Calculating the TC

[nightspirit]

There is a nice study about this subject on qfit's site. Card Counting Edge by True Count

 

[jack.jackson]

Betting indices? I’ve never heard of such a thing. Your bet spread is based on the game you are playing. If you play another game then you adjust your bets to match the new game. I’ve never heard of indices that will adjust your bets for you. Would you have to learn indices for each individual rule? Probably you would need to know indices for each rule combination since the advantage of certain rules is dependent upon other rules as well. And would each index change based on the penetration of the game? It sounds like there could be literally thousands of indices for that.



That depends on what you mean by "accurate." What are you using the count to measure?

You don't have to worry about accuracy if your bets are based on the advantage at each TC and not the TC itself. The TCs increase linearly but your advantage doesn’t. That is why it is not correct to assume that a TC of +8 is twice as good as a TC of +4, or that the advantage at a TC of +3 is halfway between the advantage at +2 and +4. The two will slowly drift apart, but that’s why you use your advantage to calculate your bets and not the TC. The TC is just an indicator.

-Sonny-

Well i had a thorough explanation of it. Then my keyboard fell on the floor and erased everything
So ill just try to do a brief summary.

 Good rules            Bad rules

 1D+2+3                +3+4

 2D+4+7                +7+10

 3D+7+10              +11+15

 4D+10+15            +15+21    

 5D+15+21            +21+27

 6D +21+27           +27+34

The first set of indices are for 2 unit bets. The second set of indices is for 4 units and above. Good rules are as follows DA2,S17, Bad rules are D10,11 only H17.

The interesting thing is, is that carlson says the betting indices are contingent of the rules in play. Not the house advantage? For this doesnt make sense to me.

 

[Sonny]

The first set of indices are for 2 unit bets. The second set of indices is for 4 units and above.

But that just tells you when to start raising your bets and when to hit your max bet. It doesn’t tell you how much to bet or what kind of ramp to use. You would be using different spreads for all of those games so that information wouldn’t help you very much. It also doesn’t consider penetration as a factor.

And are those running count indices? I don't understand.

-Sonny-

 

[EyeHeartHalves]

I can tell you what I do (It's easy):

I always truncate down to the nearest whole number of decks left unseen. This will do a few things for your game--some good; some bad:

1) It will be easier on your calculations. In your example, 3 and 1/3 would simply become 3.

2) Your win rate will increase slightly as you will be betting more money at counts that normally would have been figured to have a much lower but still advantageous TC. For example, instead of dividing a RC of 12 by 3.9 decks for a TC of 3.1, my TC would be just 4.

3) Your variance will increase because you're theoretically betting too much at TCs which are derived from a heavily truncated denominator.

4) You will make slightly incorrect (or "sometimes incorrect") strategy point decisions.

(Wow, maybe my system sucks? Oh, wait. I forgot. I make a lot of money.)

 

[jack.jackson]

l

But that just tells you when to start raising your bets and when to hit your max bet. It doesn’t tell you how much to bet or what kind of ramp to use. You would be using different spreads for all of those games so that information wouldn’t help you very much. It also doesn’t consider penetration as a factor.

And are those running count indices?

-Sonny-

And are those running count indices?

Yes,aka betting indices.

But that just tells you when to start raising your bets and when to hit your max bet

Suppose, im using the bad rules indices and with 3D remaining and the RC jumps from 0 to +16 Note: that even though +16 is high enough to make a max bet of lets say 40X40(1:8) spread. I would only ramp to 20$. Now regardless, if i win or lose and the count remains at +16 im now goin 20X20. Win or lose this and the count still remains high enough to justify a max bet, Then and only then do i go max bet. Trust me ive experimented with direct proportional betting(meaning goin straight to 40X40) and was absolutely mutilated. Time after time.
Its an essential part of the betting scheme(aka ramp)
to never increase your bet by more than two units at a time. I dont know much about other systems and their betting systems but with the A02 you'll only be making 4 different bets 1. The min(10$) 2.The bump(20$)3. The avg. bet(20X20) and of course the max bet(40X40)

The first indice is for the bump. The second one is for the avg bet. These are just a alternative to caculating the TC for betting everytime. Once youve memorized them betting becomes much easier.

Also note: there are a couple other exceptions to the betting strategy.

Have you read BJFB sonny?

 

[Kasi]

I always truncate down to the nearest whole number of decks left unseen. This will do a few things for your game--some good; some bad:
2) Your win rate will increase slightly as you will be betting more money at counts that normally would have been figured to have a much lower but still advantageous TC. For example, instead of dividing a RC of 12 by 3.9 decks for a TC of 3.1, my TC would be just 4.

3) Your variance will increase because you're theoretically betting too much at TCs which are derived from a heavily truncated denominator.

4) You will make slightly incorrect (or "sometimes incorrect") strategy point decisions.

I'm just going to ask you what would be your alternative if you do not truncate. I assume it might be rounding or, perhaps, flooring.

Sticking just with positive counts for now, I'm thinking you might otherwise round less than TC 1.5 to 0.5 as 1. Etc.

It's intuitive that the lower TC's will occur more often but give rise to a smaller average advantage.

So, while u may bet your TC +2 bet from 2.5 to 3.49 rounding, and a sim may reveal a different win rate for a TC+2 than it would if you had truncated fom 3 to 3.99, I don't think there'd be much difference overall, only in the various intervals.

And, hopefully, one knows exactly what one's betting system is based on in the first place so one knows to round or not for both betting and strategy deaprtures.

 

[Sonny]

Suppose, im using the bad rules indices and with 3D remaining and the RC jumps from 0 to +16 Note: that even though +16 is high enough to make a max bet of lets say 40X40(1:8) spread. I would only ramp to 20$. Now regardless, if i win or lose and the count remains at +16 im now goin 20X20. Win or lose this and the count still remains high enough to justify a max bet, Then and only then do i go max bet.

So you’re betting too small at first, then making your max bets a few hands too late. That doesn’t sound like a good plan to me. You’re hurting your profits by not betting enough during your advantage and by missing the few opportunities you have to make those big bets. A max bet situation is so rare that even missing one or two opportunities an hour can make a huge difference. Essentially, this system is like using a smaller bet spread with occasional overbets. Your profit is smaller and your risk is greater.

Trust me ive experimented with direct proportional betting(meaning goin straight to 40X40) and was absolutely mutilated. Time after time.

But were your mutilations unexpected? Were they significantly above your expected standard deviation? Was your ROR too high at the time? A big fluctuation is completely normal. Your only concern should be if it is unexpectedly large.

We’ve all been mutilated a few times but that doesn’t mean that proportional betting doesn’t work. The occasional mutilations have nothing to do with the betting system, it has to do with the game itself. There is no betting system in the world that can eliminate the variance of the game. The whole point of proportional betting is to minimize the fluctuations while maximizing the profit. Any system that improves one aspect will compromise the other.

If you want less variance then use a smaller unit. If you want more profit then you’ll need to accept more risk. Proportional betting is the “happy medium” that gives you the best of both worlds.

Its an essential part of the betting scheme(aka ramp) to never increase your bet by more than two units at a time.

But that’s exactly where the problem is. You won’t be earning as much from those positive counts if you’re not betting enough money on them. This is similar to Schelsinger’s “parlay” style of play. That system might be great for cover, but it is very bad for EV.

The whole point in designing a bet spread is to size your bets based on your current advantage. If you aren’t betting the correct amount of money then your system will not be as effective. If you are varying the amount of money you bet at the same TC then you are occasionally overbetting and occasionally underbetting. Your profit goes down and your risk goes up.

I dont know much about other systems and their betting systems but with the A02 you'll only be making 4 different bets 1. The min(10$) 2.The bump(20$)3. The avg. bet(20X20) and of course the max bet(40X40)

But what about those counts where you should be making a $15 or $30 bet? And if you’re using a 1:16 spread you don’t want to break that up into only four bets because you’ll be jumping from $10 to $40 to $80 to $120. That’s not good for cover or EV. It also has you raising your bet by more than two units, which you said was not allowed.

Most systems will have five or six different bet sizes that make up their bet spread. Using less will create problems like this.

The first indice is for the bump. The second one is for the avg bet. These are just a alternative to caculating the TC for betting everytime.

I still don’t understand how this works. For example:

Good Rules
6D +21+27

Does that mean you wait until the RC is 21 points above the usual spot to “bump” your bet? Or do you wait for a running count of +21 at any deck level before you increase your bets? Wouldn’t a TC adjustment be more accurate? And don’t you still have to calculate the TC every time to find out what bet level you’re at?

Have you read BJFB sonny?

It’s been a few years. Maybe I should dust it off and read through it because none of this makes any sense to me.

-Sonny-

 

[EyeHeartHalves]

I don't disagree - Kasi

I was just putting my two cents in for the first post of the thread. He seemed unsure as to whether he should divide by fractions or round. Then you responded with "round" in the second post. This is fine.

However, in single parameter balanced counts (because I don't know about much else when it comes to systems), my vote would be to "truncate."

 

[SilentBob420BMFJ]

ctrl+z? im 99% sure it was more complex than just that, but in the slim chance you didnt know ctrl+z, now you know!