Cost of Camo

[Oneoffthecount]

Let me appologize if this sounds like a definitive lack of knowledge. I was trying to estimate the cost of betting $100 on the first hand of every shoe. I looked at cvcx and say a loss rate of .23 %, so my simplistic mind thought will, an incremental $75 at .0023 equals about 17 cents per occurence. I wong out so my average first bet appears to be about one per 14 rounds.

Ok, so I played this for 13652 hands and downloaded my cvbj playlog for this period. I Identified hands played at a zero count above the min table bet. I had $4115 loss on 1108 hands and total bet for those hands was $82980. Giving me a whopping loss rate of 5% on those hands. If I assume that my loss would have been 75% less of the $4115 dollars or $3086 dollars, that gives me a per occurence cost of about $3.71 and an overall cost of 30 cents per round over 13652 hands.

Let me preface by saying that I already know the sample size is miniscule and insignificant. My question is do I chalk this up to the evil gods of variance or were my assumptions re cost of betting $100 versus $25 on the first round totally wackadoodle to begin with. If so, Is there anyone who can do the math for what that cost should be?

[Shadowless]

Are the places you play at known to sweat action? If not, cover is generally not needed if you can fly under the radar and keep your sessions short.
What's your max bet if you're betting 100 off the top?

Always betting 100 off the top and then going up or down depending on the count, even if decreasing by 50% each time only after a loss, could become a recognizable pattern after watching a few shoes of your play. I would switch it up, and I always end a session after a max bet has been made the last hand of the previous shoe.

If your play were to be imputed on BJ Survey Voice and evaluated, they'd look for a correlation between high bets and a higher count and how correct your playing decision are, so X amount off the top may be pointless.

[Oneoffthecount]

My max bet is $500. I'm trying to disquise it a little as a 10 to 1 ratio. If I win the $100 bet and the count is RC is between 1 and 4, I consider it an almost 1 TC on an eight deck shoe and bet $50. If I lose or the RC is 0 to -4, I play the table min at $25. My hope is that up front the house will see a guy who plays a few 100 and 50s so they will think its not as odd when I play a max bet later. I try to make it look like I bet based on playing on the casinos money, e.g. I won a hundred now I have thier money to bet with. Such is my simple plan. No laughing allowed.

My sense is that people look at you early and make a judgement as to whether you are a threat. I'm hoping to throw them off ealry.

[DSchles]

Your math is correct. But, while the decrease in e.v. is relatively small, the increase in variance is much greater for those hands. Variance is a function of the square of the bet size, so, in going from one unit to four units ($25 to $100) your variance for those bets is increased 16-fold. When you split to four hands off the top, double all of them, and the dealer draws to 21 and takes your $800, you'll understand why your results from a few hands are utterly meaningless.

Don

[Oneoffthecount]

Thanks Don-

My interpretation is that if my BR can absorb the variability the money in the long run is not as significant. So as long as I can ride the ups and downs, its an ok cover move versus long run expense. The variance gods owe me on this one....

[Oneoffthecount]

Thanks Don-

My interpretation is that if my BR can absorb the variability the money in the long run is not as significant. So as long as I can ride the ups and downs, its an ok cover move versus long run expense. The variance gods owe me on this one....

I brought another 14k hands of data into my model to see if the picture would improve. Knowing that this is still a meaningless sample size but hoping to see some swing in a less negative direction. Didn't really improve, over the course of about 29k hands, I would have lost $16k on hands with a real count and true count of 0 bearing in mind that I would have expected to lose $681 total on those hands played at $100 bucks with a .0023 loss rate. For the same data set my overall loss rate for all TC 0 is -2.95% or about 13 times the anticipated loss rate for true count zero using my strategy penn and rule set. So the question really isn't whether the long run cost of betting $100 on first hand is ok, its whether my bankroll can sustain a $16K or more hit due to variability in the short term (16k representing about 4 months of play). The answer for me for now is nope.

[Oneoffthecount]

Something seems wrong to me.

First, you are playing an 8 deck game with a house edge of .23%? That seems pretty low for an 8 deck game. You didn't mention rules, so maybe there is some favorable rule involved. Otherwise, I would double check this.

If you have mis-calculated this initial house edge, not only would the cost of this cover play be much greater than you are anticipating, but other areas could be effected as well, like when you say raising at "almost TC +1". If house edge is mistakenly higher than you think, you may be raising here when still at a disadvantage.

KJ thanks for questioning. The edge I was referring to was not the overall house edge but the house edge at TC = 0. CVCX has eash win loss rate displayed for each count. For a 0 count the house edge is .23%. This is an eight deck game with H17, DAS, SR, RSA and 6.5 deck penn. Strategy is HILO Sweet 16 and Fab 4. Min is 25. Spread doesn't matter for this analysis since I'm only looking at hands played at TC0. I am not an highly experienced person and sometimes, I fail to communicate clearly. However, if this still seems odd let me know.

[stopgambling]

Maybe with early surrender 10 ,dose not sound right.