At an online blackjack game, there was an odd bug.
If you split aces, and draw a 10 to the second ace (no matter what the
first ace draws), the payoff on the 21 is 3:2.
Does anyone have an idea what the affect on EV would be?
Would anyone trust a game like this to be fair? (I know I don't!)
You need a HiLo count of +2 v. 5 and +1 v. 6 to split the 10's. So, in a game where you can't count, don't split.
Hope this helps,
MF
not split 10s...
Since the hand doesn't come up often, and more often than not you will not get a 21, I don't think it'd have that much of an effect on the outcome.
I answered the wrong question. Maybe somebody out there was wondering when to split 10's with 3:2 bj payoff. Probably not though...
Regards,
MF
6 deck shoe, probability of getting two aces to split: .00568
1 deck, probability of getting two aces to split: .00439
This data is taken from Wong's Professional Blackjack which is a simulation involving penetration and two hands being played against the dealer, if there is a shuffle after every round then the frequency can be determined by the following formulas --
6 deck: ((24/312) * (23/311)) * .97796 ;
for 1 deck: ((4/52) * (3/51)) * .98694
(The .97796 and .98694 parts account for the rare you have two aces and the dealer turns over a blackjack.)
After splitting, getting a Ten on that second Ace is:
6 decks: (96/310) = .30968
1 deck: (16/50) = .32
So you get the extra half bet bonus:
6 decks: .00568 * .30968 = .00176 hands
1 deck: .00439 * .32 = .0014 hands
So assuming the standard 100 hands per hour, $25 betting unit, this earns you $0.000175 cents per hour, in the most favorable 1 deck situation.
Try not to spend it all in one place.
I suspect the code on the second hand is falling thru to "pay ace and face 3:2" routine, not having a flag that a split had occcured.
YAHOO games blackjack pays split 10s and split aces 3:2 when you get a "non-natural" natural. You can also do infinite splits and resplit aces and hit split aces multiple times and double down on any number of cards even after splits.
Despite all this, the average YAHOO player is a real yahoo and manages to lose consistently. In fact some players claim the YAHOO game is rigged, although there would be no point to it since the YAHOO game is just for fun, there is no profit motive to cheat.
I suspect your on-line game has the same experience with people losing. Why do you think they can afford to pay you that sign-up bonus I assume you are getting?
...casino, as they have just released "VIP split" blackjack, which has this rule.
Why do you think they can afford to pay you that sign-up bonus I assume you are getting?
Hmmm, lets see, industry stats (source: Gambling Online magazine Feb 2001) say that on average an online casino client loses $3000 a year. So I think they'll have a little change over after "giving" away a $25 bonus.
I read recently on the wizards site that Casino-on-Net pays 3:2 on BJs off split 10s. Said you should never split them, for the obvious reason you just stated.
This is the standard MG software. House still has an edge, though splitting TT vs. 6 is a bit more fun.
See wizardofodds.com for more details on BS for this game.
--Eliot
real cards,that's the tip for today amigo.
Unless your counting, which of course doesn't guarantee profit. Think I'll stick to online for the moment, considering how much more profitable it is, at a much lower level of risk.
You're giving money to some offshore company that you have no legal recourse if they screw you out of all your money, and you say it has a lower level of risk?
Most online casinos use 4 decks, shuffled after every round.
Thus there are 16 Aces out of 208 cards; so the chances of being dealt a pair of Aces in the first place are 16/208 * 15/207, or 0.005574.
Now there are two possibilities: either the card to the first Ace is a 10 or it isn't. If it *is* a ten, which will occur 64/206 of the time, that would leave 63/205 the card on the second Ace will be a 10. If it is *not* a ten, which will occur 142/206 of the time, that would leave 64/205 the card on the 2nd Ace will be a 10.
This makes the prob of occurence 0.005574 times the sum of:
64/206 * 63/205 + 142/206 * 64/205;
or 0.001721. This is worth an extra half unit; do your overall EV would be increased by 0.086%. Assuming "normal" rules, this would raise the EV from about -0.4% to -0.314%.
...amigo:). Touche away...
...also see Bob Fisher's post on the Internet page concerning the option to take insurance on your second split hand after seeing your first hand. Bob wasn't sure if you could get this with a hand other than aces, but I managed to confirm that you could. Whether this is ever adviseable however is combinatorics stuff I don't have time for personally.
The is nothing better than getting paid real cash after a hard days work.
the chance they will screw you is extreme.
SOTSOG,
A couple of questions about your blackjack calculation.
First, when you figure the prob that you will get AA vs a dealer blackjack , the prob for the dealer blackjack given AA by the player should be 2/50 * 16/49 * 2 or .9738. The dealer can make a blackjack by either having an ace as the first card or having a 10 as the first card.
Secondly, I think your hourly rate of $0.000175 is a little low. You give the prob of getting split aces with a 10 on the second split as occurring 14 times in 10,000 hands. If you play 100 hands per hour, you will see this .14 times. A half a bet at quarters is $12.5 so you will earn $1.75 per hour.
