Bet Spreads

[sagefr0g]

It's just the same thing as Sonny's sheet, I hope lol, but calcs a few more things like avg bet units, some trip roll stuff, goal reach stuff with no time limit, N0, SD ranges, whatever, in case anyone finds it relevant.

You can change the red cells after you input whatever in the freq, adv, SD columns. The spread is units - change roll dollars and roll $unit below in it.

What this version doesn't do is spreading to multiple hands and might not work for anything other than "play-all" scenarios.

Oh yeah - this one uses SD for input whereas Sonny's uses variance.

so Kasi just was wondering could your sheet be tinkered with using Sonny's methods to get it to go by multiple hands?

maybe i'm just confused but that covariance thing seems as if it's awfully tricky to introduce into sims. i mean i think like Norm's CVCX (my earlier version)can do it but i think it only bets all rounds in play as multiple hands. but then he has a newer version that i think you can change the number of hands played over differant TC's where i think he called it in experimental phase. i'm gonna get that version sooner or later. i don't have cvdata so i don't know what it does in that regard.
i just know that with Sonny's sheet the yeild differances can be pretty dramatic.

 

[matt21]

Hi, I am brand new to the forum and it looks like the forums contain a wealth of information (i have spent a few hours browsing various threads). It all looks very excellent!!

I have a small question regarding bet variation.

Up to this point I have been playing a 1-12 spread i.e. 1,2,4,8,10,12 for TC +1,+2,+3,+4,+5,+6. I have an (arbitrary) bankroll of 600 units (taken from business profits) plus 251 units of winnings = 851.

I am thinking of changing to a 1-10 spread in the order of 1,3,6,10,10,10 - doing some calculations (using results from simulations from bjstats.com) in xls, seems that my expected unit return per hour would be about the same, but my gut feel is that the change in spread would adversely impact on my variance and thus RoR?

Background on my game:
I have been playing for just over six months:
4D or 6D, S17, DAS No surrender 70-80% pen
count using Hi-Low Method with I18 indeces
Hours played: 223 (using either $5-$60 or ($5)$10-$120 standard unit)
Status: +251 units (1.13 units/hour)
Hands played: approx 15,000 avg 70 hands/hr (often engage in heads-on play)

Matt

Well, after several days of hard work, investing in CVBJ and CVD, examining posts on the forum, compiling data models in xls, extracting data from simulations, learning about the Kelly ratio, experimenting with multi-hand play, and compiling index plays for ENHC .... i feel i have made some good progress.

i think this time was very valuable for me to step back and assess how i have played thus far and have learned the following:
- using my existing bet varation (that i have played with over the last 7 months) i should have only expected a return of some 0.5 units per hour (so I count myself damn lucky that i have actually fallen into positive variance averaging 1.1 units) - however i did not have clear expectations prior to this point
- pen and the # of players at my table are very important drivers of EV/hour so i will continue searching for deserted tables as much as possible
- generally my set of 4D casinos provide a higher expected return than my set of 6D casinos taking into account
- i spotted several BS and index errors that i had been unaware of, due to using standard BS tables for ENHC games - i have adjusted my BS tables and index plays accordingly
- by adjusting my betting pattern, primarily spreading out to 2-3 hands as the count increases, i am able to increase my EV while cutting my ROR in half!
- the standard deviation associated with my expected return (using 68% and 95% confidence intervals) is significantly higher than what i had expected using "gut feel" - even when using horizons of 300-400 hours of playing time
- given these standard deviation values, i will need to seriously whether black jack is a good business opportunity when compared to other business and investment opportunities
- overall i now have a much better grasp and understanding of the EV and standard deviation principles as they apply to black-jack and this has increased my overall confidence

thanks for all the helpful contributions from various posters during the last few days.

 

[Kasi]

so Kasi just was wondering could your sheet be tinkered with using Sonny's methods to get it to go by multiple hands?

Sure it could. I'd use exactly the same methodology Sonny uses. It's just that Sonny uses an estimate for total variance and my question is how/if can one figure that out more exactly.

Obviously his constant 1.83 total variance for 2 hands, if that's what it was, is an estimate based on some assumption of what the co-variance for that particular game is. Whatever co-variance one assumes, one can apply the formula for more than 2 hands too.

And it's a good estimate, I think, and I doubt if "more exact" would matter that much anyway.

Just like his variance of 1.33 at each TC for 1 hand is also an estimate. It has to be since it's constant.

I suspect that neither variance or co-variance remains constant at each TC. I'd guess both Var and Co-var would be higher at -TC4 than +TC4 kind of thing.

It's probably another "tempest in a teapot" kind of question.

I mean I think co-variance would be less in a D10 game vs a DAS game.

So, I'd guess Sonny is assuming a co-variance of maybe 0.5 or so. Assume higher and lower at each TC and see what changes and if it's significant to win rate, bet-size or whatever.

Also, keep in mind that this spreading-to-multiple-hands stuff may negatively effect trip rolls, especially for shorter trips since what Sonny is assuming is the same win goal playing forever but, on trips, this can change when you consider number of finite hands played. One needs more roll on a trip when spreading to keep same ROR. I think lol.

So, are you telling me that CVCX or CVDATA, whichever one does this stuff, that, if you plug in spreading to 2 hands * x units somewhere compared to 1* x units, let's say at positive counts lol, the SD column doesn't "magically" change from, say, maybe 1.15 or so to 1.35 or 1.37 or so??

 

[Kasi]

- given these standard deviation values, i will need to seriously whether black jack is a good business opportunity when compared to other business and investment opportunities...
- overall i now have a much better grasp and understanding of the EV and standard deviation principles as they apply to black-jack and this has increased my overall confidence

Wow.

Thanks Matt.

This is like an epiphany for me.

You've responded to every doubt I ever had, done absolutely everything I think everyone should do.

Not that it matters what I think because you're the one who did it.

I got too much "I" in this so I apologise. You'll never hear from me again. I'll answer questions if I can but I got no worries, mate.

I join you in thanking the people here for giving you the tools to make your own decisions about how to invest your own money.

My last question of you - would you consider your investment in CVCX and CVD worthwhile?

 

[sagefr0g]

.....

So, are you telling me that CVCX or CVDATA, whichever one does this stuff, that, if you plug in spreading to 2 hands * x units somewhere compared to 1* x units, let's say at positive counts lol, the SD column doesn't "magically" change from, say, maybe 1.15 or so to 1.35 or 1.37 or so??

good question IMHO.
i'll check it out and report back.
machin schnell here. lol

edit:
but wait. for equivalency purposes maybe it shouldn't be 1*x units and then 2*x units. point being isn't it when you play two hands you would go to an amount on each hand that makes the total bet of the two something like 150% of the one hand bet?

 

[sagefr0g]

Quote:

Originally Posted by Kasi
.....

So, are you telling me that CVCX or CVDATA, whichever one does this stuff, that, if you plug in spreading to 2 hands * x units somewhere compared to 1* x units, let's say at positive counts lol, the SD column doesn't "magically" change from, say, maybe 1.15 or so to 1.35 or 1.37 or so??

i guess it doesn't change? or maybe i screwed something up?
well irregardless of my question about changing the size of x going from one hand to two hands here is maybe close sims to your question above:

edit: ok hold on check the third image the custom bets. maybe more what your askin. problem is it doesn't show the stdev for the various tc's...... i'd have to try and fudge some other sim to get the bets to match the custom and then we could see the std dev.........

 

[Kasi]

good question IMHO.
i'll check it out and report back.
machin schnell here. lol

edit:
but wait. for equivalency purposes maybe it shouldn't be 1*x units and then 2*x units. point being isn't it when you play two hands you would go to an amount on each hand that makes the total bet of the two something like 150% of the one hand bet?

I hope "machin schnell" means something like "now" lmao. I just saw Cartman on "South Park" today choosing an Adolph Hitler costume for Halloween. Funny stuff.

To your edit - yes.

But in Sonny's sheet, on the last tab, he bets 1 hand at $60 at TC+4 with a variance of 1.33 but 2 hands of $60 at +TC5 and above with a total variance of 1.83 per hand or total variance of twice that per round. At least I assume that 1.83 is per hand, not per round.

Whether that means he only would have bet 1 hand of 8 units otherwise at +TC5 and above, maybe at a total variance per hand (and round) of 1.33, compared to 2 hands of 6 units, thereby betting 150% more money when spreading at a total variance of 1.83 per hand, twice that for 2 hands, I'll leave it to you and CV... to decide how EV and ROR changes.

I guess I'm a little fuzzy right now but I think that 1.83 variance means per hand when playing 2 hands so double that for total variance of 3.66 per round?

I could be way way off here lol.

I throw this cr*p out hoping someone who knows what they are doing replies.

But Sonny, QFIT, Ken, etc never reply lmao.

Meanwhile "Sieg heil" or "machin schnell". I'm so ignorant I don't know what either means. I can do French and Greek for some reason though, lmao.

 

[sagefr0g]

I hope "machin schnell" means something like "now" lmao. I just saw Cartman on "South Park" today choosing an Adolph Hitler costume for Halloween. Funny stuff.

To your edit - yes.

But in Sonny's sheet, on the last tab, he bets 1 hand at $60 at TC+4 with a variance of 1.33 but 2 hands of $60 at +TC5 and above with a total variance of 1.83 per hand or total variance of twice that per round. At least I assume that 1.83 is per hand, not per round.

Whether that means he only would have bet 1 hand of 8 units otherwise at +TC5 and above, maybe at a total variance per hand (and round) of 1.33, compared to 2 hands of 6 units, thereby betting 150% more money when spreading at a total variance of 1.83 per hand, twice that for 2 hands, I'll leave it to you and CV... to decide how EV and ROR changes.

I guess I'm a little fuzzy right now but I think that 1.83 variance means per hand when playing 2 hands so double that for total variance of 3.66 per round?

I could be way way off here lol.

I throw this cr*p out hoping someone who knows what they are doing replies.

But Sonny, QFIT, Ken, etc never reply lmao.

Meanwhile "Sieg heil" or "machin schnell". I'm so ignorant I don't know what either means. I can do French and Greek for some reason though, lmao.

lol check my edit. i gotta get some :sleep:
http://www.blackjackinfo.com/bb/showpost.php?p=93038&postcount=26

edit: i think in my elephant alter life i remmember you maybe asked QFIT about if the std dev didn't appear to change? and then i did also. or maybe it was something else. but unfortunately he must have missed it and no response.

 

[matt21]

Wow.
Thanks Matt.
This is like an epiphany for me.
You've responded to every doubt I ever had, done absolutely everything I think everyone should do.
Not that it matters what I think because you're the one who did it.
I got too much "I" in this so I apologise. You'll never hear from me again. I'll answer questions if I can but I got no worries, mate.
I join you in thanking the people here for giving you the tools to make your own decisions about how to invest your own money.
My last question of you - would you consider your investment in CVCX and CVD worthwhile?

Hi Kasi, i am a little puzzled by your post. Are you saying that it's good that I evaluated card-counting as an investment opportunity for myself? And why will I never hear from you again? It would be great if you could clarify what you wrote earlier so hope to hear from you on this soon!!

Investment in CVCX/CVD - this cost only $200 as a package and i was able to download it and unlock immediately for use. the software is very good for doing analysis, running simulations and completing drills - a very big time saver and valuable tool. So i think that investment was definitely worthwhile.
$200 very well spent particularly if you are going to spend a lot of time playing.

 

[blackjack avenger]

A Little of This, A Little of That

According to Wong and Sch has it in his book

For 2 hands you bet .77 of your one hand bet.

$100 bet becomes 2 hands of $77 or $75.

Considering the amount of rounding we have to do in the real world, wether it is acually $78 or $74 does not seem that important.

 

[sagefr0g]

According to Wong and Sch has it in his book

For 2 hands you bet .77 of your one hand bet.

$100 bet becomes 2 hands of $77 or $75.

Considering the amount of rounding we have to do in the real world, wether it is acually $78 or $74 does not seem that important.

you wouldn't happen to know how standard deviation per tc behaves say if you bet some amount one hand only as opposed to betting that amount on each of two hands would you? maybe know of a formula that can figure it?
a formula that could be used in a spread sheet.
what would that be covariance if your playing two hands.

 

[blackjack avenger]

Perhaps

you wouldn't happen to know how standard deviation per tc behaves say if you bet some amount one hand only as opposed to betting that amount on each of two hands would you? maybe know of a formula that can figure it?
a formula that could be used in a spread sheet.
what would that be covariance if your playing two hands.

From Bryce Carlson:
Total variance of 2 hands is 3.52 (1.8762 SD). The variance for one hand is 1.76. This was for a game with a 2% adv.

If you want to spread a $100 bet over 2 hands then that would be equivalent to cutting your bet by a third. Also, I imagine cutting your ROR.
$100 = DBL $75
$66 = DBL $50

 

[N&B]

The OP's Bet ramp could be altered a little as 1-2-4-2x3-2x5. This spread is only 1-10, and bumps up a bit better. If your forced to play single-handed, 1-2-4-7-10 is fine.

 

[Kasi]

edit: i think in my elephant alter life i remmember you maybe asked QFIT about if the std dev didn't appear to change? and then i did also. or maybe it was something else. but unfortunately he must have missed it and no response.

I think maybe you're right - maybe he said it just calculates it internally lol.

 

[Kasi]

Are you saying that it's good that I evaluated card-counting as an investment opportunity for myself? And why will I never hear from you again?

That's a yes to the first lol.

Of course you'll hear from me if you ask questions and I can help answer them. I just meant it sounded like you have the tools to answer almost all your own questions yourself :grin:

Can't you take a compliment lmao?

Ask anyone - never hearing from me again is GOOD NEWS :grin:

One question always leads to another - ask away lol.

 

[Kasi]

you wouldn't happen to know how standard deviation per tc behaves say if you bet some amount one hand only as opposed to betting that amount on each of two hands would you? maybe know of a formula that can figure it?
a formula that could be used in a spread sheet.
what would that be covariance if your playing two hands.

No, I don't know a formula that can figure out co-variance. I almost sometimes also think maybe one needs to sim and then it is what it is. Just as SD is what it is.

I know it will change based on the rules of the game.

Yes, there is a formula to calculate the variance of 2 hands but you need to know the co-variance first to apply it. (I think lol.)

Apparently 0.5 isn't a bad estimate of co-variance to use. Wong says it's .43 for a D10 game, .48 for a shoe game (I think lol).

Katrina actually lists co-variance at each TC in her book. It varied alot from low TC's to high TC's. With lots of indexes, it actually reversed itself and went from low to high as opposed high to low with BS.

I haven't worked on your CVCX sims yet but maybe I can work backwards from the 2-hand numbers to the 1-hand numbers and see if the implied co-variance at that count actually is changing. The formula is in Sonny's sheet.

 

[Kasi]

From Bryce Carlson:
Total variance of 2 hands is 3.52 (1.8762 SD). The variance for one hand is 1.76. This was for a game with a 2% adv.

Yes, but I believe he also assumed a co-variance of 0.5 to get that?

Had he assumed 0.4 instead, would that have invalidated his point? Would it only have changed the amount to bet?

That's what I thought would happen but I didn't actually do it lol.

Great article though. I liked it alot without understanding it lmao. But I was thinking maybe that stuff down below, all that linear motion stuff, might have something to do with that "chances of halving a roll stuff" which I'm still trying to figure out lmao.

 

[sagefr0g]

No, I don't know a formula that can figure out co-variance. I almost sometimes also think maybe one needs to sim and then it is what it is. Just as SD is what it is.

I know it will change based on the rules of the game.

Yes, there is a formula to calculate the variance of 2 hands but you need to know the co-variance first to apply it. (I think lol.)

Apparently 0.5 isn't a bad estimate of co-variance to use. Wong says it's .43 for a D10 game, .48 for a shoe game (I think lol).

Katrina actually lists co-variance at each TC in her book. It varied alot from low TC's to high TC's. With lots of indexes, it actually reversed itself and went from low to high as opposed high to low with BS.

I haven't worked on your CVCX sims yet but maybe I can work backwards from the 2-hand numbers to the 1-hand numbers and see if the implied co-variance at that count actually is changing. The formula is in Sonny's sheet.

ok here is i think some better comparison sims as it goes like you were talking about one hand of X and then two hands of X.
note: the std dev's per tc don't show as changing while the per hand std dev does change.

 

[blackjack avenger]

One and the Other

ok here is i think some better comparison sims as it goes like you were talking about one hand of X and then two hands of X.
note: the std dev's per tc don't show as changing while the per hand std dev does change.

I think the std per tc doesn't change because that is the std for that true count, for that counting system. It is separate from bets.:joker:

The std per hand changes I believe based on the actual bet, one or two hand.:joker:

 

[blackjack avenger]

Like a Thief in the Night

If you cut the bankroll for the one hand sim by about a third; keeping the same bets, you will have close to the same ror for both sims.