Casino War

[sharkbyte]

Has anyone else looked into this game? It has piqued my interest.

I'm going to be in Vegas, the beginning of November. Does anyone know what casinos have tables? The casino, here, is MGM so maybe some of their other properties...

[220Inside]

Looks like you can find it pretty much everywhere on the strip

https://www.onlineunitedstatescasinos.c ... egas-9599/

[London Shooter]

Casino war makes baccarat seem highly exciting

[heavy]

Down and dirty - it's a no brainer. High card wins. 2.88% house edge. No particular strategy is better than another. As with most casino games, you can beat it over the short run on variance. Over the long run it'll eat your bankroll. Avoid the side bets. Avoid the tie bet. IF I were going to play War, which I won't be doing, I'd play a combination Paroli-Fibo strategy. I know. Sounds crazy. Up as you win. Up as you lose. But you go up at different rates and you're looking for different things. I'd go up on a win and continue to increase my bet until I lost. Then I'd regress to one unit. If I lost two in a row I'd go into fibo mode. On the third loss my next bet would be the sum of the last two and I'd be looking for two wins in a row before regressing back to on unit and starting over. Gets a little convoluted at times but it works for me.

[r_ventura_23]

My wife loves it. Tropicana in AC has a $5 game. We can play for hours and drink with very little swing in money.

[heavy]

There you go. It all depends on your priorities.

[r_ventura_23]

Happy wife happy life. Or so they say.

[sharkbyte]

Heavy - It appears just your basic up on a win, regress on a loss, holds okay. Surrendering the tie. But I am actually looking at the tie as its own bet/strategy. I don't yet have enough information about the game, for any certainty, but initial testing is positive. Going to reach out to another, on the board, to see if he is willing to check my work.

Testing sets are:
1 player, 70,000 hands, 6-deck shoe, rolling played cards back in after 124 cards played***
5 players, 20,000 hands, 6-deck shoe, rolling played cards back in after 124 cards played***

Initial Results - 9-bet sequence*, total of 19 units for a series, average win is 14.4 units:
1 player - 10.22% hit rate, +12.1% adv, +10606 units

5 players -
player 1 - 9.41% hit, +6.39% adv, +1477 units
player 2 - 10.51% hit, +15.41% adv, +3790 units
player 3 - 9.95% hit, +8.21% adv, +1968 units
player 4 - 10.21% hit, +16.9% adv, +4028 units
player 5 - 9.52% hit, +4.62% adv, +951 units

Observations:
- It's a grind. Getting to a higher unit value will have the most impact
- More players at the table does not appear to water-down the hit rate

*I have calculated out to a 30-bet sequence, all that are longer than 5 bets result in a positive outcome

*A 28-bet sequence is less than 200 units, average win is 14.9 units

***I noticed the dealer loading cards back into the auto-shuffler, at a little over a decks worth. I assume they're running a 6-deck shoe, but have never before seen a shuffler where you just add cards back in and they are mixed back into the deck. Anyone have more information?

***I think my 124 card number is too high, but I need to spend some more time watching, and getting a rough count for how many cards play before the dealer loads them back in.

[London Shooter]

Wizard of Odds shows tie bet on 6 decks has house edge of 18.64% There's no way you are getting such a big advantage over a bet that start so far into negative territory it's unreal.

[sharkbyte]

On an individual hand basis I would agree. But I'm looking at it to happen once within a 9-hand stretch.

Also why i want someone else to look at my results. i don't claim to be perfect so I may have missed something in my calculations.

[heavy]

It this were baccarat I'd say "see a tie - bet a tie."

[sharkbyte]

Never mind. I found a more accurate way to design the RNG, and reduced the card count to 75. Hit rate dropped to a high of 7.7% and all results went negative. Low hand count (10k), and I will confirm with more...to be thorough, but I don't see a reason to think there will be any improvement.

Heavy's up on a win, regress on a loss, still holds fairly steady...so a way to pass the time at least. I also experimented a little with 3 losses (at minimum), then a 4-step martingale. Ties are a wash on the hand count. It has its bright spots, but 8+ losses happen often enough to knock it down close to even, so why bother with risking the money...

thanks for everyone's input.