EV per hour

Kasi

Well-Known Member
#21
newb99 said:
My calcs are for a six deck game, play all approach with a 1-8 spread. I've used one of Mr Snyder's tables from Blackbelt in Blackjack (%ages of hands at a given favourable and house edge) as the basis for them. The spreadsheet's nothing fancy, and calcs the wins/losses at each count/edge, based on an entered spread, and tallys the lot up to give a bottom line expected win/loss for "x" number of hands played. It doesn't take into account the effect of pushes (7% of all hands played across all counts? - I might build that in a bit later), and the Std Dev calc is one that is programmed into Excel for calculating it on a single figure (STDEVPA)

So it certainly isn't pin-point accurate, but I'm not sure it needs to be at my level of play. My results do fall into line with what others post and they do provide an instant indication of what is the maximum expected loss for any given amount of play - handy for knowing how much I need to take to the House of Chance so as not to blow out.

If I was playing several times a week with a decent amount behind me I would want something more robust and would invest in a decent sim program. But for the time being Excel, combined with an understanding of the underlying numbers gleaned over the last 8 months or so, suffices.

Newb99
Sounds good to me. Maybe you could look into Powersim too since it's free. I think it does ENHC.

Don't know what Arnold's tables are like or what that Excel function does. Or why you calc win/loss %'s? If you have freq and adv, that's all you need. SD too but just assume 1.15 at all counts - can't be that far off. Like in Sonny's sheet. Can you model after that? Maybe you are?

If you want to compare notes lol feel free to send me the raw data you are using. Or more. Or not. See how close we come to the same conclusion :)
 

Kasi

Well-Known Member
#22
ccibball50 said:
CVData and yes I will definately send you a message about question I might have on the program - the winning per 100 hands was about 60 cents though..
If you have CVDATA why are you asking these questions lol.

Hey, I don't even have it, it sounds like it has so many bells and whistles it could be easily confusing. No doubt will be to me too.

Looking forward to hearing from you.
 

Kasi

Well-Known Member
#23
FLASH1296 said:
Score =15.64 [with a spread of 8-1]
It means that with a $10,000 bankroll and "perfect" play and betting "optimally" you will earn $15.64 per hour on average. Your swings (standard deviation will be over 20 units per hour, (meaning that losing several hundred units in a day's play will not be very remarkable.)]
Actually, for that game, for 1 hours play 1 SD is almost 36 min units. Yet you could play 1000 hands and 1 SD would only be 125 min units, only 110 unit loss compared to starting roll. Losing several hundred units in a day's play actually would be pretty remarkable.

FLASH1296 said:
If your bankroll is $2,500 you can expect an average profit of a whopping $3.91
Actually if your bankroll was $2500 and you used the same $spread as you did with a $10K roll, you'd still make $15.64/hr. But with a higher risk.

If you bet your $2500 optimally, as you did with $10k, then yes that would be your win rate. Except then your min unit would be below table min so you would not be able to bet that way.
 

UK-21

Well-Known Member
#24
Kasi said:
If you want to compare notes lol feel free to send me the raw data you are using. Or more. Or not. See how close we come to the same conclusion :)
I'll PM the spreadsheet over to you shortly and you can test it for reasonableness. Do let me know if you find any glaring deficiencies.

Thanks for taking an interest.

Newb99
 

FLASH1296

Well-Known Member
#25
on risk ...

Software is really not needed.

The more you money you have behind you, the lower your risk of going broke.

The R O R numbers are something that are not "facts on the ground"

They are essentially "theoretical"

They exist only in the world of mathematics, not in the world of your wallet.
 

ccibball50

Well-Known Member
#26
Kasi said:
If you have CVDATA why are you asking these questions lol.

Hey, I don't even have it, it sounds like it has so many bells and whistles it could be easily confusing. No doubt will be to me too.

Looking forward to hearing from you.
Well the reason is I have been to websites and have received different EVs than what I am getting in CVData.
 

UK-21

Well-Known Member
#27
FLASH1296 said:
[Software is really not needed.

The more you money you have behind you, the lower your risk of going broke.

The R O R numbers are something that are not "facts on the ground"

They are essentially "theoretical"

They exist only in the world of mathematics, not in the world of your wallet.
Silly me. So looks like I've been wasting my time analysing the probabilities and calculating how much I need to take to avoid a blow out. And all along I just needed to take loads and loads of money and to forget about the maths . . . .
 

bj bob

Well-Known Member
#28
Agree and disagree

FLASH1296 said:
Software is really not needed.

The more you money you have behind you, the lower your risk of going broke.

The R O R numbers are something that are not "facts on the ground"

They are essentially "theoretical"

They exist only in the world of mathematics, not in the world of your wallet.
R.O.R. calculations can be compared to taking a trip by car. Let's say you're driving from Albuquerque to St. Louis. You just don't head out the door with a $20 bill in your pocket and think you're going to be fine. What you do need to do is at least roughly "ballpark" the driving distance and travel time and then allow for the gas/ food/ room money needed for that trip. You use the "theory" to guide your calculations, but on the other hand it's not really necessary to sim the exact amount to the penny. A quick pencil exercise along with the usual "Murphy's Law" (10-20%) over funding will do the trick nicely. The credit card in the wallet takes care of the rare "doomsday" scenario.
 
Top