Insurance Play - Question for Don

I just read something that I never heard anywhere before. When making an insurance decision (using Hi-Lo), should one disregard the dealer ace for purposes of the TC? For example, suppose the RC is 9 factoring in all cards on the table including the dealer ace. Would it make sense to increase the RC by 1 and use a RC of 10 for the insurance decision (effectively ignoring the dealer ace)? After all, insurance is a bet that the dealer has a 10 in the hole. Why reduce your count by 1 when the ace has no bearing in this particular decision?

Of course, once the dealer closes insurance then the RC goes back to 9. While it won't make or break one's game, but it would make the insurance decision slightly more accurate, no?

Thanks,
MJ

The simplest answer is to make the insurance bet in regard to the running count based upon all cards observed.

While the index for insurance can actually change depending on which cards you hold, you have to count them all. The indices are generated with the knowledge that the dealer shows an ace. You have to count it.

Don

Since it is hi-lo, you could keep an ace side count, and adjust the count for a surplus or shortage of aces, for the insurance bet only. You would include dealer up card in this case. I do that with single deck in Wendover, but shoes would be too much for me.

You STILL count the ace in the primary Hi-Lo count, even when you keep a side count for insurance purposes. You then adjust the running count, depending on the number of aces, before you convert to true count. Regardless, you still count the dealer's ace when you see it.

Don

Why not generate the insurance index without including the ace? Would that not yield at least some theoretical advantage? It certainly wouldn't hurt EV, or would it?

Thanks,
MJ

I don't understand your question. What is the point? Why don't you want to include the ace? What difference does it make to you? Why would not counting it do anything beneficial? Why would ignoring a card ever make things better?

Don

Don,

The reason you do not want to count the ace (for only this particular play, namely insurance) is because it does not add any information of value.

Insurance is simply a side wager of whether or not the dealer has a 10 in the hole. The hi-lo does not distinguish between aces and tens as they are both lumped into the category of high cards, each having a tag value of minus one. Wouldn't factoring in aces that have been played weaken the accuracy of the insurance decision? I believe this is why the 10-count has an insurance correlation of 1.0 whereas the hi-lo only has an insurance correlation of about 0.97.

Now why would you want to generate an index value using a method that is sub-optimal to begin with? This would be like building a house on a weak foundation.

Suppose you are playing a single deck game--6 low cards and 7 high cards have come out of the shoe yielding a RC of -1. Included in those high cards are 3 aces. For the purposes of insurance the RC is actually +2, no? Isn't this the reason why some counters use a side count that tracks aces?

Thanks,
MJ

Based upon your post wouldn't it be ideal to keep an ace side count while playing hi-lo? Just make sure you deduct aces that have been played from the RC prior to making the insurance decision.

If this is the case, then wouldn't it make sense to generate an insurance index that accounts for this particular style of play?

MJ

Yes, for all pitch games. See Victor's Insurance Parameter.

If you keep a side of aces for insurance purposes (fairly useless in shoe games for other plays), then the index changes ever so slightly from Hi-Lo 3.1 or 3.2 to 3.0. Not worth talking about.

But you keep asking the same question, and I keep answering it. If you ask three more times, I'm not going to change my mind! :-) Hi-Lo reckons the ace. Period. Whether you side count or not has nothing to do with counting the ace in the primary count for Hi-Lo.

Don

You keep a side of aces in Hi-Lo to adjust the RC to be more accurate for insurance. That has NOTHING to do with the fact that you count the aces in the primary count, which you must do for Hi-Lo, or it isn't Hi-Lo anymore!!

You didn't mean IC of 0.97 for Hi-Lo. It's 0.76. BC is 0.97.

Don

I don't think you are aware how much more trouble it is to keep an Ace side count.
I can speak from experience since I'm doing exactly this in just one casino that offers a special side-bet which is triggered when you get a natural. It is clear that for this side-bet the ace density is crucial.But for the insurance bet it's only the ten density that counts and with the ace adjustment you get only little improvement as Don pointed out. You still don't know how many neutral cards are remaining or do you also want to make a side count of those?
By the way. in spite of using two different languages for high-low and ace side count it still happens that I get mixed up, especially when several aces fall in the same round.
Bottom line, I wouldn't bother to count aces just for insurance.

If you want to be perfect with insurance, I have a suggestion for you: start at -24 (6 decks); count the tens -2 and all the others +1; you insure whenever the count is positive.

Francis Salmon

Ace side count is easy in single deck. Easy to keep and easy to adjust the hi lo running count when a potential insurance bet comes up. I use foot position to keep track. See Professional Blackjack for running count adjustment method. I play Wendover only once a year but it is worthwhile for me. When I played single deck only in the old days I used hi-opt II and had to track aces. When shoes took over more and more I switched to hi lo. Tracking aces in shoes was too much for me.

If the insurance index number is barely there for taking insurance after counting ALL seen cards, just quickly make the slight adjustment for the dealers ace and perhaps any other aces currently on the felt. Then act accordingly. I wouldn't even consider worrying about it on shoe games.

You could do something similar if Dealer shows 10 and you have a hard 16 in a multi-deck game and the count is 0. Look at the layout primarily for 5's. Then act accordingly to either hit on not.

A professional level chess player is probably doing the rough mental equivalent of five side counts.

Of all the people I trained/observed it doesn't take more than about six months at the extreme to master an ace side count, with most players, much less.

The issue really is whether the additional difficulty is going to compensate for increased error and fatigue. This is more an area of psychology than mathematics. It may surprise you to know that some, though by no means all, players actually find they find they make fewer mistakes and can play for longer. This only happens with players who find hi-lo beneath them and tedious-the ace side count adds a level of challenge for them.

A professional level chess player is probably doing the rough mental equivalent of five side counts.

I happen to be an International Chess Master and therefore in a good position to tell you that this assertion is grossly exaggerated.
A Chess player calculates one line after the other and imagining or recalling a sequence of moves has nothing to do with the simultaneous memorization of several (moving) numbers.By the way very few chessplayers become card counters.

Of all the people I trained/observed it doesn't take more than about six months at the extreme to master an ace side count, with most players, much less.

When after one week of training for a new count a havent reached casino speed I usually stop the project. With the ace side count I succeeded. But the problem is the number of errors you will make in actual casino play.It seems to me that the additional theoretical benefit should be at least 30% of the win rate for the greatly increased effort in order to leave some room for errors.That's why I use it only in that special bonus game.

The issue really is whether the additional difficulty is going to compensate for increased error and fatigue. This is more an area of psychology than mathematics. It may surprise you to know that some, though by no means all, players actually find they make fewer mistakes and can play for longer. This only happens with players who find hi-lo beneath them and tedious-the ace side count adds a level of challenge for them.

I don't think that vanity is a good criteria for choosing the level of card counting,the guideline must always be the actual win rate you can achieve. Frankly I don't believe in this fairy tale that these players will actually make fewer mistakes.

Francis Salmon