fatcat519: Parker re Question

fatcat519 · 12-30-2005, 05:34 PM

I posted a question about insurance below, (Dec 26, 9:25 a.m.), which didn't get any answers. I'm wondering whether it was a dumb question, or whether it wasn't seen by the right people so close to Christmas, or whether it just wasn't thought to be interesting.
I would still appreciate an answer, if there is one.

Thanks, fatcat519
(P.S.- that name doesn't really fit , but it was assigned and I never bothered to change it

Parker · 01-01-2006, 10:59 AM

> I posted a question about insurance below, (Dec 26,
> 9:25 a.m.), which didn't get any answers. I'm
> wondering whether it was a dumb question, or whether
> it wasn't seen by the right people so close to
> Christmas, or whether it just wasn't thought to be
> interesting.
> I would still appreciate an answer, if there is one.

I generally leave questions of this nature to the math gurus, who, for whatever reasons, have declined to comment. It isn't a dumb question by any means, although I am unsure how much practical value it may have.

You asked:

> I don't know whether this question is simple or
> complicated; but if a person is playing, say , HiLo
> and always takes insurance at or above the correct
> index +3, what percentage of these insurance bets
> would he expect to win?
> If it depends on the game, let's assume A/C rules,
> 8-decks, 75%.
> How does this connect to the insurance correlation for
> the system?

Well, since you are only insuring when you have a positive expectation, and insurance pays 2:1, you would expect to win more than 33.3% of the insurance bets you made. The exact number would depend on number of decks, penetration, and the insurance correlation of the counting system in use. I think you would need to run a sim to get an exact number for a specific game.

And of course, we are talking long-term expectations here. In any given shoe, you may "expect to win" all of your insurance bets, none of your insurance bets, or anything in between.

BTW, this illustrates an important difference between the Parker Pages and Don's Domain, where all serious questions will receive an answer from Don and/or one of the other Masters.

Norm Wattenberger · 01-01-2006, 11:11 AM

Good answer. If you are using the right Insurance index, you'll win a bit more than one-third of the time.

Designated Driver · 01-03-2006, 05:02 AM

> Good answer. If you are using the right Insurance
> index, you'll win a bit more than one-third of the
> time.

So Norm, are you saying that if you use a "perfect" insurance count with the appropriate index, of course, that you would only win a little more than one-third of the time? However I surmise that you would realize a profit as it pays 2:1, right?

Also Norm, I have a few questions about CVBJ 4(being able to customize a count). May I post them here?

Thanks,
Desi. D.

Norm Wattenberger · 01-03-2006, 09:02 AM

> So Norm, are you saying that if you use a
> "perfect" insurance count with the
> appropriate index, of course, that you would only win
> a little more than one-third of the time? However I
> surmise that you would realize a profit as it pays
> 2:1, right?

At the Insurance index you will win slightly more than one-third. At higher indexes, which occur less often, you will win at greater frequencies and may bet more.

> Also Norm, I have a few questions about CVBJ 4(being
> able to customize a count). May I post them here?

As long as they aren't too complex on the Computing for Counters page. Complex questions are better by e-mail.

fatcat519 · 01-03-2006, 10:59 AM

> I generally leave questions of this nature to the math
> gurus, who, for whatever reasons, have declined to
> comment. It isn't a dumb question by any means,
> although I am unsure how much practical value it may
> have.

Thanks for your answer, and to those who answered below.
I agree that there is probably not much practical value in this. I was just curious about it, and knowing that I won't win many more than one out of three makes it easier to accept the ones that lose.

Norm Wattenberger · 01-03-2006, 03:47 PM

Well put. In the particular case that you mentioned in your original post, you will win 34.3% of the time.

> Thanks for your answer, and to those who answered
> below.
> I agree that there is probably not much practical
> value in this. I was just curious about it, and
> knowing that I won't win many more than one out of
> three makes it easier to accept the ones that lose.

Designated Driver · 01-04-2006, 12:19 AM

> At the Insurance index you will win slightly more than
> one-third. At higher indexes, which occur less often,
> you will win at greater frequencies and may bet more.

But if the count is at or above the Insurance index, you would always place the full insurance bet(not a fraction) if offered, right? So, like doubling, if it is right to insure, it is right to insure fully or not at all(if you have the necessary funds), correct? And although you may wish to insure yourself for more at higher counts as buying life insurance can be quite profitable at times, especially when facing/biting a bullet(ace), this isn't allowed, is it? You can only insure for an amount up to half of your original wager, right? I hope that wasn't too many questions. :-)

> As long as they aren't too complex on the Computing
> for Counters page. Complex questions are better by
> e-mail.

Okay, thank you.
Desi. D.

Norm Wattenberger · 01-04-2006, 04:23 AM

Insure for less makes little sense unless you are in a tournament or are below the index and trying to throw off counter-catcher software. Insure for more is very rarely possible. But with a larger spread, you may bet more on the underlying hand at higher counts and therefore have a larger Insurance bet at higher counts.