Which deviations are most valuable?

[aceside]

Can CVData analyze deviations and, like Don did with the Ill18, list them in order of EV/SCORE? I know it's a hard thing to do computationally, and it depends on your bet spread and probably a bunch of other stuff. But it would be nice to know which deviations to learn first. (https://www.youtube.com/watch?v=E8Tdvgpdbrc)

I feel like I can't be the first to ask this question, so if there's already a thread on this that I can't find, I'd appreciate a link to it.

I do not quite understand this question very well. I guess the most important thing in Blackjack is bet spread, and play deviations are minor. If you work so hard on minor important things, what is the value of it?

[PitBoss321]

...play deviations are minor. If you work so hard on minor important things, what is the value of it?

But they aren't minor! If we take Colin's word in the video I linked, 20 - 40% of our edge comes from playing deviations, not betting variation. But, a lion's share of the value from play deviations comes from Insurance and 16v10. So my question is how fast does the value of deviations drop off (for example, https://blackjackincolor.com/truecount3.htm here you can see in the first chart that the I18 get you most of the value of full indexes bet -3 and +6), and which are next on the list?

And, as a corollary, what indexes are valuable under different systems than those the I18 were made for? (3/4deck, s17, 1-12 bet spread)

[aceside]

But, a lion's share of the value from play deviations comes from Insurance and 16v10.

I watched that video clip but intuitively I don't believe what is stated there. It is very clear to me that the most frequent encounter of hands is 16v10, so this must be the most valuable deviation if flat bet. The insurance bet is a lot less frequent, and thus a lot less important. I guess 15v10 should be a lot more important than insurance too. However, the ultimate question I ask is this: If we flat bet, how much edge can a player gain from playing deviations only? Additionally, if a player follow a bet ramping spread of 1/2/4/8/16 for step increasing true count, how much can a player get from combing deviation and spread?

[PitBoss321]

I watched that video clip but intuitively I don't believe what is stated there. It is very clear to me that the most frequent encounter of hands is 16v10, so this must be the most valuable deviation if flat bet. The insurance bet is a lot less frequent, and thus a lot less important.

Your intuition will lead you astray! Just because insurance may be rare (1 in 13 hands doesn't seem rare to me, but that's beside the point) doesn't mean it's not valuable. If you get a lot of EV from a rare event, it can balance out to be profitable. From Blackjack Attack, insurance is actually the most valuable deviation you can learn, one third of the total value of the Illustrious 18! (Also, it occurs twice as often as 16v10)

If we flat bet, how much edge can a player gain from playing deviations only? Additionally, if a player follow a bet ramping spread of 1/2/4/8/16 for step increasing true count, how much can a player get from combing deviation and spread?

This is a very good question. These days, you can't really make money flat betting. You'd need a single or double deck game with very lenient rules to do that, or some other way to get an edge.

However, you can make money with a spread. I plugged in your 1/2/4/8/16 spread into CVCX, and you can make $38 an hour with a $10,000 bankroll and 2% risk of ruin

[aceside]

Just because insurance may be rare (1 in 13 hands doesn't seem rare to me, but that's beside the point) doesn't mean it's not valuable.

Insurance is offered 1 in 13 hands, but you take the insurance bet when the true count is above 3. I mean this is very rare compared to 16v10 when the true count is zero. I still trust myself, beside, I believe the surrender option is the most important deviation.

[aceside]

This is a very good question. These days, you can't really make money flat betting. You'd need a single or double deck game with very lenient rules to do that, or some other way to get an edge.

However, you can make money with a spread. I plugged in your 1/2/4/8/16 spread into CVCX, and you can make $38 an hour with a $10,000 bankroll and 2% risk of ruin

Can you also run your CVCX to find out: If we flat bet, how much edge can a player gain from playing deviations only? I do not have CVCX. thank you in advance.

[PitBoss321]

I mean this is very rare compared to 16v10 when the true count is zero. I still trust myself, beside, I believe the surrender option is the most important deviation.

If you keep "trusting yourself" and not the math you are not going to be a successful card counter. I want to be nice, to help you out, but that mindset will destroy any chance of winning money consistently.

As for surrender, late surrender adds 0.07% to your edge off the top, it is a valuable rule if you can get it. (https://wizardofodds.com/games/black...le-variations/) It's hard to compare to deviations, especially since different bet spreads, rules will result in different edges, different values for insurance. But both are valuable.

Can you also run your CVCX to find out: If we flat bet, how much edge can a player gain from playing deviations only? I do not have CVCX. thank you in advance.

In the same situation I simmed before, flat betting with the most valuable (I think) 25 indices still has you at a negative EV, -0.2%. If you wong out, leave the table when the count gets to a true -1, you can make money flat betting though. 10K bankroll and 2% risk of ruin lets you bet $20, for $7.25 an hour.

One issue with this is the N0 (https://www.blackjackapprenticeship....advantage-play) Your bankroll will swing a lot with that play. After 500 hours, you could be anywhere from $10,000 up to $3000 down.

[aceside]

(Also, it occurs twice as often as 16v10)

I do every calculation myself. Let me show you that the player 16v10 is a lot more frequent than dealer Aces. Frequency of the player 16v10 is 42%*4/13=13% for six-decks. Frequency of the dealer Aces is 5%*1/13=0.4%. Therefore, 16v10 is a lot more frequent that insurance offers. I do not believe publications. The 16v10 must be the most important deviation. Besides, the surrender option will make this deviation more important. You cite of wizard of odds "late surrender adds 0.07%" is not correct for card counting. It should be 7% when counting. However, your calculation earning when flat bet with solely play deviation is very helpful for me. Thank you.

[G Man]

Ace, you're wrong in your thinking. Insurance IS the most valuable index play of all. You seem to forget that when you take insurance at +3, you may easily have a bet out there that is 8 to 10 or 12 times as valuable as when the TC is 0.

[Norm]

Exactly. Also, penetration affects the importance of different indices.

Can you also run your CVCX to find out: If we flat bet, how much edge can a player gain from playing deviations only?

Yes.