Hi everyone,
I have a game of online 6D blackjack with good rules where I can play up to 7 spots.
My question is, on the 7th spot, i will have seen around 18 cards more or less so theoretically I can make better decisions of when to hit stand or surrender. If I were to play perfect computer stategy based on the seen cards, how much better is my EV?
Thanks
Firstly, avoid online blackjack like the plague. You can't count the cards properly that way because the cards aren't dealt without replacement as they are in a brick-and-mortar casino.Originally Posted by dqg
Secondly, your change in EV after seeing the first 18 cards depends on the specific cards you see. 5s have the largest positive effects of removal in a six-deck game at about +0.1212% per card removed from the shoe, and aces have the largest negative effects of removal in a six-deck game at about -0.0973% per card removed from the shoe. If you see the first 18 cards are 18 of the 24 available 5s, then your change in EV is approximately +2.5148%. If you see the first 18 cards are 18 of the 24 available aces, then your change in EV is approximately -1.9082%. Hence, your change in EV will be somewhere in between these two (vanishingly unlikely) extremes.
It's unlikely that anyone here will be able to give you a perfect answer, as "playing perfectly" is not something that most have spent time trying to calculate. This, of course, is what the old "Casey" and "David" concealed computers by Keith Taft did. In any event, we would surely need all the rules, which you haven't expressed.
As a sheer guess, for the shoe game, I'd say that playing perfectly, which involves reckoning every seen card and every possible subset, for each hand, is unlikely to add more than 10% to systems such as Hi-Opt II.
Don
Just read John's answer, above. Perhaps he better understood what you meant. If your online game is replacing all the cards after each round, then perfect strategy on the last hand, for having seen 18 cards, surely isn't worth much at all. For a similar discussion, but involving single deck, see BJA3, pp. 45-47.
Don
Thank you for pointing out that the rules are important to know to answer this question with any degree of accuracy, Don. To clarify, my EORs were calculated using a combinatorial analyzer assuming six decks, S17, double any 2 original cards, no surrender, split only once, no DAS.
Thanks for the answers. I was under the impression that it wouldn't be much better after seeing 18 cards, but the reason for the question is that I thought it might be useful for beating online promos. So maybe a promo is not worth it with .5% house edge blackjack game but it is worth it with a .35% house edge blackjack.
The idea comes from a book I read where there is a person playing 7 spots, 100 in the first, 200 in the second,... and 700 in the last one, so by the time he gets to the spot number 7 he already has good information on the composition of cards and can make better strategy decisions. (he was probably playing a super low edge game to start with)
Please note the following, which is rather important for your analysis. Most online shoe games, and all others, for that matter, have the hands dealt open, which is to say that all 14 cards plus the dealer's upcard are available to be seen and counted, whether you're playing first base or third base. The only difference, therefore, when playing the seventh spot, is the ability to have seen any hit cards for the first six hands. As the average number of cards used for a hand is approximately 2.7, this implies that 6 x .7 = 4.2 additional hit cards will be seen before you play the seventh hand. Hardly worth getting excited about!
Don
Generally speaking the gain from strategy variation with such a small fraction of the deck dealt out will be at best hundredths of a per cent. So you might get the house edge down to 0.34 or 0.33% or something when it was 0.35%.
Almost certainly the gain from doing this is going to be eaten by the extra time involved in making the calculation.
Peter Griffin calculated effects of removal for every possible play in "Theory Of Blackjack". You could get chat gpt to write you a simple python program to calculate perfect play for every conceivable hand combination based on that data. Not worth it for this though, though this approach might be helpful in other online games.
Archvaldor is correct. This isn't just about the small gain against the house edge. If you bet $1 on the 6 spots and then $100 on the 7th spot in efforts to clear the bonus then the whole thing will take much longer just because you still have to play all of those $1 hands before you finally get to the one hand with the larger bet when you have the additional information.
Sometimes it is about the speed of the hands as well. If you are at 0.30% house edge for a bonus that will take 10 hours to clear but you could also clear it in one-fourth the time at 0.34% house edge then I would argue that the additional gain of a couple bucks is not worth the extra hours you are spending to achieve that gain.
I would rather make $300 in 2 hours than $325 in 8 hours. Something like that.
Posting Permissions
Reply With Quote
Bookmarks