Is it true?

Sonny

Well-Known Member
#2
That sounds pretty far off, although it depends on how long his sessions are. For a 1-hour session a decent counter will have around a 52% chance of being ahead. If he plays 10-hour sessions then he will be closer to 3-out-of-5 winning sessions.

-Sonny-
 
#3
FUHGEDABOUDIT said:
I read or heard somewhere that a good card counter should win two out of three sessions. True?
Don't know about that. First 5 years I NEVER had a losing year. The last 2 nothing but downhill. So figure the math? The only thing that matters to me is the bottom line and mine is still a plus! I remember I had one streak of 18 winning sessions in a row. The losing streaks you want to forget. 60 months of + minus 24 months of - = + 36 months = 71.4% of the time a +.
 
Last edited:

toastblows

Well-Known Member
#4
InPlay said:
Don't know about that. First 5 years I NEVER losing had a year. The last 2 nothing but downhill. So figure the math? The only thing that matters to me is the bottom line and mine is still a plus! I remember I had one streak of 18 winning sessions in a row. The losing streaks you want to forget.
no kidding, because 1 losing session usually wipes out about 10 winning sessions....:laugh:
 
#5
toastblows said:
no kidding, because 1 losing session usually wipes out about 10 winning sessions....:laugh:
Depends on how long those session are. When I first started I used to play to make a days wages. Short and Sweet. Last 2 years have been longer sessions maybe thats why I am losing? Seems like I am up always at the start then its downhill from there. I used to treat it as a business before.
 

BJinNJ

Well-Known Member
#6
I read that too, recently.

Let me see if I can find the reference.
No guarantees, but I did read it.

I thought it was in Bluebook II or Blackbelt, but maybe not.
When I read 2 of 3, I presumed it was an approximation.
Maybe 64-65% is closer.

BJinNJ :cool:
 

BJinNJ

Well-Known Member
#7
Couldn't find a reference

I skimmed Bluebook II but the only thing I could find was a small
table in the Kiss II/III section comparing 100 hr. win rates for
various counts including Kiss, KO and Hi/Lo. These win rates
were all in the 60-65% range.(nearly 2 of 3)

I also went through Blackbelt, without success. I also recently
referred to Million Dollar BJ, so I looked there, too. No luck.

Then I tried Gamemaster's BJ school. No luck.
I also went thru BJ Blueprint. No luck.

It just seems to me that I read, somewhere, that about 2 out of
3 sessions should be winners, or you need to re-evaluate the
games you choose to play, your skills, etc.

Maybe it was just an article somewhere. I doubt that it was in
BJA3, which is the book I'm currently reading.

If I see it again, I'll definitely make a note and try to post it here.
There are alot of little details like this that slip thru the cracks, while
engrossed in reading about this game.

BJinNJ :cool:
 

SystemsTrader

Well-Known Member
#8
FUHGEDABOUDIT said:
I read or heard somewhere that a good card counter should win two out of three sessions. True?
I would think that's about right. I believe it was Schlesinger who said, but don't quote me on this that a counter should win somewhere between 55% and 65% of his sessions. I guess it also comes down to how you define a session? For myself my solo play has yielded a 66% session win rate which is right on the 2 out of 3 mark and my team play results are at 57%.
 

ihate17

Well-Known Member
#10
what is a session?

If I take 30 years of AP and count it as 30 sessions, then I am winning 77% of sessions. If I took the first 5 years and throw them away, calling it my long term learning curve, then I am winning 84% of my sessions.

Without counting up every single (session), if 30 minutes or several hours, my best guestimate would be something in the area between 55%-60%, but I am not sure that these numbers mean anything, as I note that cardcounters who are still learning tend to have long sessions with big losses quite often which can destroy the wins of many short sessions.

ihate17
 

Guynoire

Well-Known Member
#11
Win Limits

Many people, either consciously or unconsciously, set win limits on their sessions. This has no overall effect on the results other than limiting the number of hours played but will shift the distribution of winning to losing sessions. Setting win limits will increase the proportion of winning sessions but these winning sessions will typically be smaller than the losing ones.
 

Kasi

Well-Known Member
#13
QFIT said:
Interesting question. I put up an answer at http://www.blackjackincolor.com/blackjackvariance3.htm. Basically, Sonny's numbers are right.
I was wondering if the BJA formula you referred to on your link above was Formula 1 for goal given a time constraint.

Even if it wasn't, lol, is there anywhere you know of I could see a step-by-step calc with an example? Is it solvable in a spreadsheet? I can't even find a function for summing an infinite series.

No matter what I do the second part always comes up zero and it's been a stone in my shoe for I don't know how long.

Obviously, I'm clueless, maybe not completly but mostly, lol, on that infinite series stuff lol.
 

QFIT

Well-Known Member
#14
I would not suggest that anyone attempt to use formula 1 by hand. But this is much simpler than formula 1:

1 - normp((b - t * mu) / (sigma * t ^ 0.5))


where:
mu = win rate
sigma = std. dev.
b = bankroll
t = time
normp = Cumulative normal probability distribution function
 

BJinNJ

Well-Known Member
#15
I found it..

at least the reference I recently read.

Page 22 of BJA3

"You're planning a one week, 40-hour assault on the A.C. 6 deck game?
Well, you should be a winner two out of three times, but don't expect
any more."

BJinNJ :cool:
 

FLASH1296

Well-Known Member
#17
yes, sort of

A talented Card Counter will have between 60 and 70 percent winning sessions.

The exact figure depends on several variables, i.e. rules, penetration, betting spread.

The wins will tend to be larger than the losses as well.

Anyone who claims that they nearly always win, or words to that effect is either a liar or someone who has only played a few dozen sessions and/or keeps no records.
 

Kasi

Well-Known Member
#18
QFIT said:
I would not suggest that anyone attempt to use formula 1 by hand. But this is much simpler than formula 1:

1 - normp((b - t * mu) / (sigma * t ^ 0.5))


where:
mu = win rate
sigma = std. dev.
b = bankroll
t = time
normp = Cumulative normal probability distribution function
Thanks so much Norm.

So it's kind of maybe a variation of the "chances of being ahead at any point during a specified number of hours of play by at least $1" but assumes you you start with a 0 unit bankroll?

Man, if they had this formula around, why even invent the other one? :)

Anyway, thanks again - I'll give it a shot.
 

Kasi

Well-Known Member
#19
Guynoire said:
Many people, either consciously or unconsciously, set win limits on their sessions. This has no overall effect on the results other than limiting the number of hours played but will shift the distribution of winning to losing sessions. Setting win limits will increase the proportion of winning sessions but these winning sessions will typically be smaller than the losing ones.
I think, like System said it depends greatly on how you define a session.

Some might choose to define a session by time too rather than win amount. Maybe 1-2 hours because that's all the longer he wants to play in one place. When he wns, he wins. When he doesn't , he doesn't.

So a guy who goes to Vegas every week and plays 6 hours a day in 3 2-hour sessions won't come out ahead in maybe, like Sonny says, but, say, 53% of the sessions.

Then he goes back home and calls a session a week's play of 42 hours a "session". Now he might be winning 65% of his weekly sessions.

Then, after 40 years of playing, he calls a year a session and can now likely brag to his friends he wins 95% of his sessions.

If you do set win limits, don't leave in a good shoe!
 
Top