Why do runs occur?

Why do runs occur?
I find it interesting that runs seem to actually occur. If the table is going in one direction it actualy does so for some time and I either win or loose so now part of my stratage is to start low and see which way the table is going. But why????

This is more a function of the human brain's natural tendency to look for patterns. You tend to remember the runs more than the times without much of a pattern. If you keep detailed notes over many sessions you will notice that runs actually occur less frequently than your memory leads you to believe.

Selective memory. We're trained very young to notice and identify patterns. For the sake of argument, let's say it's 50/50 you win or lose. Over 6 trials, you can get results such as:

1. WWWLWW
2. WLLWLW
3. LLLLWL

You're more likely to remember 1 and 3 and forget all about 2. They stick out in your mind. So it just "seems" like it happens that way most of the time, but in reality it doesn't.

This article has been one of my favorites to explain the tendency to see patterns when gambling:

How a Lack of Control Leads to Superstition, by Michael Shermer, Scientific American, January, 2010

They actually do occur. In an experiment where people are asked to generate a random string of Heads and Tails, the average person puts less streaks, and shorter ones, than is actually expected. Then when we do notice truly random streaks, they seem unusual.

Quote: jc2286

Selective memory.
1. WWWLWW
2. WLLWLW
3. LLLLWL
You're more likely to remember 1 and 3 and forget all about 2.

Its the casino owner who knows that all those pattern seekers will windup losing to him eventually.

Quote: teliot

This article has been one of my favorites to explain the tendency to see patterns when gambling:

How a Lack of Control Leads to Superstition, by Michael Shermer, Scientific American, January, 2010

Very interesting, thanks Eliot.

What I experience is that I always bet the same thing no pass full odds. One day recenty in 15 minutes I doubled my money a seven came up almost every other roll for a win. yesterday I lost with the pass line betters doing very well over a one hour period and my no pass crushed. I don't change my tech based on the table action.

read it thanks

Quote: mikeman6kids

If the table is going in one direction it actualy does so for some time and I either win or loose

so now part of my stratage
is to start low and see which way the table is going. But why????

What exactly do you mean by a "run"? A run of what??
What in your mind is a "trend"
Why would you bet less if you "think" your "perceived run or trend" is against the way you play?

Here is some sim data
$10 Dpass, 5X Lay odds.
Many perceived that the larger the buy-in the greater chance one had to double a bankroll.
Looks true, well...
Is it true across the board??

1 million players in each sim
DPass, $10 Flat bet and 5X Lay Odds
Double starting Bankroll or Bust trying.
No time limit on # of games played

10 unit Bankroll($100 buy-in $10 flat bets)
Bankroll was busted . . = 55.640% of the time ( 556396)
Win goal was met . . . = 44.360% of the time ( 443604)

20 unit Bankroll($200 buy-in $10 flat bets)
Bankroll was busted . . = 53.943% of the time ( 539430)
Win goal was met . . . = 46.057% of the time ( 460570)

30 unit Bankroll($300 buy-in $10 flat bets)
Bankroll was busted . . = 52.887% of the time ( 528869)
Win goal was met . . . = 47.113% of the time ( 471131)

40 unit Bankroll($400 buy-in $10 flat bets)
Bankroll was busted . . = 52.562% of the time ( 525617)
Win goal was met . . . = 47.438% of the time ( 474383)

50 unit Bankroll($500 buy-in $10 flat bets)
Bankroll was busted . . = 52.436% of the time ( 524357)
Win goal was met . . . = 47.564% of the time ( 475643)

100 unit Bankroll($1000 buy-in $10 flat bets)
Bankroll was busted . . = 52.714% of the time ( 527139)
Win goal was met . . . = 47.286% of the time ( 472861)

There were some long "runs" of players doubling their bankroll.
I see 17 in a row at one point.
Does this mean the don't pass bet is the best way to double a bankroll if the "perceived trend"
is that the don't pass in the short run observed is winning more than the pass line??

Or is the OP saying,
if the pass line is "winning more" than the don't pass, very probable,
one should still bet the don't pass but just bet less on it until what event(s) happens to make larger bets???

inquiring minds want to know

Good Luck!

I have more concrete evidence than perception of trends I watch the money go or come from the rail even though my bet is the same and almost 50/50

Almost but not quite 50/50. That's the key.

Streaks happen because you are talking a random walk, and at times the results will be all in one direction. Past performance does not predict future events, or if it does, you are not dealing with a random variable.

Quote: mikeman6kids

What I experience is that I always bet the same thing no pass full odds. One day recenty in 15 minutes I doubled my money a seven came up almost every other roll for a win. yesterday I lost with the pass line betters doing very well over a one hour period and my no pass crushed. I don't change my tech based on the table action.

Let me complete your sentence: until I have no more money to bet.

Even those who claim to have consistent strategies have to take bathroom breaks and other inconsistencies as well as determining a time to stop.

Just because it's not based on a pattern you recognize, does not mean you aren't subconsciously responding to stimuli to affect your bets.

I'm not criticizing, just putting something out there to think about: the brain plays tricks on you.

My big thing is just not looking backwards. If something horrible just happened, keep looking forwards. If something fantastic just happened .. eyes AHEAD.

Quote: mikeman6kids

I have more concrete evidence than perception of trends

concrete evidence is good.
So is having fun playing Craps.
Yes, we did figure that one out. It is a main reason you started this thread

Quote: mikeman6kids

I watch the money go or come from the rail even though my bet is the same and almost 50/50

OK.

Are you saying the don't pass is a better bet than the pass when more money is moving which way from the rail??
This is very interesting.

They BOTH win about 49.3% of the time.
The difference is "when" they win

But you have something that tells you when to bet less $$$ on the don't pass (the odds part I guess)
It can not be just watching the money move.

Streaks, positive or negative, are nothing but exceptional periods in the probability of die combinations. After the come out roll if 7, which should be 1 in 6 rolls, doesn't come up before point then you have a positive streak. And then given the same situation except 7 comes up within the 6 rolls and prior to point you have the negative streak. I for one watch the dice and play the statistical odds of die combo's and usually play for long periods of time. Sometimes I win and sometimes I lose, but never too much.

I saw a wicked mean streak at the M Sunday morning around 8am.

A japanese fellow bet $10 do's with max odds 3x4x5x doing a passline and come bet with every roll.

He lost $5,000 going straight down like a car careening off a cliff.

He swapped to don'ts at about -$3500 and back to do's at about $4500 all to no avail.

It was the most horrible display of bad luck I have ever witnessed.

Never increased his bets on the downslide, just REALLY bad luck!

To get it back, which he did, he did start making bold plays and got lucky.

I think he had about 80% of it back by the time I left.

He got REALLY pissed. He held it together emotionally for the first negative four grand, but when he got down to his last $400 he was pretty pissed.

Moral of the story: streaks happen in everything random. I have never seen a streak this long and this bad before in my life. But now I have. I think he was sure it would turn around soon all the way down and he was just wrong.

I'm glad I wasn't witnessing my own bad streak, though!!!

If events were evenly spaced, they would be predictable, and therefore not random.

Quote: 7craps

They BOTH win about 49.3% of the time.
The difference is "when" they win

Choppy tables are the worst. Some players who are only Right Bettors just bet less and seem able to weather a storm a bit better than those who shift to the Dark Side when the skies darken and don't survive to hear Winner.

Quote: FleaStiff

Choppy tables are the worst. Some players who are only Right Bettors just bet less and seem able to weather a storm a bit better than those who shift to the Dark Side when the skies darken and don't survive to hear Winner.

People talk about choppy referring to specific bets or strategies.

If you believe in randomness, then you can bet randomly and measure the choppiness of your luck.

Screw what other people are doing! But streaks are in the random data, not in the table or in the cards or in any vehicle of the casino.

Just randomly pick a bet each time you are ready to make one. Then look only at your luck as a gauge for what to do next.

Too many people think towards patterns when they compare what is happening now to what has happened before.

You can easily break that trend by betting differently and expecting things to happen in a unique way each and every time you take action always ready to improvise!

Try this on for size:

Bet 1: Pass line
Bet 2: Don't come
Bet 3: come
Bet 4: flip a coin bet the come or don't come accordingly
Bet 5: place the 6
Bet 6: take everything down that you can and no new bets
Bet 7: nothing
etc ... randomly improvising random low edge bets....

You will have streaks doing this, but they will be streaks in YOUR luck not in the table or in the dice, or anything else. It will be unique to you.

And you don't have to vary the bet amounts, just the bets that you choose to take or take down with each roll.

Also, manage your total exposure. Imagine the worst case that could happen each roll and make sure you can stomach that.

Quote: mikeman6kids

Why do runs occur?
I find it interesting that runs seem to actually occur. If the table is going in one direction it actualy does so for some time and I either win or loose so now part of my stratage is to start low and see which way the table is going. But why????

Eating too much of the wrong food?.....

Quote: Ahigh

I saw a wicked mean streak at the M Sunday morning around 8am.

A japanese fellow bet $10 do's with max odds 3x4x5x doing a passline and come bet with every roll.

He lost $5,000 going straight down like a car careening off a cliff.

He swapped to don'ts at about -$3500 and back to do's at about $4500 all to no avail.

It was the most horrible display of bad luck I have ever witnessed.

Never increased his bets on the downslide, just REALLY bad luck!

To get it back, which he did, he did start making bold plays and got lucky.

I think he had about 80% of it back by the time I left.

He got REALLY pissed. He held it together emotionally for the first negative four grand, but when he got down to his last $400 he was pretty pissed.

Moral of the story: streaks happen in everything random. I have never seen a streak this long and this bad before in my life. But now I have. I think he was sure it would turn around soon all the way down and he was just wrong.

I'm glad I wasn't witnessing my own bad streak, though!!!

Holy crap! That's horrible. Even I (the unluckiest craps player ever) haven't had a single-session streak that bad. Been in that mindset, though: ("It can't get any worse, it can't get any worse . . .oh, shit")

I like this simulation and if I understand it correctly you need a bank role of about 20 unit to maximize what odds you can get 52 vrs 47 loss win. I assume a unit is the 10 dollar no pass? Can you run the simulation with 6 dollar minimum 5x lay odds no pass. Can you post the graph of the dollar amounts or if not just the dice rollls?

So you need to have at least a 20 unit bankroll to have the best odds. What aobut if the minimum bet is 5 or 6 dollars? How big a bankroll do you need and I assume you mean 20 unit to be 20 times 10?

Quote: mikeman6kids

So you need to have at least a 20 unit bankroll to have the best odds.
What aobut if the minimum bet is 5 or 6 dollars?
How big a bankroll do you need and I assume you mean 20 unit to be 20 times 10?

That is why I listed Bankroll Units.
You are right.
A $10 bettor would have a 20 unit buy-in with 20 times $10 or $200
A $25 bettor would have a 20 unit buy-in with 20 times $25 or $500
A $5 bettor would have a 20 unit buy-in with 20 times $5 or $100
A $6 bettor would have a 20 unit buy-in with 20 times $6 or $120

All have the same *unit* buy-ins but different total $ buy-ins.

Quote: Ahigh

He swapped to don'ts at about -$3500 and back to do's at about $4500 all to no avail.

One of my superstitions is that the dice hate side-switchers. Sure enough [evidently] he timed it perfectly wrong.

Quote: Ahigh

People talk about choppy referring to specific bets or strategies.

If you believe in randomness, then you can bet randomly and measure the choppiness of your luck.

Screw what other people are doing! But streaks are in the random data, not in the table or in the cards or in any vehicle of the casino.

Just randomly pick a bet each time you are ready to make one. Then look only at your luck as a gauge for what to do next.

Too many people think towards patterns when they compare what is happening now to what has happened before.

You can easily break that trend by betting differently and expecting things to happen in a unique way each and every time you take action always ready to improvise!

Try this on for size:

Bet 1: Pass line
Bet 2: Don't come
Bet 3: come
Bet 4: flip a coin bet the come or don't come accordingly
Bet 5: place the 6
Bet 6: take everything down that you can and no new bets
Bet 7: nothing
etc ... randomly improvising random low edge bets....

You will have streaks doing this, but they will be streaks in YOUR luck not in the table or in the dice, or anything else. It will be unique to you.

And you don't have to vary the bet amounts, just the bets that you choose to take or take down with each roll.

Also, manage your total exposure. Imagine the worst case that could happen each roll and make sure you can stomach that.

This is a great illustration of this fact: no matter how you bet, you are "chasing a pattern". If you progress your bets, you are chasing a winning streak; if you regress, you are chasing a "choppy" table, I suppose. If you simply bet pass or don't pass, then the only thing that matters is your W-L record, not when the win and losses occur. A guy on rec.gambling.craps came up with what he called the "flip-skip" method, which needs a flip-flop pattern of passline wins for success.

I had an idea a few years ago about pitching in baseball: in order not to be predictable, have a few sets of probability distributions for different pitches, depending on the situation, then use an RNG to select the next pitch. >:-)
Cheers,
Alan Shank

Quote: MrWildCard

Streaks, positive or negative, are nothing but exceptional periods in the probability of die combinations. After the come out roll if 7, which should be 1 in 6 rolls, doesn't come up before point then you have a positive streak. And then given the same situation except 7 comes up within the 6 rolls and prior to point you have the negative streak. I for one watch the dice and play the statistical odds of die combo's and usually play for long periods of time. Sometimes I win and sometimes I lose, but never too much.

What does the bolded sentence mean?

Quote: bbvk05

Quote: MrWildCard

Streaks, positive or negative, are nothing but exceptional periods in the probability of die combinations. After the come out roll if 7, which should be 1 in 6 rolls, doesn't come up before point then you have a positive streak. And then given the same situation except 7 comes up within the 6 rolls and prior to point you have the negative streak. I for one watch the dice and play the statistical odds of die combo's and usually play for long periods of time. Sometimes I win and sometimes I lose, but never too much.

What does the bolded sentence mean?

If I'm playing on a table that offers higher than 2x odds ill bet more on the point with the most die combos. Things like if the point is a 6/8 ill bet higher on my back odds and press the place bets. If the point is 4/10 ill go even odds and same bets and let the high odds payout be my profit. But as I said before this style of play won't make you rich. It'll buy you time on the table and the ability to leg out a couple c notes.

Quote: MrWildCard

Quote: bbvk05

Quote: MrWildCard

Streaks, positive or negative, are nothing but exceptional periods in the probability of die combinations. After the come out roll if 7, which should be 1 in 6 rolls, doesn't come up before point then you have a positive streak. And then given the same situation except 7 comes up within the 6 rolls and prior to point you have the negative streak. I for one watch the dice and play the statistical odds of die combo's and usually play for long periods of time. Sometimes I win and sometimes I lose, but never too much.

What does the bolded sentence mean?

If I'm playing on a table that offers higher than 2x odds ill bet more on the point with the most die combos. Things like if the point is a 6/8 ill bet higher on my back odds and press the place bets. If the point is 4/10 ill go even odds and same bets and let the high odds payout be my profit. But as I said before this style of play won't make you rich. It'll buy you time on the table and the ability to leg out a couple c notes.

The easiest way to get rich is to borrow steal 100x what you need to live comfortably for the rest of your life. Then bet against an unlikely event with low edge bets once, then return the stolen money before it is noticed.

If you lose the bet, you are going be a thief, but that probably won't happen.

Stealing that much money without being noticed and returning it in time could get tricky.

Hmmmmmmm...tempting. Didn't work out so well for Dan Mahowny.