Perhaps a silly question but...

[k_c]

Even with statistical rounding, the average TC (assuming we integerize TCs) will be very slightly negative. This is an affectation of integerizing numbers when the distribution curve has a non-zero skew. It should not occur if we did not integerize TCs.

What I have found is that the average of all pre-round running counts always tends to be negative. For example, using HiLo as an example, if a billion rounds were simmed using basic strategy and each pre-round running count was recorded then at the end of the sim the average of these billion pre-round running counts will be less than zero.

By their nature, running counts are always integerized. Since true counts are computed from running counts if true counts are rounded say to the nearest .1 rather than integerized then the rounded true count average will also be negative.

If true counts are intgerized like with flooring then that adds skewing.

It seems kind of counter intuitive that average HiLo running count would be something other than zero. However the fact that it is being measured pre-round in a dealt game of blackjack (especially with a cut card) may seem to make it so.

 

[Renzey]

I ran three sims, 6D, S17, DAS, HiLo, Sweet16.

1. 75% penetration
2. 25% penetration
3. 75% penetration., but stop counting after 25% and assume the same TC through the remainder

SCORES
1. 15.4
2. -7.1
3. -0.8

Now, let’s look at the win rates. I have added sim 4 -- Basic strategy.
1. $9.38
2. -$2.27
3. -$1.00
4. -$2.83

Now, if we play BS for two hours and count for one hour, we make $9.83-2*-$2.83 or $3.72. But, if we use your method for three hours, we lose $3.

Norm,
Out of curiosity, what was the max. bet used in mode #1 -- vs. the max bet used in mode #3? And in harmony with the definition of SCORE, were the ROR's the same for each mode?

 

[QFIT]

Norm,
Out of curiosity, what was the max. bet used in mode #1 -- vs. the max bet used in mode #3? And in harmony with the definition of SCORE, were the ROR's the same for each mode?

Min and max bets were the same in all three sims. 1-15. RoR cannot be the same in harmony with SCORE as SCORE assumes no goal or time limit, meaning that sims #2 and #3, with negative EVs, had 100% RoRs.

 

[assume_R]

It seems kind of counter intuitive that average HiLo running count would be something other than zero. However the fact that it is being measured pre-round in a dealt game of blackjack (especially with a cut card) may seem to make it so.

Indeed it does seem counter-intuitive, and do you have any idea why this is????

 

[muppet]

suppose it's because a hand (and therefore a round) tends to 'end' on a high card more often than on a low card

 

[Renzey]

Min and max bets were the same in all three sims. 1-15. RoR cannot be the same in harmony with SCORE as SCORE assumes no goal or time limit, meaning that sims #2 and #3, with negative EVs, had 100% RoRs.

Norm,
Did mode #3 bet 15 units as often as mode #1? What I'm driving at is, did both betting modes have similar hourly SD's and a similar "average" bet size? Or did they both bet 15 units at say, +4 TC, making mode #3's average bet size at advantageous counts much smaller than that of mode #1?

 

[QFIT]

Norm,
Did mode #3 bet 15 units as often as mode #1? What I'm driving at is, did both betting modes have similar hourly SD's and a similar "average" bet size? Or did they both bet 15 units at say, +4 TC, making mode #3's average bet size at advantageous counts much smaller than that of mode #1?

No, of course not. That is the problem with front-counting. You are simply not able to identify that many opportunities to make the max bet. If you rigged the sim to make the same number of max bets, you would be grossly overbetting the front count or underbetting HiLo. When the Speed Count sims were run, the sims were rigged to give the same S.D. In doing so, the HiLo sim was increasing bets at a negative EV. No wonder their sims made SC look good in comparison. You can't compare systems in this manner. Each system must be simmed to its best ability. You can't cripple one system, as the Speed Count sims did. (The OPP sims used a similar technique.) The front-counting sim had a lower average bet size because the count cannot identify many instances where it is reasonable to increase the bet.

 

[Renzey]

No, of course not. That is the problem with front-counting. You are simply not able to identify that many opportunities to make the max bet. Each system must be simmed to its best ability.

I think that betting a front counting method to its most effective ability would be to reach its max bet at around +2 TC, since you'll very rarely see anything higher in such an early part of the shoe. It of course should not be 15 units, but probably 6-to-8 units. Yes, it would constitute overbetting, but would get multiple units out there whenever you're playing with a "heavy" shoe.
Doing this, in conjunction with wonging out at the front count checkpoint on negative RC's, along with betting 1 unit the rest of the way on near-neutral front count RC's, I'm pretty sure would yield a positive net EV.

This I say from calculating RC frequencies at a given front count check point and applying the "zero unit/1 unit/4unit/7 unit" bets the rest of the way based on those frequencies, and subtracting the expected loss in the initial part of each shoe.
I then do my best to simulate the same, although my dated simulation program can't replicate that strategy precisely. Being the simulation guru, if you can do so such that the hourly SD's would be the same as counting all the way thru and spreading 1-to-15, it would shed more comparative light on the concept.

 

[QFIT]

If you add Wonging, and reduce the count at which you max bet (increasing risk), yes you can manage a positive EV. But, you would also substantially increase EV if you add Wonging and continue the count. You would still be better off counting one-third of shoes, and using BS 2/3rds.

Edit: For that matter, if you add Wonging, you are now counting more than one-third of the rounds.

 

[sagefr0g]

for what it's worth (probably absolutely nothing, lol ) the images below contain some results where i was trying to fool around (from about a year or so ago) with an excel shuffle program regarding this sort of stuff.
lmao to tell the truth i can't fully make sense of it now, just throwing it in there for everyones amusement i guess.

 

[Renzey]

If you add Wonging, and reduce the count at which you max bet (increasing risk), yes you can manage a positive EV. You would still be better off counting one-third of shoes, and using BS 2/3rds.

Are you saying that your net positive EV for each shoe would still be less than $1.24 (since using modes #1 and #4 would yield a combined average of $1.24 per shoe?)

For that matter, if you add Wonging you are now counting more than one-third of the rounds.

You made me stop and think there, and technically I see that is correct.

 

[QFIT]

You're now comparing one method with Wonging and the other without. And one counting one-third of the time and the other more often. The comparison makes no sense to me.

 

[Renzey]

You're now comparing one method with Wonging and the other without. And one counting one-third of the time and the other more often. The comparison makes no sense to me.

Norm,
The sense of it was to offer the original poster a viable and productive way to fill his request, which was to count only part of the time. He sounds like a recreational level player who's looking for a simple step over the threshhold from negative EV to positive EV, while still enjoying the game and not working too hard.
Counting even one-third of the shoes all the way thru requires more than a recreational level of sustained concentration, a very aggressive betting spread, and a larger bankroll.
Counting all the shoes thru the first one-third or so requires less focus, fewer index plays, a more palatable betting spread and probably a smaller bankroll. It's not so much a moneymaker as a "satisfier" with a minimum of extra effort.

As always, your simming expertise is appreciated. You can always be counted on to impart a keen point of view here.

 

[QFIT]

I see the exact opposite. Counting one-third of the way requires a much more aggressive betting ramp, higher bets, and larger bankroll or higher risk due to the very low EV and inaccurate info. Weak systems require risky betting to create a positive EV.

OTOH, counting one-third of the shoes means that you can lower the bets during the non-counting portions and are betting with far more accuracy when the big bets are out.

 

[k_c]

Indeed it does seem counter-intuitive, and do you have any idea why this is????

I would guess that when positive cards appear in the course of a blackjack game they tend to end the round for the hand to which they are dealt in a significant percentage of the cases.

Examples: player or dealer blackjack ends round for that hand with no opportunity for balancing low cards to be dealt; pat hands like player T-T v dealer T-T - again no balancing low cards can be dealt; bust cards, which are most likely to be tens and cannot be 1,2,3,4,5 end the round for either player or dealer with no opportunity for balancing low cards.

I think that if you were to simply burn cards and stop at random points then average running count would be zero for HiLo but the side condition of dealing cards in a blackjack game seems to make the pre-round HiLo average running count negative maybe because the stopping points aren't completely random.

 

[QFIT]

I then do my best to simulate the same, although my dated simulation program can't replicate that strategy precisely.

I used a trick to simulate this with multi-betting strategies. Just realized I have a "Stop counting at n" option that simulates this exactly. Getting old.

 

[assume_R]

thanks, k_c that makes sense that there is a higher probability of high cards ending a hand, which would make the average pre-round RC slightly negative. keen insight there, buddy!

 

[Fun_at_21]

Thanks qfit and renzey. You're right that the heart of my original post was in basically exploring how card counting "elements" might be vaguely/infrequently used to just help reduce a quality BS player's disadvantage.

Although I should probably highlight that my curiosities with this are not with any hope of gaining a true positive EV. Although a dead even game is fair in my eyes. Even a true 0% goal is not precisely necessary either.

The main reasons I look to use just mild card awareness benefits to help my disadvantage rather than totally vanquish it by doing the hard work, is because I prefer subdued betting patterns rather than having to do outlandish spreads. I don't have a problem "raising" bets if it's in the realm of minor raises if I'm aware of an advantage. But I just like/prefer making drastic spreads or "big bets" (even if that's supposed to be the route of overcoming the disadvantage).

So besides playing BS as proper as possible and even using some of the composition dependant plays to shave a few additional points closer to the even mark, I'm really just exploring which "non-work" card awareness methods might best help to just knock off points even further even if I'm NOT raising my bets properly high.

In essence, my "style" and preference is:

Playing normal BS as much as possible

Flat betting (on average) as much as possible

Only raising bets "insuspiciously" and typically never more than 1-2 extra units

Playing more to not lose (playing defensively) rather than bet wildly (and properly) to get an outright edge (I'm big on betting the minimum at all times if I "know" that I'm in a definite disadvantage situation)

Under such a content and conservative style as above, I don't expect ANY card awareness methods to be of enormous help. Yet I figure even slight bet raises in positive situations can only help get closet to that even mark rather than farther away. But I don't know if the "card quantity" method I dabble with best succeeds in this? Or perhaps Dubey's methods of playing off previous hand types? Or perhaps a form of an Ace/five method (if I didn't have to constantly or perfectly use it on all shoes)?

P.S. I meant I DON'T like making big bets or spreads in that last sentence of the third paragraph I wrote above. It looks like I left out that word by mistake... LOL