Indices

[Count Luckula]

I recently ran a simulation for 8D DOA DAS 82%pen using Hi-Lo. I set the risk aversion ratio to 1. My question is, how should I use this risk aversion ration. The literature I have is hard to understand and I am wondering if this ratio is affecting my results. For example, for A4 vs. 4 I am getting an index of +2 and basic strategy says to always double. Basic strategy doesn't take into accouunt that you have more money on the table at higher counts, so why would BS tell you to double always? Other interesting ones are A7 vs A, A2 vs 5, and insurance.

 

[Rob McGarvey]

RA Indices

You are seeing what I was recently talking about when I asked about RA for onLine BJ. If you set the RA to 2 instead of 1 your + numbers will go up a bit further. Being RA with these plays can also be construed as cover as you will not be playing standard BS rules. People will often tell you not to double your bet after a loss ($16-$32) to try and win $2. The same can hold true with risk aversion. With a RA value of 1 the program tells you to skip the double $? x 2 for the expected .03% gain (.59 to .62) for A,4 v 4

 

[Abraham de Moivre]

I don't know what software you are using, or what this risk aversion ratio setting actually does.

However, the numbers you are giving seems perfectly reasonable. (you forgot to specify H17/S17, which is important.)
A,4 vs 4 and A,2 vs 5 are very marginal plays. You might as well flip a coin. Basic strategy says double because when you do it a million times you come out slightly ahead versus just hitting. Using the count, you avoid the variance of putting up twice as much money doubling on marginal plays, but when the count is high and you are more likely to win, (and as you point out, you have the larger bets out) then for sure you double on these.

FYI - I use 0 as my index for A,4 vs 4 on the games I play. I only double when the count is positive. A,2 vs 5 I use -1. If the count is negative, I just hit and eat a few more small cards.

A,7 vs A - are you seeing a +1 or something like that? These means you should stand, not hit. A few extra 10s in the deck make this play a stand. I think the count would have to be so high that it is only in theory that you would ever double on it. I think the theory behind this play is that it takes two 10 cards for the dealer to bust a Ace up hand. One 10 to turn his soft hand into a stiff, and another 10 to bust his stiff. So don't waste a 10 card turning your soft 18 into a hard 18 in a high count. In a low count, the dealer is less likely to bust, and you are more likely to catch some small cards and improve your 18.

Insurance? What do you find interesting about that? Actually I prefer to call it, "Does the Dealer have a Ten Card in the hole?" The count for this play is +3. However, this is the break even point. Meaning at +3, placing the "Does the Dealer have a Ten Card in the hole" bet is just like flipping a coin. I don't like to depend on paying my rent based on the luck of a coin flip. So I only bet the "Does the Dealer have a Ten Card in the hole" bet when I have a count that is slightly higher than +3.

Reducing Risk is all about reducing variance. If things always happen the way they are supposed too (zero variance) then you will always win. One way of reducing variance is to avoid coin flips. Why let the variance of a bunch of small bets, making small marginal plays, gone wrong creep into eating the profits of what is happening with the big bet sure things (hopefully, they are sure things, sometimes you get hammered on those too. LOL).

 

[Count Luckula]

I am using SBA and the game is S17. I understood that the table was saying that you are suppose to stand on A7 vs A at TC's higher than +1, I had just never seen this index until running my own sim. As far as insurance, you were right on the button, the sim gave me a result of +4 (where the Mayor's index for multideck is +3). Does making these plays really seem like cover, or do the "eyes in the sky" recognize these types of moves and can tell you are well educated in the game.

 

[Abraham de Moivre]

Cover

Since the average ploppie stands on soft 18, no matter what, standing on soft 18 vs A @ +1 is probably a good thing. The reason you have never seen it before is because it is a low frequency play. A,7 vs A. You have to have two aces in play at the same time.

Ploppy - stands on soft 18
Player who knows a little something - hits soft 18 vs 9,T,A
Counter - hits soft 18 vs 9,T sometimes stands on A
Ian Anderson - using his ultimate gambit, stands on soft 18 while asking for a comp
ZenGrifter - using his power of hypnotism, convinces dealer his 18 is actually a blackjack and collects 3:2 payoff.

As far as insurance goes, the less you take it the better the cover, since everybody knows insurance is a bad bet and never takes it.
This is a two-edged sword, however. By taking insurance at lower counts, or even in negative counts with a really small bet, the pit will see you sometimes LOSING an insurance bet. By taking insurance less frequently, and always winning, you are waving an even bigger counter flag.

 

[T-Hopper]

Ultimate Gambit

> Ian Anderson - using his ultimate gambit, stands on soft 18 while asking for a comp

Only do this in a stand on soft 17 game. This will cost far more than it is worth for cover if the dealer hits soft 17.

 

[zengrifter]

Re: Ultimate Gambit

hit v 9 stand v 10 would be a H17 gambit trade-off. zg