Thread Rating:
(Banker or player wins one more hand than the other over the shoe)
In my experience I have found it to be much higher (approx 4-5 on average)
Does anyone have any thoughts on this?
Here is the results of the simulation:
Quote: mustangsally
******
how about this interesting tidbit of info.
I sampled 10 million 8 deck Bacc shoes and have this table of data
Banker wins minus (-) Player wins per shoe
stats show on average 0.93 more Banker wins
looks like the top 2 are 0 and -1
0 means equal Banker and Player wins (difference of 0)
-1 means 1 more Player win than Banker win
(no method of bet selection)
difference count count/10million 0 470876 0.0470876 -1 464159 0.0464159 3 461944 0.0461944 1 452870 0.045287 2 440194 0.0440194 5 432053 0.0432053 -3 421225 0.0421225 4 408527 0.0408527 -2 407685 0.0407685 6 392305 0.0392305 -5 385988 0.0385988 -4 381998 0.0381998 7 357713 0.0357713 8 340837 0.0340837 -6 328947 0.0328947 -7 298758 0.0298758 9 291604 0.0291604 10 272797 0.0272797 -8 258235 0.0258235 11 239853 0.0239853 -9 235278 0.0235278 12 220101 0.0220101 -10 213561 0.0213561 -11 182223 0.0182223 13 180280 0.018028 -12 152011 0.0152011 14 151358 0.0151358 15 132195 0.0132195 -13 129223 0.0129223 -14 102517 0.0102517 16 100993 0.0100993 17 83963 0.0083963 -15 77411 0.0077411 -16 65432 0.0065432 18 55442 0.0055442 -17 52227 0.0052227 19 48445 0.0048445 -18 44470 0.004447 20 36014 0.0036014 -19 31716 0.0031716 -20 31271 0.0031271 21 24844 0.0024844 22 24738 0.0024738 -21 17794 0.0017794 23 17776 0.0017776 -22 12020 0.001202 25 10435 0.0010435 -23 10426 0.0010426 24 9703 0.0009703 -24 5812 0.0005812 -25 4653 0.0004653 -26 4636 0.0004636 26 4274 0.0004274 -27 3859 0.0003859 29 2697 0.0002697 28 2325 0.0002325 -29 1927 0.0001927 30 1925 0.0001925 -28 1167 0.0001167 32 1160 0.000116 31 1153 0.0001153 -30 773 0.0000773 -31 770 0.000077 27 411 0.0000411 33 5 0.0000005 -32 4 0.0000004 34 4 0.0000004 -34 2 0.0000002 35 2 0.0000002 37 2 0.0000002 -38 1 0.0000001 -33 1 0.0000001 38 1 0.0000001 40 1 0.0000001
Sally
In an 8-deck shoe, there are 416 cards; taking burnoffs into account, and assuming 5 cards per hand, that is about 80 hands per shoe.
About 10% of them will be ties.
Of the other 72 hands, 50.6825% of them are expected to be banker wins; this is about 36.5 hands.
The other 35.5 hands would then be player wins.
And, the commission on the banker wins will set you back the equivalent of about two hands then for every one that the banker gains over the player.Quote: ThatDonGuy0.93 sounds about right.
In an 8-deck shoe, there are 416 cards; taking burnoffs into account, and assuming 5 cards per hand, that is about 80 hands per shoe.
About 10% of them will be ties.
Of the other 72 hands, 50.6825% of them are expected to be banker wins; this is about 36.5 hands.
The other 35.5 hands would then be player wins.
Quote: DanthemanI was recently shown the results of a simulation which shows that the average difference in wins between player and banker over a baccarat shoe is 0.93
(Banker or player wins one more hand than the other over the shoe)
In my experience I have found it to be much higher (approx 4-5 on average)
Does anyone have any thoughts on this?
Dan,
When I have questions about Baccarat data, I usually check the Wizard's simulation of 250,000 8-deck Baccarat shoes here. The Wiz seems to call attention to his having used a Mersenne Twister random number generator and a Fisher Yates shuffle when he constructed his simulation to maximize true randomness.
My examination of the WoO data shows the Banker wins 9,272,972 hands versus 9,025,043 Player wins. So, the Banker wins 247,929 more hands than the Player, for an advantage of 0.991716 hands extra per shoe. So, based on the WoO data, expect Banker to win about 1 extra hand per shoe (in a fair game).
I now agree that the average margin of wins between Banker and PlayerQuote: DanthemanI was recently shown the results of a simulation which shows that the average difference in wins between player and banker over a baccarat shoe is 0.93
(Banker or player wins one more hand than the other over the shoe)
In my experience I have found it to be much higher (approx 4-5 on average)
Does anyone have any thoughts on this?
is closer to .992 (that looks to be about 1)
the program I used was not adding up the counts correctly
and I do not have the actual code to change/correct it.
so I did it another way instead of doing it all over.
take the probabilities for the number of B&P hands per shoe
and multiply that by the binomial probability of the number of wins to produce the margin
lots of math but that is what spreadsheets are for
0=0.046183611 a Tie in wins for each
so for (X more Banker wins)
1=0.0465052
2=0.046196064
3=0.045277644
4=0.043775277
5=0.041760316
(X more Player wins)
1=0.045252736
2=0.043741293
3=0.041717068
4=0.039246613
5=0.036431778
this all sums to 0.476087601
so more than 5 Banker or Player wins = greater than 50%
this is what the OP has felt playing many shoes (me too)
but the average margin is still 1 win
when all is added up and divided among the shoes
I finish the calculations later
for a more accurate probability table
-X = Player margin
X=Banker margin
margin | probability |
---|---|
-5 | 0.036431778 |
-4 | 0.039246613 |
-3 | 0.041717068 |
-2 | 0.043741293 |
-1 | 0.045252736 |
0 | 0.046183611 |
1 | 0.0465052 |
2 | 0.046196064 |
3 | 0.045277644 |
4 | 0.043775277 |
5 | 0.041760316 |
Sally
Quote: mustangsally
this all sums to 0.476087601
so more than 5 Banker or Player wins = greater than 50%
this is what the OP has felt playing many shoes (me too)
but the average margin is still 1 win
when all is added up and divided among the shoes
Sally
Yes I think that may be where I missed the point, by including when I see player win by 4-5 hands over shoe ASWELL AS when banker wins by a similar amount and not averaging the 2 out.
So when the player wins more hands than banker over a shoe, what is the average amount of hands it will win by?
Player = 2.9Quote: DanthemanSo when the player wins more hands than banker over a shoe, what is the average amount of hands it will win by?
Banker=3.9