Last Hand Based Bet Adjustment

For a while, when explaining to friends how to play blackjack, I referred them to simplified and basic strategy, but there always arises the question of how to bet. There I just suggest to remember the more cards come out, the fewer are left, so one should lower the bet after blackjacks, especially multiple, and raise, if so desired, only after long hard hands. It's not counting and won't beat the house, but beats betting on a whim, it seems.

Lately I thought it could actually make sense to formalize this into a few simple rules. First of all, of course, I'm sure I'm not the first one to have come up with this, so maybe someone has done it already and there is something, somewhere already describing and detailing it?

If not, then, how effective can such an approach really be, i.e. by how much can it reduce the edge, and how can it be improved?
So far it's the following set of rules, mostly for double-deck games. It's assumed the player doesn't scan the table, but blackjacks are hard to miss.
Blackjack push - take a break
Multiple blackjacks or soft hands - drop to minimum
Blackjack - lower the bet
T+T vs T+T - lower the bet
Mixed hands - stay
Long hands (4+ cards, all hard) - ready to raise
Repeated without any lowering events - raise

It's pretty easy, the intention is not to keep any counts, just bring some reason into betting, but I'm not sure how correct it is. Assuming the average bet remains the same (e.g. playing $5-$20 instead of flat $10), can a player expect to lose noticeably less on the average, or is such behavior too far from proper strategy to be an improvement?

Quote: P90

It's pretty easy, the intention is not to keep any counts, just bring some reason into betting, but I'm not sure how correct it is. Assuming the average bet remains the same (e.g. playing $5-$20 instead of flat $10), can a player expect to lose noticeably less on the average, or is such behavior too far from proper strategy to be an improvement?

In essence, what you are recommending IS counting--adjusting your bet according to what cards have come out. The player is only considering his own hand, but that's valid as far as it goes---the remaining cards are effectively as randomly distributed as they would be if they were all in the undealt portion of the deck (note: that isn't COMPLETELY true, but for the sake of this discussion, that's good enough).

This will work because it's a betting-decision count only (not: a playing decision count), and it properly weights aces the most, followed by tens. Is it accurate? Not really. Will it make a difference? Yes. How much? Probably 0.3-0.5%.

Another baby step might be for the player to guesstimate the proportion of high and low cards in the other players' hands. For instance, if the dealer shows a ten, but most of the players have quickly tucked their cards under their money anyway, the deck is probably ten-poor, all other things being equal.

Of course, the above only applies to a single-deck game. The information wouldn't be worth much at all in a shoe game--not enough to justify adjusting your bets.

Quote: mkl654321

In essence, what you are recommending IS counting--adjusting your bet according to what cards have come out. The player is only considering his own hand, but that's valid as far as it goes---the remaining cards are effectively as randomly distributed as they would be if they were all in the undealt portion of the deck (note: that isn't COMPLETELY true, but for the sake of this discussion, that's good enough).

Yeah. Just instead of adding numbers in his mind, the player adds or subtracts from his bet (within his normal betting range, of course). One of the first points is to replace the practice of increasing bets after a good hand, which is damaging to the player.

While casual counting (like in estimating the proportion of cards) can be useful, my concern with it is that it's prone to errors from wishful thinking and distractions, and these errors can be dangerous if accumulated. So the idea is to keep it to a few formal rules that are unaffected by perception and can be as automatic as basic strategy, for a player who doesn't intend to go to beyond recreational gambling. As for determining the proportion of high and low cards, I prefer using hand length, since it has decent correlation and can be formalized.

What I'm looking for is balancing it, so that in an evenly depleting deck it keeps the bet either stable or slowly decreasing (as a safeguard). I've noticed that casual counting tends to lead to escalating counts.