It is about the certainty of BR growth. Your long run BR will be best suited if you insure optimally. But in the short term the certainty of BR growth is best if you don't lose both insurance and your original bet. Without insurance on a strong hand you are vulnerable to lose what is a strong likelihood to be a winner. On a weak hand insuring likely will cause you to lose both the insurance bet and the original bet. Solely for the reason of certainty of BR growth many choose to insure strong hands a little below the index and weak hands not until the index is exceeded by a TC or 2. It is not optimal for BR growth but it is optimal for the certainty of BR growth. In other words you give up the tiny EV of mistakes on either side of the index very close to the index where there isn't much of a difference (bear in mind insurance accrues/loses EV very quickly on either side of the index) for added short term swing control and possible heat moderation. Generally the ploppy "wisdom" is not to insure weak hands so you make sure you don't have the likely outcome of losing 1.5 bets and insure strong hands so you to "lock in" and 0.5 bet win or push. When it comes to RA insurance or heat averse insurance it is the same principle.
Last edited by Three; 08-18-2016 at 07:00 AM.
It should also be pointed out, regardless of what t3 or myself is saying - at the end of the day, insurance is nothing more than a side bet. We may ramp our bet sizes and ramp our side bets based on factors, but the best EV say to play insurance is full bore at or exceeding the index. Deviations from the above would be based on factors stated above.
I'll answer 2 ways.
1. Hi LO most counters will insure everything at true 3.
2. T3 used an interesting phrase that I was going to use but didn't - risk averse insurance. In other words, why insure a crap hand at hi LO at exact true of 3.0, since the gain over your lifetime will be marginal while your day to day variance will be increased. At higher true counts, the insurance decision is obvious. The balance of his comments about bankroll certainty, though wordy, are, in my view valid. He us getting better, kinda sorta, on the wordy part
I am picking where wordy id more necessary versus not necessary. When talking to a beginner or one that sounds like they would benefit from more detail I often do so just for that individual and any others in the same boat.
It is hard to say but based on comments on the forum it seems most never care to learn enough to understand more in-depth decisions. This should be a concern of those with a small BR. A huge BR often doesn't worry about the short term at all. Some in that situation still like greater certainty of short term income rather than a slightly larger long term income.
But they need to learn. The way to do that is to explain things to them. They might not understand all of the explanation but in trying to fill in the blanks they will learn how to teach themselves from the various resources available to consult. The lazy ones that just want everything spoon-fed to them are almost all destined for failure. The ones that seek out info are the ones that have the best chance. This is how you get them to seek out info on their own and ask questions to get interactive clarification on the site that they can't from non-interactive resources. Spoon-feeding them everything is setting them up for failure. You get 2 types that are more likely to fail. The first tries to learn everything from the forum and the second asks lots of questions but refuses to or can't listen to the answers. Not challenging the newer players makes them at higher risk to be in one of the ore likely to fail groups. There are 2 prime examples of each these types on this site, Zeebar and LoneWolf respectively. Neither has failed quite yet. It seems Zee is getting the right idea about being more proactive in use of the great reference material and software but LoneWolf never seems to want to listen to anybody or even everybody even after asking for their opinions.
As a humor point on the subject, a post, some years back by Don addressed some of the issues. The personal story (I've forgotten the exact details, so creative license will be utilized)
He referred to his pat 20, against dealer ace with a true count of 2.9645328 which was below index of 3.00000, and accordingly, rejected insurance as true was below index. Odds were not with him, as dealer hit the snapper.
Since then, I've always believed his deck estimation to be "above average"
Actually the index of hilo for insurance could possibly be 3.1, 3.2, etc. Most authors wont write in the decimal form into the index chart and most just floor down to a whole number to make it as user friendly as possible. The true indices for each count on each play deviation are all in the decimal forms. 1.5, 1.7 , 2.4, 3.6 etc, but all rounded down to an even integer.
This is not how indices are created as has been discussed many times. CVData, SBA, and Cacarulo's index generation routines do NOT calculate decimal indices and then integerize them. This would result in some incorrect answers as blackjack is inherently non-linear. They create integer indices.
Last edited by Norm; 08-19-2016 at 06:09 AM.
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
Rounding down positive integers of x.5 of higher (.5 being the decimal in a positive count) would result in bad/incorrect indices. You have to understand that there will be a higher frequency for the decimals closer to 0 for any integer so you would be making more costly mistakes more frequently after x.4something. You are forgetting the TC frequency bell curve and its affect on TC frequencies.
Bookmarks