffin and John Gwynn Jr. There they state that if seven players colluded perfectly they would enjoy a 2.3% player advantage. However, they don't state what the edge would be in a five-player game. I suspect that the odds would swing back to the house.

You, are right that on average a royal flush will occur once in every 649,740 hands, and that in 1,299,480 hands the expected number of royal flushes is 2. However, this is only the average. With every hand that goes by you are no closer to getting a royal. Every game of independent trials has this memory-less property so a royal flush is never overdue.

The probability of zero royals in 1,299,480 hands is 13.53%.

In games with a maximum pay out you should never bet so much that the maximum winning would be affected by the maximum. For example if the maximum pay out is $2000 and the biggest win pays 100:1 then you shouldn’t bet more than $20. Assuming the royal flush there pays 100-1, you shouldn't bet more than $30. As long as you stay under these limits the odds have not changed. The version where you can switch a card is called Oasis Poker.

According to table 3, a four of a kind pays $500, a full house pays $100, and a flush pays $50. If m is the amount of the jackpot meter then the return per dollar bet is (1121800+4*j)/2598960. The meter would need to reach $369,290 for this to be a positive expectation bet.

You correctly calculated the number of player combinations as combin(52,5)=2,598,960. From there, the dealer can have combin(47,5)=1,533,939 possible hands. Then any one of five dealer cards can be face up. So 2,598,960*1,533,959*5=19,933,230,517,200.

You’re not the first to ask me about this. I’m afraid I haven’t worked out the odds for this variation. If this twist ever makes it to Vegas I’ll make it a higher priority.

p.s. (Feb 21, 2006) I now address this rule variation in my Caribbean Stud Poker section.

Yes, knowledge of other player cards can help if you use the information correctly. I haven’t studied this in depth but what you are already doing is a good idea. When you have an ace/king you don’t want the dealer to form a pair. If you or your wife can match the dealer’s up card that lowers the probability of the dealer forming a pair, and thus increases the probability that the dealer won’t qualify. However if you’re willing to fight to cut down the house edge marginally I wouldn’t waste your time and money on Caribbean Stud Poker but rather on a lower house edge game like blackjack or video poker.

No. If you’re interested in comparing expected loses it would be better to use the house edge. My section on the house edge shows the blackjack house edge to be 0.43% (Atlantic City rules) and that of Caribbean Stud Poker to be 5.22%. The expected loss for 1 hand of Caribbean Stud Poker at a $5 ante is $5 * 5.22% = 26.10 cents. The expected loss for 6.7 hands of blackjack at $15 per initial bet is 6.7 * $15 * 0.43% = 43.22 cents. So given these two options you will lose less in Caribbean Stud Poker. The ratio of the house edge of Caribbean Stud Poker to blackjack is about 12. So the expected loss of a $1 initial Caribbean Stud Poker bet is about the same as a $12 initial blackjack bet.

No. This is a hard and fast rule, you never raise in Caribbean Stud Poker with less than ace/king.

No

Thanks for the compliment. Trust me, you should raise on any pair, even a pair of deuces. It isn’t just that you hope the dealer won’t qualify, but you also win the ante and raise if the dealer gets an ace/king. You will still have a negative expectation on a low pair, but the expected loss by folding is even more.

You are referring to Caribbean Draw Poker. This one would be a very difficult game to analyze. At this time I have no plans to analyze it but once it reaches a certain saturation point it will be hard to keep putting it off.

The math is quite easy. The probability of a royal flush is 1 in 649740. So the expense of reseeding the jackpot is $10,000*(1/649740) = 1.54%. For every dollar bet you keep 40% for profit and reseeding the jackpot. 40%-1.54% = 38.46% profit/house edge. It does not make any difference what you pay on the smaller jackpot or if there is a maximum win. Ultimately the 60% that goes to the meter goes to the players one way or another, it doesn’t matter to you how it gets divided up.

I believe what happens in this situation is the player to the dealer’s right would win the progressive jackpot and the other one would win only $10,000. This is because the dealer pays players from right to left so player to the right would be paid first, the meter reset to $10,000, and then the second player paid. However I think the second player would have a legitimate complaint on his hands. The probability of this happening in a full table is 1 in 20,103,110,301. So I would doubt this has ever happened or ever will happen.

As a general rule, you can make any bet for the dealer in any game. In general you should tell the dealer which bets are his, except blackjack where its common practice that any bet outside the betting circle is for the dealer.

Blackjack: Yes to all three. The usual way to bet for the dealer in blackjack is to put the tip on the edge of the betting circle. If you split or double most people also split or double the dealer’s bet, although it is not required.

Caribbean Stud Poker: I asked a dealer and he said raising for the dealer is optional. I haven't studied it but I think this would result in the tip having an advantage.

Let it Ride: I'm told that the player should put out three tips initially but must pull them back in the same manner that they pull back their own bets. Bets that are pulled back go to the player, not the dealer.

Craps: Yes, you can make any bet for the dealer. The most common ones are the yo-11 and the hard ways. If you make a line bet for the dealers and back it up with the odds it is implied the odds are a tip too.

Roulette: As in craps you can make any bet for the dealer. Just tell them in advance.

You should try to sit as far to the left as possible, assuming you play the side bet. If you don’t make the side bet it doesn’t matter. The reason is if two more players get a straight flush or higher the player furthest to the left will get the benefit of the full meter, because the dealer pays players from her right to left. Subsequent players will get less after paying the first player. In the event of two royal flushes the first player would get the full meter and the second only $10,000, which is what the meter is generally reset to. However the odds of this are extremely unlikely. I would just play where you have the most elbow-room and are furthest away from any smoking players. Finally, yes, there are generally six spots.

Coincidentally I heard of a Vegas casino doing the same thing because they wanted to take out their Caribbean Stud game. Here is a general formula for calculating the expected return when a straight flush pays the full jackpot.(((5108*FL+3744*FH+624*FK+40*J)/2598960)-M*0.052243-1)/(M+1)where

FL = Flush win
FH = Full house win
FK = Four of a kind wi
n J = Jackpot amount
M = Minimum ante bet

In your case we have (((5108*150+3744*300+624*1500+40*155000)/2598960)-5*0.052243-1)/(5+1) = 36.858%. So the player advantage is 36.858% of the combined ante plus $1 side bet, or an expected profit of $2.21 per hand.

I strongly believe the makers of the shuffling machines at least attempt to make the shufflers as fair and random as possible. A deliberately gaffed machine I’m sure would violate Nevada law. It is fairly easy to see good x-card hands in x+1 cards. For example the probability of a three of a kind in three cards is 0.235%, and in four cards 0.922%, or almost four times higher.

Thanks for playing it. Yes, Ties Win Blackjack was a good choice for this promotion. The probability of a full win is 43.314%, a half win is 8.75%, and a loss is 47.936%. So the probability of any win is 52.064%. The probability of five consecutive wins is 0.52064⁵ = 3.825%. Flat betting this results in an extra 3.825% of return for the player. The house edge normally is 0.247%, so the player advantage under this promotion would be 3.5785%. However I find no mention of this promotion on the casino web site and given my usual 2-3 week delay to answer e-mail it is probably over.

For the benefit of my readers, table 3 pays $500 for a four of a kind, $100 for a full house, and $50 for a flush. The breakeven meter on the Northern Indiana side bet pay table is $369,290.00

Thanks for the kind words. Yes, I gave Casino Life permission to use my material. I’m happy they gave me a good plug. No, they didn’t pay me. I also have never been paid for my appearances on the Travel Channel here in the United States. I do it for the fun and the publicity.

Thanks. To answer your question, no, I don’t think it is a good game. The house edge is too high. If you are looking for a big win you could play a progressive betting system in blackjack, pressing your bets as you win. Of course this comes at the cost of frequent smaller losses.

Those ideas sound like flagrant cheating. However, putting ethical issues aside, I would find a game of Caribbean Stud Poker with a high progressive jackpot. Then I would arrange the cards to give myself a royal flush and the dealer a qualifying hand. After resuming time make the $1 side bet and the maximum ante bet.

Please see the following table. This table also shows the house edge assumed for player rating purposes. My source is an executive with a major Strip casino here in Vegas, who wishes to remain anonymous.

Hands per Hour and Average House Edge

Games	Hands/Hour	House Edge
Baccarat	72	1.2%
Blackjack	70	0.75%
Big Six	10	15.53%
Craps	48	1.58%
Car. Stud	50	1.46%
Let It Ride	52	2.4%
Mini-Baccarat	72	1.2%
Midi-Baccarat	72	1.2%
Pai Gow	30	1.65%
Pai Pow Poker	34	1.96%
Roulette	38	5.26%
Single 0 Roulette	35	2.59%
Casino War	65	2.87%
Spanish 21	75	2.2%
Sic Bo	45	8%
3 Way Action	70	2.2%

I don’t think that is fair, but I have heard of other casinos following that rule. The way I think the rule should be followed, and the way most casinos do it, is one player receives the full 10% of the jackpot, and the other 10% of whatever they reseed the jackpot to. Most casinos reseed the jackpot at $10,000. This begs the question, which player should get 10% of the full jackpot? Some argue the player to the dealer’s right would get it, because his hand is adjudicated first. Others argue the player to the dealer’s left, because he received his hand first. I have received passionate e-mails both ways. I tend to side with those who say the player to the dealer’s left would be paid first. It may depend on what the table games manager decides to do at the particular casino.

Thanks, I stand corrected. I thought I heard somewhere that the first hand to the right would get paid first, because the dealer pays from the right. However your rationale makes sense too.

Thanks. As I say about machine jackpots, 0.5% to 1% of the jackpot amount after taxes is good. Whether the dealers pool their tips or not should not make a difference.

The rate of return is 34.53%, plus 3.08% for every 100,000 Kronor in the jackpot. The breakeven meter is 2,126,825 Kronor.

For the side bet to have zero house advantage, the jackpot would need to reach $578,842.11.