Insurance bets-Ideal results?

[bj bob]

Any competent AP knows when to take insurance, at which count, and is aware of the fact that this is a substantial boost to one's EV. Theoretically, if one's insurance bets are > 33% accurate they have a winning situation. My question is what is the optimum result, in terms of percentage, that an AP can expect to achieve and which system out there has the highest IC?

 

[zengrifter]

My question is what is the optimum result, in terms of percentage, that an AP can expect to achieve and which system out there has the highest IC?

I forget what % "perfect insurance betting" can generate... BUT the only count that can provide perfect insurance is the 10-count, either traditional 10s/non-10s ratio method -or- 10s -2/non-10s +1 RC point count method. zg

 

[Automatic Monkey]

I forget what % "perfect insurance betting" can generate... BUT the only count that can provide perfect insurance is the 10-count, either traditional 10s/non-10s ratio method -or- 10s -2/non-10s +1 RC point count method. zg

A practical count with strong insurance correlation is to count 10's as -2, 2 through 9 as +1, and aces as 0, with or without a sidecount of aces. With the sidecount it actually compares reasonably well to other counts, and can give close to perfect insurance estimation.

 

[EasyRhino]

I can't comment on the theory, but when I first started I didn't understand that insurance bets only needed to win more than 33% of the time before they paid off. I was getting pissed insurance in a high count seemed to be less than half the time.

... if only there was some place they would clearly print that insurance pays 2 to 1...

 

[Mimosine]

... if only there was some place they would clearly print that insurance pays 2 to 1...

hahahaha.

i just re-read the other night either in bluebook or KO that the house's edge on insurance is 2.3:1 or 2.4:1 (if i recall), without knowledge of count.

 

[bj bob]

A practical count with strong insurance correlation is to count 10's as -2, 2 through 9 as +1, and aces as 0, with or without a sidecount of aces. With the sidecount it actually compares reasonably well to other counts, and can give close to perfect insurance estimation.

OK Mr. Monk, You've clearly answered the first half of the question, i.e. what is the best count for IC. Now, could you calculate what sort of percentage one could expect from this ideal insurance strategy? (35-40% return tops?)

 

[zengrifter]

OK Mr. Monk, You've clearly answered the first half of the question, i.e. what is the best count for IC. Now, could you calculate what sort of percentage one could expect from this ideal insurance strategy? (35-40% return tops?)

No, I answered "whats the best insurance count" - His is the most practical. zg

 

[Automatic Monkey]

OK Mr. Monk, You've clearly answered the first half of the question, i.e. what is the best count for IC. Now, could you calculate what sort of percentage one could expect from this ideal insurance strategy? (35-40% return tops?)

My sim gets 6.074% on insurance bets with a flat bet, 8D.

8.285% for DD.

13.36% for SD, RO6 pen.

You get higher advantage at lower numbers of decks because you end up in extreme highcounts more often.

 

[zengrifter]

My sim gets 6.074% on insurance bets with a flat bet, 8D.

8.285% for DD.

13.36% for SD, RO6 pen.

You get higher advantage at lower numbers of decks because you end up in extreme highcounts more often.

What do these percentages mean? zg

 

[Automatic Monkey]

What do these percentages mean? zg

That's the advantage on insurance bets you will experience with a flat bet, and taking insurance at a true count of +5 using the {0,1,1,1,1,1,1,1,1,-2} count.

Using a spread will give you a higher advantage still.

 

[bj bob]

Sorry, ZgMan

No, I answered "whats the best insurance count" - His is the most practical. zg

Forgot to give you your due credit. MEA CULPA. So, what's your best estimate as to pct. of Ins. accuracy in a perfect world?

 

[bj bob]

My sim gets 6.074% on insurance bets with a flat bet, 8D.

8.285% for DD.

13.36% for SD, RO6 pen.

You get higher advantage at lower numbers of decks because you end up in extreme highcounts more often.

Wow! Didn't realize the huge difference between 1D-2D-almost half again the EV. So that means it must be fantastic at Bxxxx(Ro7). Can't wait!

 

[zengrifter]

So, what's your best estimate as to pct. of Ins. accuracy in a perfect world?

A tiny bit higher than the "practical" version. zg

 

[bj bob]

Vote For Zg!

A tiny bit higher than the "practical" version. zg

Zg, With an answer like that you'd make a great politician. You probably met your share during your stay at the Federal C.C. That would make a great campaign platform. "Hi folks! My name is Zg and I'm running for national office of(fill in the blank) and unlike the rest of my oponents I spent my time behind bars BEFORE I got elected." Can't miss with that strategy, it's different. Anyway, sign me up for you next run and don't forget to cut me in.

 

[SystemsTrader]

I was playing around with the efficiency calculator in CVData and some counts will yield a perfect insurance correlation of 1. or .999. I understand that IC is the correlation between the tag values and the value of the cards in the insurance situation but I'm unsure how a system can correlate to 1. which would be perfect? Does that mean that at a certain count you would be able to predict a ten 100% of the time? I've searched through Theory of Blackjack but couldn't come up with the answer.

 

[QFIT]

I was playing around with the efficiency calculator in CVData and some counts will yield a perfect insurance correlation of 1. or .999. I understand that IC is the correlation between the tag values and the value of the cards in the insurance situation but I'm unsure how a system can correlate to 1. which would be perfect? Does that mean that at a certain count you would be able to predict a ten 100% of the time? I've searched through Theory of Blackjack but couldn't come up with the answer.

No it doesn't mean that you can predict Tens 100% of the time. It means you have 100% of the information that you can get via straight counting.

 

[sagefr0g]

arrrrrggggg this insurance thing drives me wacko.
i probably even asked this before.
just having trouble understanding what it is that makes insurance a profitable bet.
i mean yeah you take it at the correct TC or higher and it's profitable in the long haul........ but why?

i mean i think here is how it goes.....

case A
any cards you got (except a blackjack) and the dealer has an ace showing and you take insurance. then if the dealer has blackjack, she takes your origninal bet and pays you 2:1 on your insurance bet which was half your orignal bet. so ultimately thats a push.

case B
but if you have a blackjack and the dealer has an ace up and you take insurance then your blackjacks push and you get payed 2:1 on your insurance bet. so there you actually come out ahead.

so it seems to me that case A only keeps you from losing money while it is really case B where you actually realize profit.

 

[Sonny]

so it seems to me that case A only keeps you from losing money while it is really case B where you actually realize profit.

Yes and no. Case B obviously wins you money so that’s a good thing. Case A gives you a push instead of a loss. That’s a “gain” of 1 unit (0 instead of –1). In that situation a push is actually a win because it is better than a loss.

Imagine if you were playing against a dealer who made a lot of mistakes. You notice that the dealer sometimes doesn’t take your bet when you have a losing hand. Wouldn’t that give you a pretty big advantage? It’s the same thing with insurance. The dealer is occasionally not collecting your losing bet. A push is better than a loss, so you can think of it as “winning” 1 unit.

-Sonny-

 

[sagefr0g]

Yes and no. Case B obviously wins you money so that’s a good thing. Case A gives you a push instead of a loss. That’s a “gain” of 1 unit (0 instead of –1). In that situation a push is actually a win because it is better than a loss.

Imagine if you were playing against a dealer who made a lot of mistakes. You notice that the dealer sometimes doesn’t take your bet when you have a losing hand. Wouldn’t that give you a pretty big advantage? It’s the same thing with insurance. The dealer is occasionally not collecting your losing bet. A push is better than a loss, so you can think of it as “winning” 1 unit.

-Sonny-

yes & no. i like that lol.
so maybe it's a gain over EV? to where a push on normal hands isn't?

 

[zengrifter]

Zg, With an answer like that you'd make a great politician. You probably met your share during your stay at the Federal C.C. That would make a great campaign platform. "Hi folks! My name is Zg and I'm running for national office of(fill in the blank) and unlike the rest of my oponents I spent my time behind bars BEFORE I got elected." Can't miss with that strategy, it's different. Anyway, sign me up for you next run and don't forget to cut me in.

Sounds good, I'd like to run it by MDLBJ first. zg