Beating "Instant 18"

POST DELETED BY REQUEST

Dorothy

Even accepting the premise that a preponderance of 7s in the deck makes a dealer 17 more likely, you would also have to have a condition where 19s and 20s have not become more likely as well. And the likelihood of a dealer hard 17 would have to be increased enough to confer an edge that beats the inherent -EV of the side bet.

I would expect a preponderance of 7s to turn 60% of dealer stiffs (12, 13, 14) into 18-killers. After all, in many modified hi-lo counts, a 7 is counted as +0.5, meaning that two 7s out of the deck is equivalent to two, say, 4s out of the deck. It follows that a deck with a preponderance of 7s is mildly unfavorable to the player. That alone might be enough to negate any side bet advantage, because you have to make a plain ol' blackjack bet in order to make the side bet.

Quote: mkl654321

...a deck with a preponderance of 7s is mildly unfavorable to the player. That alone might be enough to negate any side bet advantage, because you have to make a plain ol' blackjack bet in order to make the side bet.

A shortage of aces, as well as a preponderance of 7's, is good for the sidebet. So, the main blackjack bet would tend to be really bad when the sidebet is at positive EV. Counters who "play all" could have a side count of aces vs 7's to play Instant 18 as a side bet, but counters who "wong out" at negative counts might do better overall.

POST DELETED BY REQUEST

Dorothy

The piece of data that is missing is the house edge on the "normal" hand with these various counts. If the normal hand is sufficiently -EV to negate the side bet 18 hand's +EV, then the count wouldn't help.

I also doubt very much that the 7s and the As have equal weight, which would lead to gross inaccuracies if that was the case.

You may be on to something, but I doubt that a simple Ace-7 count alone would do the job here.

I'm sorry to see Dorothy deleted her posts, which were outstanding. I remember the gist of the tables, and if I have time will try to recreate them myself. However, I would appreciate a comment on why Dorothy deleted them.

The following analysis is for blackjack dealt from 6 decks, H17, dealt to 240 cards. I simulated 25 million shoes dealt to this depth, using a count system of A = +1 and 7 = -1, all other cards were 0. This is a balanced count (obviously), so I applied true count conversions by rounding (not truncating or flooring ... rounding ... ok?). This analysis demonstrated that for a true count of 2 or higher the player has the edge over this bet. I also determined the frequency of the various counts. In doing this analysis, I didn't assume any fixed number of players. Rather I dealt 2 to 7 cards at random to the "players" to simulate their depletion of the shoe.

So, here it is again ... many thanks to you, Mr. Wizard, you're the best ...

True Count	Edge	Frequency
-8	-10.35%	0.03%
-7	-9.47%	0.07%
-6	-8.67%	0.17%
-5	-7.78%	0.36%
-4	-7.03%	0.73%
-3	-6.04%	1.49%
-2	-5.11%	3.11%
-1	-4.10%	7.02%
0	-2.37%	55.67%
1	-1.01%	18.38%
2	0.09%	7.00%
3	1.10%	3.10%
4	1.97%	1.49%
5	2.84%	0.73%
6	3.60%	0.36%
7	4.60%	0.17%
8	5.12%	0.07%

Overall, if you played this bet every time the true count was 2 or more, then you would play it on 12.96% of the hands and have an average edge of 0.91% over the house. I would rank this side bet as beatable, but not worth it. There may be a stronger count available, what do I know?

Ms. D.

Interesting. Given the re-post, can you shed some light on who requested the deletion of your first two posts?

Quote: MathExtremist

Interesting. Given the re-post, can you shed some light on who requested the deletion of your first two posts?

Glad you found it interesting ... the bet is not especially beatable ... if the max wager is $100 then on 12 hands per 100 you'll make about 91 cents, for a win rate of $10.92 per 100 hands. Not exactly time to retire and play this bet, is it? If the bet had any real issues, the deletions would have stood up.

Ozzily yours,

Ms. D.

I feel as if this is for a stand soft 17, rather than hit. I see you say in your post it is H17, but I saw someone else ran this scenario, and this was the table for standing on all 17's.

Quote: Instanto18

I feel as if this is for a stand soft 17, rather than hit. I see you say in your post it is H17, but I saw someone else ran this scenario, and this was the table for standing on all 17's.

It would be better for S17, but the only place that offers this bet, Santa Fe Station, is all H17.

I agree. I am saying the paytable Dorothy Gale made up is for S17. Is there a confirmation one way or another possibly?

Quote: Instanto18

I agree. I am saying the paytable Dorothy Gale made up is for S17. Is there a confirmation one way or another possibly?

I double checked. It is for H17, as it says.

--Dorothy