2 deck poker, has it been done?

I mean that is less ridiculous than every single one of the 7893 games that other guy has proposed

I haven't done the math but does a full house still beat a flush? Does 5-of-a-kind beat a straight flush?

That second one is interesting at least. In old books I have, they indicate that five of a kind breasts a straight flush but loses to a royal when considering games with wild cards. I can't think that simply duplicating every card affects the maths to any significant degree

Yet going the opposite direction and removing cards does such as in Short Deck where a Flush beats a Full House.

Well there you are doing things disproportionately compared to the status quo.

Since in a normal deck without wild cards 5 of a kind is impossible and it can occur in your game clearly the odds are effected to a significant degree on certain hands.

The other even more fun variant was to have 2 different backs on the cards, say red and blue.

So you can now see if your opponent has 2 been delt 2 cards from the same deck or not, and the same with your own hand showing.

Slightly deeper, you can be looking at the backs of the community cards to see which deck they are from. You may also want to keep track which backs were folded and even the burn cards.

Again could be an interesting change.

That would be a really bad game. A better Holdem game would be adding an ante from all players. Or Short Deck. Or PLO for a more options poker variant.

all in or fold on flop and run it 3x with 2 decks holdem or omaha.

The whole reason one hand beats another is because they have a lesser chance of happening. In Short Deck Hold'Em where cards are removed from the deck, the odds of getting a Flush are lower than the odds of getting a Full House so a Flush beats a Full House in that game. You have to calculate the odds of getting each hand to determine hand ranking or you'll end up in a situation where Two Pair beats a Straight.

would the 2nd deck affect the hands. 5 of a kind is still harder to get than a straight flush so it would be added to the correct place.

Why would anyone do that? It would just make the game less exciting. For example, you start with pocket aces, you know there can only be 2 other aces in the deck. But if there are 6 other aces, then what?

As a practical situation suppose we have AJ facing a pot sized river bet on a rainbow board of AQ932. We give our opponent a range of JJ+, AQ+.

We beat KK and JJ, 9 combos. We lose to AK and AQ (8 each) plus AA (1) and QQ(3) for 20 total. We win 9/29=31%. We need 33% equity to call the pot sized bet so we fold

With two decks same spot, there are 28 KK and 21 JJ combos so we beat 49. There are 15 AA, 21QQ, and 48 each of AK and AQ. Total losers are 132 combos. Our equity is 49/132=37.1% so we can call the pot sized bet.

This is just one spot, but it looks like adding another deck could change the equity calculations enough to generate more action.

I know. But there's also the added time you get dealt the exact same card, say 7h 7h which is suited pair. Still the same changes of getting a set but now we can get involved in flush stuff.

Also you may have 7h 6s but the flop comes 7h 8h Qh or something. More interesting spots

I did the math on 2 deck hand probabilities. The order of hand ranks is not affected. I am not sure why people insist on calling a royal flush a different hand than a straight flush, but if you must do so, then a royal flush should beat five of a kind. If you more logically regard a royal flush the same way as an ace high straight - just the highest possible hand of its class - then five of a kind beats any straight flush. Again, by this logic, we could say an ace high straight beats a full house in standard poker 1020 ace high straight combos (4 possible aces, kings, queens, jacks and tens for 1024 combos minus 4 combos that make royal flushes) vs 3744 full house combos (13 ranks for trips, 4 combos per, then 12 ranks for the pair, 6 combos per for a total of 13*4*12*6).

Anyhow relative to standard, all hands other than nothing, straights and straight flushes (including royals) are more probable in 2 deck poker. In some cases fairly significantly so. Trivially, 5 of a kind is more probable since it has zero probability in standard poker. Four of a kind is almost five times more likely; full houses about twice as likely.

Hand 1deck. 2 decks
5K. 0. 0.0000079
RF. 0.0000015. 0.0000014
SF. 0.000014. 0.000013
4K. 0.00024. 0.00094
FH. 0.0014. 0.0027
Fl. 0.0020. 0.0028
St. 0.0039. 0.0035
3K. 0.021. 0.033
2P. 0.048. 0.058
1P. 0.423. 0.445
HC. 0.501. 0.453

All numbers are fractional probabilities. If numbers donÂ’t total 1, itÂ’s either due to rounding or the fact that I entered these manually and made a typo.

It just occurred to me that there is an analogous situation that presents itself in real-world casinos. Most casinos have three card poker as a table game. Almost invariably, three card poker tables have a “pair plus” bet where the player bets that his hand will be a pair or greater, with a pay table depending on the actual hand - 1:1 for pair, 3:1 for a flush, etc.

In a few casinos I have seen a similar side bet on blackjack tables. The player bets that his starting two cards plus the dealerÂ’s up card will make a three card poker of a pair or greater. In all the places IÂ’ve seen this the pay tables AFAIK are identical.

I never really questioned this since I just figured they were identical bets. But after reading this thread I realized they are not identical. Blackjack tables are dealt from a multi-deck shoe, usually 6 or 8 decks. Three card poker tables use a single deck. IÂ’m sure the house edge for the blackjack side bet is still positive (or the casino wouldnÂ’t offer it), but based on this thread I would think itÂ’s at least a slightly less -EV bet than the pair plus bet on the three card poker table.