Quote:
Originally Posted by WheelDraw1020
If there's a Baccarat dealer who never gives the banker a card if leading after the player's 3rd and always hits the banker if behind after the player's 3rd but follows all other rules correctly how much is a Banker bet worth?
|
I did a simulation of 10 million hands, and I got that a bet on the banker would have a positive EV with an edge of about +0.9%. That should be accurate to within 0.1% to about 99.9% confidence. That's for an 8 deck game with 5% commission. You didn't explicitly specify what the dealer does when the player takes a 3rd card and he is tied with the player, so I assumed that he follows the normal rules in that case based on his total and the player's 3rd card.
The R simulation is below. I first checked it without the dealer error and got the right value, which is about -1.06%, to within the accuracy of the sim. The dealer error lines are marked in the comments.
Code:
sims = 10000000
decks = 8
cards = c(rep(1:9,4*decks), rep(0,16*decks))
hit = c(1,1,1,1,0,0,0,0,
1,1,1,1,0,0,0,0,
1,1,1,1,1,0,0,0,
1,1,1,1,1,0,0,0,
1,1,1,1,1,1,0,0,
1,1,1,1,1,1,0,0,
1,1,1,1,1,1,1,0,
1,1,1,1,1,1,1,0,
1,1,1,0,0,0,0,0,
1,1,1,1,0,0,0,0)
dim(hit) = c(8,10)
wins = 0
losses = 0
for (i in 1:sims) {
deal = sample(cards, 6, F)
player = (deal[1] + deal[2])%%10
banker = (deal[3] + deal[4])%%10
if ( (player < 8) & (banker < 8) ){
if (player <= 5) {
player = (player + deal[5])%%10
if (banker < player) { #Dealer error
banker = (banker + deal[6])%%10 #Dealer error
}
else {
if (banker == player) { #Dealer error - should always do next line
if (hit[banker+1,deal[5]+1] == 1) {banker = (banker + deal[6])%%10}
}
}
}
else {
if (banker <= 5) {banker = (banker + deal[5])%%10}
}
}
if (banker > player) {wins = wins + 1}
if (player > banker) {losses = losses + 1}
}
(wins*19/20 - losses)/sims