House Edge and Noobs

GeorgeD · Aug 11, 2008

It seems like people keep coming here thinking that the house edge "resets" every shoe, table switch, day, week, session, whatever. IOW: many think if they "beat" the edge for some time period that the next period is a whole new ballgame. They think they can always lock in winnings, but they can always recover their losses.

I think this is kind of like a variation of the gamblers fallicy

What (maybe in FAQ's) might easily explain this fault in logic to the average newcomer?

shadroch · Aug 11, 2008

It's hard to explain,because in a way,it does reset. If you have a white coin and a black coin and they are in a pot in equal amounts,you should get them in equal amounts. But if you pull out twenty White ones in a row,then over the next hundred,you'll still get about a fifty-fifty division, not 70-30 Black.Its only over the long run that it will even up.

GeorgeD · Aug 11, 2008

Good point, but you can't exploit that change unless you count the number of white and black you have pulled so far ..... a great analogy, but if you have a billion of each color in the box it will be an ever changing edge that eventually converges to 50/50. No?

shadroch said:
It's hard to explain,because in a way,it does reset. If you have a white coin and a black coin and they are in a pot in equal amounts,you should get them in equal amounts. But if you pull out twenty White ones in a row,then over the next hundred,you'll still get about a fifty-fifty division, not 70-30 Black.Its only over the long run that it will even up.

bj bob · Aug 11, 2008

GeorgeD said:
Good point, but you can't exploit that change unless you count the number of white and black you have pulled so far ..... a great analogy, but if you have a billion of each color in the box it will be an ever changing edge that eventually converges to 50/50. No?

The critical difference here is whether you put the coins back in the pot after each draw or not. I believe Shad is talking about returning them as he goes which would result in the 50:50 conclusion.

sagefr0g · Aug 12, 2008

this is the general blackjack forum so it's ok.

GeorgeD said:
It seems like people keep coming here thinking that the house edge "resets" every shoe, table switch, day, week, session, whatever. IOW: many think if they "beat" the edge for some time period that the next period is a whole new ballgame. They think they can always lock in winnings, but they can always recover their losses.
.....

i think maybe the quality of the shoe does reset every shoe or at least it might and is probably likely to do so. if your counting the shoe is most likely crap i forget what is it on average over the long haul count the shoes up and 75% were crap for you the counter and 25% were juicey? i think those numbers are kind of pie in the sky but maybe those numbers are more accurate for hands played sort of thing. but yeah like you say as far as anyone knows any shoe off the top has some house edge that you expect. definately true if your just playing basic strategy and not counting. well it's true too even if you are counting. you just know that it's in the long haul gonna be a lot of crap and a little juice. and admittedly you never know how it's gonna come at you.
but count me in on the idea if i beat the edge for some time period that the next time period is a whole new ballgame. like for me it's pretty likely i'm playing a weak game from the get go. if i win some money more than the expectation that i'd have in the case for which i was playing a strong game that gives me at least two alternatives. i can now choose to play a stronger game with that money won from the weaker play or i can now risk the money won from the weaker play and continue to play a weak game in hopes of lightening striking twice in the same place. you can take the same tacts in the unfortunate case where you are below expectation.
so but technically i guess your right you can't lock the money in and you can't always recover loss's. it's like what i'm talking about isn't advisable orthodox advantage play. but you can fiddle with those aspects a bit though for the fun of it.

jay28 · Aug 12, 2008

GeorgeD said:
It seems like people keep coming here thinking that the house edge "resets" every shoe, table switch, day, week, session, whatever. IOW: many think if they "beat" the edge for some time period that the next period is a whole new ballgame. They think they can always lock in winnings, but they can always recover their losses.

I think this is kind of like a variation of the gamblers fallicy

What (maybe in FAQ's) might easily explain this fault in logic to the average newcomer?

The house edge does reset for each session. For an example imagine a roulette table with no house edge, (no greens) just red and black numbers. All black numbers have came in for the last 20 spins......

Would you put you money on red because the stats have to even out?

The answer is no...it's still 50/50 for either a red or black result.

Every shoe in blackjack is individual, It's not until you have the data from the long haul that the house or player edge becomes predictable. The next game is a new ball game, however the next 10,000 isn't.

GeorgeD · Aug 13, 2008

I see what you all mean .. each shoe and each hand is a new event.

What I'm trying to say is many people seem not understand that the long term odds will play out in the end ... either + for an AP or - for a Non AP. They don't see each days session as just part of the overall "long term". That's why people think things like stop loss plans or money management can make them a winner overall.

Most people will say that if they flip 10 heads in a row that the next flip is more likely to be tails but we know it is still 50/50 because the last ten flips are history.

Even extreme Positive or negative variance doesn't change the future ... it just becomes less significant to the long term as more and more hands are played out. The gamblers fallacy makes people believe otherwise.

Kasi · Aug 13, 2008

GeorgeD said:
What I'm trying to say is many people seem not understand that the long term odds will play out in the end ...

No big deal - said it before and I'll say it again lol.

What does "long-term" mean?

In theory, it's "forever". In practice, people only live so long. Or play so many hands/roulette/crap decisions. With different rolls and goals and the possibility of changing unit size at any time, although "systems" never include that possibility.

Brock Windsor · Aug 13, 2008

Kasi said:
.
What does "long-term" mean?
In theory, it's "forever". In practice, people only live so long. Or play so many hands/roulette/crap decisions.

The consensus I've received seems to be that 20000 hands puts you in the long run, at least for BJ. Anyone else have an opinion on a "reasonable" long run?
BW

Click to expand...

N&B · Aug 13, 2008

George:
IMHO they seem to be of the "dice-rolling" mentality. Every toss is no indication of prior or future results. And it seems to them that the cards do not have a memory. So when they win, they press, and when they lose, they press. Anything except the dreaded break-even.

sagefr0g · Aug 14, 2008

GeorgeD said:
I see what you all mean .. each shoe and each hand is a new event.

What I'm trying to say is many people seem not understand that the long term odds will play out in the end ... either + for an AP or - for a Non AP. They don't see each days session as just part of the overall "long term". That's why people think things like stop loss plans or money management can make them a winner overall.

what your saying is in line with orthodox AP thinking no doubt. can't argue with that. i'm currently wondering if things like stop loss plans, stop win plans, money management and goal shooting coupled with orthodox AP play might be used with relative impunity. and perhaps if good 'luck' is taken advantage of and the effect of bad 'luck' minimized maybe similar results to advantage play might be achieved with probably a effeciancy penalty.

please critique if your so inclined: http://www.blackjackinfo.com/bb/showthread.php?t=11134

Most people will say that if they flip 10 heads in a row that the next flip is more likely to be tails but we know it is still 50/50 because the last ten flips are history.

your math is flawless if i had a clue.

most people would say that because they have the integrety to believe the coin is fair or not question that the coin is fair if they are told that it is. the thng is there probably is no such thing as a fair coin.
that's probably true for cards in any given 'short term'. could that be taken advantage of? don't ask me cause i don't know lol

Even extreme Positive or negative variance doesn't change the future ... it just becomes less significant to the long term as more and more hands are played out. The gamblers fallacy makes people believe otherwise.

exactly right all things being normal. life and even one's casino experience's are full of abnormal virtually chaotic experiences. so maybe variance can't change anything. question is can you?

but again your math is right on if i had a clue. :grin:

edit: and i just want to add i hope i'm not being unfair putting forth such silly arguements because i really do think you are right.

Kasi · Aug 14, 2008

Brock Windsor said:
The consensus I've received seems to be that 20000 hands puts you in the long run, at least for BJ. Anyone else have an opinion on a "reasonable" long run?BW

That's fine - maybe to you it is 20,000 hands. Maybe it's even 50,000 hands. Maybe it means you have not yet gone broke, maybe it means you have doubled your roll, maybe it means you are simply ahead by some amount in that time, maybe you have already gone broke and never made the "long-run". Maybe it means 20,000 hands seen but not hands played. Maybe it means, 200 hours later, one will assume 20,000 hands have been played regardless of how many hands have actually been played.

Whatever it is, if that's "long-term" for BJ, then why not apply the same standard to someone using stop-loss, money-management, betting systems, voo-doo, whatever, to that guy too?

Whatever - guess it always goals, rolls, time and risk lol.

sagefr0g · Aug 14, 2008

Brock Windsor said:
The consensus I've received seems to be that 20000 hands puts you in the long run, at least for BJ. Anyone else have an opinion on a "reasonable" long run?
BW

most AP's wouldn't say how ever long it takes to get the job done. :cat:

N&B · Aug 14, 2008

True, never reveal N(zero) of your play.

Brock Windsor · Aug 15, 2008

Kasi said:
That's fine - maybe to you it is 20,000 hands. Maybe it's even 50,000 hands. Maybe it means you have not yet gone broke, maybe it means you have doubled your roll, maybe it means you are simply ahead by some amount in that time, maybe you have already gone broke and never made the "long-run". Maybe it means 20,000 hands seen but not hands played. Maybe it means, 200 hours later, one will assume 20,000 hands have been played regardless of how many hands have actually been played.

Whatever it is, if that's "long-term" for BJ, then why not apply the same standard to someone using stop-loss, money-management, betting systems, voo-doo, whatever, to that guy too?

Whatever - guess it always goals, rolls, time and risk lol.

As I understand it, after about 20000 hands (or slightly less, maybe 16?) the sample is large enough that the typical expected return counting surpasses the standard deviation in $$. Everything beyond this the standard deviation curve (in $) is continuing to flatten while the expected return remains constant. This is what I was alluding to. This is why very few voodoo systems will show a profit after 20000 hands. The casinos expected return has surpassed 1 standard deviation so there is a greater than 68% chance the casino is ahead of any system by that time. I haven't run these myself so am not certain, but for some reason 20000 stuck out in my head as a rough estimate.
BW

sagefr0g · Aug 16, 2008

Brock Windsor said:
As I understand it, after about 20000 hands (or slightly less, maybe 16?) the sample is large enough that the typical expected return counting surpasses the standard deviation in $$. Everything beyond this the standard deviation curve (in $) is continuing to flatten while the expected return remains constant. This is what I was alluding to. This is why very few voodoo systems will show a profit after 20000 hands. The casinos expected return has surpassed 1 standard deviation so there is a greater than 68% chance the casino is ahead of any system by that time. I haven't run these myself so am not certain, but for some reason 20000 stuck out in my head as a rough estimate.
BW

isn't it basically N0 your getting at here? probably that number is ballparkish. thing is i guess it's game dependent, dependent on the count method, dependent on the bet ramp and dependent on probably a whole lot of things.

penetration, number of players ect........

Kasi · Aug 17, 2008

Brock Windsor said:
This is why very few voodoo systems will show a profit after 20000 hands.

That's all - be ahead after 20,000 hands lol?

Of course it depends on a lot of things but it could be extremely likely lol.

Can I have a 10,000 unit roll ?

Verboten72 · Aug 20, 2008

House Edge 'Reset'

Very interesting topic...and one I've thought about quite a bit.

Consider the following: all other things being equal, would you rather play a game of single-deck or a game using an 'infinite' deck? Or better yet, would you want to enter a game at the beginning of a shoe or when all the counters at the table are Wonging out?

Even non-counters would agree that fewer decks are more favorable for the player...which is why multi-deck games almost always have otherwise better rules than single-deck games. The reason there are different tables for single-deck and eight-deck games is because the expected values and probabilities for each hand played are, in fact, composition dependent.

Whether you want to call it a 'reset' or whatever else...it should be clear that house edge is different (however miniscule the difference) at different entry points into the game given the composition of the remaining deck(s).

Just my two cents...

Kasi · Aug 22, 2008

sagefr0g said:
isn't it basically N0 your getting at here? probably that number is ballparkish. thing is i guess it's game dependent, dependent on the count method, dependent on the bet ramp and dependent on probably a whole lot of things.
penetration, number of players ect........

You got it lol.

Some games it could be 4000 hands. Others 1,000,000 lol.

Here's some voodoo.

I got $10K and flat bet $1/hd for 10000 hands in a -0.4% game. After 10000 hands, I hit my EV and I'm down $40.

I change my $1 unit to $40.

How often do you think I could finish at least 1 unit up flat-betting with a 249 unit roll if I played until I lost all 249 units or won at least 1 unit?

God forbid it's a -0.1% game and I have a 999 unit roll left.

If and when I do finish at least 1 unit up, I go back to $1 for the next 10,000 hands.

How long is the "long-run" then? How long must I live to lose my $10K? Maybe I knew in the first place I never would play more than 300,000 hands in the first place.

Just throwing cr*p out here lol.

sagefr0g · Aug 22, 2008

Kasi said:
....

Here's some voodoo.

is that allowed here? :joker:

I got $10K and flat bet $1/hd for 10000 hands in a -0.4% game. After 10000 hands, I hit my EV and I'm down $40.

I change my $1 unit to $40.

How often do you think I could finish at least 1 unit up flat-betting with a 249 unit roll if I played until I lost all 249 units or won at least 1 unit?

God forbid it's a -0.1% game and I have a 999 unit roll left.

If and when I do finish at least 1 unit up, I go back to $1 for the next 10,000 hands.

How long is the "long-run" then? How long must I live to lose my $10K? Maybe I knew in the first place I never would play more than 300,000 hands in the first place.

Just throwing cr*p out here lol.

ok so i been workin on this problem. at this point i'm a lost puppy.
just give me one hint. it doesn't seem as if you've given enough info to know the EV.

i got this one number 6,225 hands and don't know what to do with it or if it's even not bogus.
like maybe you could bet the $40 unit a thousand hands 6.225 times? probably bogus cause i'm not using any known procedure. to lazy to look it up.
and probably a bogus ev=-0.040 which seems ridiculous to my mind

am i gonna have to use that z-stat thing the wiz of odds talks about?
lmao i think i'll go look up the definition of EV.
so maybe i just need to look up the EV for a basic strategy only player? for what game though?

House Edge and Noobs

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

New Member

Well-Known Member

Well-Known Member