This reminds me of the old BBS days. Someone would figure out the BBS code for displaying variables, and start posting messages with stupid things like "If your name is on this list, you will be banned: Joe, Frank123, [$name],...".
The computer doesn't know your name. It echoes a variable.
You can make a small bankroll last a long time in craps if you place the right bets. It's a mish-mosh of different things, and I can never remember them. (I don't play Craps). The wizard's page on Craps has them listed somewhere.
Pai Gow is also really good for the long haul. It has a low edge, and it takes forever to play a hand. You have an advantage on any hand where you're the banker, so take that option whenever you can.
If your answer to this is anything other than "The house, of course", then it may be a good idea to let someone else hold all of your money next time you are so much as in the same state as a casino.
Actually, depending on the number of decks left, it's either an even-money bet, or up to a 3% advantage for me.
IF you to gamble, and someone said they'd play coin-flip with you, and that you were allowed to use a weighted coin to your advantage, wouldn't you?
Yup, the teams are more of a worry, simply because any well organized team will have a huge bankroll. A red-chipper with a maximum bet of $100 isn't a worry. A team who can drop $10,000 bets with a 2% advantage is.
Of course, there are two reasons the casinos aren't too concerned:
The MIT team did it first. And the casinos figured out how they did it, so now they know what to look for. The longer a team operates, the easier they are to spot. The truth is, the MIT teams probably has made more from book sales, movie rights, public appearances and course fees than they did from counting.
Teams are unbelievably difficult to get working. Just the trust issue is a major factor. Name five people in your life who you would blindly and 100% implicitly trust with $50,000 of your money. Keep in mind that you pool your bankroll, and share the profits. It just takes one bad egg to realize that they can slog out the counting for a year to see a 1-2% return on their investment-- OR they can take out $10,000 and lose it in a "bad session" (ie: they just pocket it and tell you they lost). Also, one bad player on the team (who isn't up to perfect snuff) can wipe out any profits the team will see. (If you could name five people, how many of them do you trust to be able to do simple math for 8 hours straight?) It takes months of training, analysis and testing to ensure every member is trustworthy and competent.
So the casinos just balance the likelyhood of an effective team coming to town vs. the chance that they'll spot them in operation. Then they tally up the amount they'll earn from their tables from bad players. That'll tell them how much it's worth spending on anti-counting training/technology/etc. Why spend $500,000 on a new system to keep someone from earning $250,000 / year off you-- especially if there's only a 0.001% chance that team will come around. The numbers are fudged, but it's all just a numbers game. Somewhere along the lines, someone gets paid a bunch of money to tell them what the numbers are.
This is one reason why many blackjack tables, particularly high limit tables, do not allow new players to sit down in the middle of a shoe (i.e. you have to wait until the shoe currently being played is finished, the cards are shuffled, and the next shoe is loaded)
NMSE - No Mid-Shoe Entry. Yup, just about all the high-roller tables have that. It's rarer on the mid or low level tables, though, since their bread and butter is unskilled, transient traffic. I've seen a $50-$2500 table that allowed midshoe entry. Let me tell you, seeing people drop $500 a hand on a game they don't know how to play-- that's quite a sight to watch. Maybe I should just open a casino.
Yup, a bet ceiling can make it nearly impossible to get an advantage. You have to be able to bet more in a positive count than in a negative count-- and quite a bit more. 10x as much would be nice, for most fair games.
So if a casino opens a $10-$100 table, you MIGHT see ~$10/hour from it if you play perfectly. A few mistakes can cost you a good $1 or $2 per hour. Can you imagine doing a mind-numbingly boring job, sitting still for 8 hours a day, for just $10 per hour? Well, probably-- this is Slashdot-- but also imagine you have to personally pay for every error you make out of your own pocket.
You can try to mitigate the negative counts by not playing those hands, and only playing positive counts. But then you get fewer hands per hour. And some games might not let you join after the shuffle has occurred. Or you might lose your seat and miss out on the high counts.
I once spent five hours walking around a blackjack pit, and didn't see a single high-count that had an open seat. Tons of fun.
If you bet the same every hand, you get that advantage.
No. If you vary your bet according to the count, you get the advantage. You need to be proportionally more as the count goes up, to make up for the small bets you made when the count was down.
If you work with a team and the next guy bets BIG, then its hugely in your advantage.
Again, close. If you work with a team, then the next guy will not be playing at all during a low count. If you are counting, they you don't vary your bet at all. But when the count goes up, you signal the Big Player to come in. They place a large bet-- something that would seem suspicious if YOU put it down, but is normal for him. He only bets that amount and doesn't vary.
In that case, your advantage is EXACTLY THE SAME (~0.5% per count)-- but the EV will be greater. The more you bet, the more you'll earn, but the rate will be the same. If you have a 1.5% advantage, it doesn't matter if you bet $100 or $10,000. You will only "earn" 1.5% of that.
Your numbers are WAY off how multi-person counting works.
Not really. And it depends on the type of team you are working with. If you have small player/big player (as above), then you will still need ~12,000 hands to overcome one standard deviation. If you are sharing a bankroll amongst many counters, then, well, you still need 12,000 hands, but you will be able to pool your hands. (Assuming you are all playing at separate tables). You reach the longterm much quicker, and lower your risk of ruin.
Interestingly, I've had dealers help me count before
Uhhhg. Never rely on the dealer. They don't know what they're doing. They're just a flawed gaming machine made of flesh.
Doing simple "count the tens" helps your odds on a non-continuous-dealt game
No. No it doesn't. It's useless to count the tens unless you are also counting the low cards that balance it. It's useless to know that five 10s have left the deck, unless you know how many low cards have also left the deck. The whole point about counting is to know the estimated composition of the remaining deck.
Example: You are counting the tens. 6 tens come out of the deck. You assume a count of -6, and lower your bet. I am hi/lo. I see those 6 tens come out, and then 12 low cards. I KNOW a count of +6, and raise my bet to take advantage of it. Guess who is coming out on top
I had a dealer, who was watching me pull back as the tens had largely made their appearances actually told me "you don't want to take this next hit".She was right.
And she could just have easily have been wrong. She doesn't know what the next card is. Neither does a counter. A counter just knows the estimated composition of the deck, and can vary their bet or use an "index play". IE: Basic Strategy says 12 vs. 2 is a hit, because that move is the best possible play statistically. But at a count > 0, it becomes stand, because now that move is the bes possible play statistically. That doesn't mean the next card is a 10. It just means that you'll lose less by standing than by hitting.
You cannot point to a single hand and use that as proof for anything. Remember, 12,000 hands is where "long term" begins. Everything else is indistinguishable from luck. If anything, the dealer was taking a blind shot hoping for a tip.
There is a romantic view of card counting. People assume it's a magical skill you pick up, then can just roll into a casino and use it as a personal ATM overnight.
The truth is, it gives you about a 1% edge over the house. That means that for every $10 you bet, you'll "earn" $0.10. You can get, maybe, 100 hands of Blackjack per hour on a good day.
And the "margin of error" (standard deviation) means that your long-term swings won't balance out until after about 12,000 hands. 120 hours of Blackjack, just to statistically be guaranteed to at least break even.
And all that is assuming you count perfectly, and play perfectly.
So after card counting gets hyped, you'll get a whole ton of people who want the quick win. They'll learn a quick hi/lo system. They won't practice. They won't learn basic strategy perfectly (quick, what's the proper move when you have 44 vs. a dealer's upcard of 5?). And they'll go into the casino. Maybe they'll double up quickly and walk away. More likely they'll just keep playing, have a few drinks, and either make a bit of money, or get frustrated and lose everything, or just play for a while and have fun. But in every case, they'll be playing with a disadvantage. Making a couple mistakes or missing a couple counts, maybe they're playing an even money game, or just 0.5% house edge. If they start steaming and making the big errors, they'll be giving the house 4-5% of their money on every hand.
And for the one in a thousand counter who does a good job and earns 1-2% on her money, they'll be 999 players who give it all back.
If card counting had the ability to destroy the casinos, they'd have been out of business a long time ago. Blackjack is profitable for the casino.
At $1/track they'd have to sell more than one track to every man, woman, and child in the US
Not quite. The tracks will probably be a loss leader. The profit is going to come from two different places.
The myriad of Beatles fans who will buy the albums, and then go on to buy a bunch of other "while I'm here I might as well" tracks.
The business that they won't lose to a competitor. If Amazon signed an exclusive with the same people, then people who wanted Beatles would go there, and probably stay there. Apple is paying a premium to keep their customers on iTMS.
I wouldn't have known I even owned a Trinitron if it wasn't for this article. I've had this 19", 2,874lb beast on my desk since I lugged it home from the thrift shop, but never noticed the brand name up in the corner. So that's what those lines are. I just figured they were what-I-paid-for bugs.
Though, I also have translucent diagonal lines that run across the screen that remind me of a CRT projector than needs its edges blanked. And the pincushioning has always been off. And on a cold day, I have to turn the contrast waaaaaaay down to keep it from shutting itself off. But aside from that, best $20 I ever spent.
I call dibs on the patent for "The abstract representation of a generic object from which all other objects, abstract or concrete, are henceforth derived."
And why shouldn't some malicious attacker use every single exploit they've ever written/downloaded/stolen in their botnet payload? It's not like they're really concerned about bandwidth or CPU loads.
It's not in an abandoned nuclear bunker. It's in an "abandoned" nuclear bunker. Wikileaks goes chugging along-- but if anything ever hits their servers that's a bit too sensitive for the US government, someone presses a button, and the "abandoned" nuclear weapons in the bunker go "foom".
Advertising on Slashdot is a chancy business. Never mind the click through rate, if I were an advertiser, I'd be concerned about the likely high percentage of AdBlock users who never see the ads at all.
If I were an advertiser, I'd be concerned about the possibility that I run a good ad campaign, and got so many clicks my adserver got Slashdotted
If you can create features where I rank my social insurance number against celebrity's horoscopes, and then play hangman with my bank account information against friends to see who is better, then I'm totally in!
And according to a Microsoft press release, they feel confident that there are key indicators signaling IBM's adoption of Vista for the new supercomputer.
Slashdot Mirror
That's what lackies are for. "Hey, Merv, touch that for me, will you?" *zzz*. "Hey, Johnson, touch that for me, will you?"
"friend"
The computer doesn't know your name. It echoes a variable.
Pai Gow is also really good for the long haul. It has a low edge, and it takes forever to play a hand. You have an advantage on any hand where you're the banker, so take that option whenever you can.
Actually, depending on the number of decks left, it's either an even-money bet, or up to a 3% advantage for me.
IF you to gamble, and someone said they'd play coin-flip with you, and that you were allowed to use a weighted coin to your advantage, wouldn't you?
The correct answer is split. "Never split fours" is true in all cases, except against a 5 or a 6. In that case, you pair if you can, hit otherwise.
Of course, there are two reasons the casinos aren't too concerned:
So the casinos just balance the likelyhood of an effective team coming to town vs. the chance that they'll spot them in operation. Then they tally up the amount they'll earn from their tables from bad players. That'll tell them how much it's worth spending on anti-counting training/technology/etc. Why spend $500,000 on a new system to keep someone from earning $250,000 / year off you-- especially if there's only a 0.001% chance that team will come around. The numbers are fudged, but it's all just a numbers game. Somewhere along the lines, someone gets paid a bunch of money to tell them what the numbers are.
This is one reason why many blackjack tables, particularly high limit tables, do not allow new players to sit down in the middle of a shoe (i.e. you have to wait until the shoe currently being played is finished, the cards are shuffled, and the next shoe is loaded)NMSE - No Mid-Shoe Entry. Yup, just about all the high-roller tables have that. It's rarer on the mid or low level tables, though, since their bread and butter is unskilled, transient traffic. I've seen a $50-$2500 table that allowed midshoe entry. Let me tell you, seeing people drop $500 a hand on a game they don't know how to play-- that's quite a sight to watch. Maybe I should just open a casino.
So if a casino opens a $10-$100 table, you MIGHT see ~$10/hour from it if you play perfectly. A few mistakes can cost you a good $1 or $2 per hour. Can you imagine doing a mind-numbingly boring job, sitting still for 8 hours a day, for just $10 per hour? Well, probably-- this is Slashdot-- but also imagine you have to personally pay for every error you make out of your own pocket.
You can try to mitigate the negative counts by not playing those hands, and only playing positive counts. But then you get fewer hands per hour. And some games might not let you join after the shuffle has occurred. Or you might lose your seat and miss out on the high counts.
I once spent five hours walking around a blackjack pit, and didn't see a single high-count that had an open seat. Tons of fun.
No. If you vary your bet according to the count, you get the advantage. You need to be proportionally more as the count goes up, to make up for the small bets you made when the count was down.
If you work with a team and the next guy bets BIG, then its hugely in your advantage.Again, close. If you work with a team, then the next guy will not be playing at all during a low count. If you are counting, they you don't vary your bet at all. But when the count goes up, you signal the Big Player to come in. They place a large bet-- something that would seem suspicious if YOU put it down, but is normal for him. He only bets that amount and doesn't vary.
In that case, your advantage is EXACTLY THE SAME (~0.5% per count)-- but the EV will be greater. The more you bet, the more you'll earn, but the rate will be the same. If you have a 1.5% advantage, it doesn't matter if you bet $100 or $10,000. You will only "earn" 1.5% of that.
Your numbers are WAY off how multi-person counting works. Not really. And it depends on the type of team you are working with. If you have small player/big player (as above), then you will still need ~12,000 hands to overcome one standard deviation. If you are sharing a bankroll amongst many counters, then, well, you still need 12,000 hands, but you will be able to pool your hands. (Assuming you are all playing at separate tables). You reach the longterm much quicker, and lower your risk of ruin. Interestingly, I've had dealers help me count beforeUhhhg. Never rely on the dealer. They don't know what they're doing. They're just a flawed gaming machine made of flesh.
Doing simple "count the tens" helps your odds on a non-continuous-dealt gameNo. No it doesn't. It's useless to count the tens unless you are also counting the low cards that balance it. It's useless to know that five 10s have left the deck, unless you know how many low cards have also left the deck. The whole point about counting is to know the estimated composition of the remaining deck.
Example: You are counting the tens. 6 tens come out of the deck. You assume a count of -6, and lower your bet. I am hi/lo. I see those 6 tens come out, and then 12 low cards. I KNOW a count of +6, and raise my bet to take advantage of it. Guess who is coming out on top
I had a dealer, who was watching me pull back as the tens had largely made their appearances actually told me "you don't want to take this next hit".She was right.And she could just have easily have been wrong. She doesn't know what the next card is. Neither does a counter. A counter just knows the estimated composition of the deck, and can vary their bet or use an "index play". IE: Basic Strategy says 12 vs. 2 is a hit, because that move is the best possible play statistically. But at a count > 0, it becomes stand, because now that move is the bes possible play statistically. That doesn't mean the next card is a 10. It just means that you'll lose less by standing than by hitting.
You cannot point to a single hand and use that as proof for anything. Remember, 12,000 hands is where "long term" begins. Everything else is indistinguishable from luck. If anything, the dealer was taking a blind shot hoping for a tip.
There is a romantic view of card counting. People assume it's a magical skill you pick up, then can just roll into a casino and use it as a personal ATM overnight.
The truth is, it gives you about a 1% edge over the house. That means that for every $10 you bet, you'll "earn" $0.10. You can get, maybe, 100 hands of Blackjack per hour on a good day.
And the "margin of error" (standard deviation) means that your long-term swings won't balance out until after about 12,000 hands. 120 hours of Blackjack, just to statistically be guaranteed to at least break even.
And all that is assuming you count perfectly, and play perfectly.
So after card counting gets hyped, you'll get a whole ton of people who want the quick win. They'll learn a quick hi/lo system. They won't practice. They won't learn basic strategy perfectly (quick, what's the proper move when you have 44 vs. a dealer's upcard of 5?). And they'll go into the casino. Maybe they'll double up quickly and walk away. More likely they'll just keep playing, have a few drinks, and either make a bit of money, or get frustrated and lose everything, or just play for a while and have fun. But in every case, they'll be playing with a disadvantage. Making a couple mistakes or missing a couple counts, maybe they're playing an even money game, or just 0.5% house edge. If they start steaming and making the big errors, they'll be giving the house 4-5% of their money on every hand.
And for the one in a thousand counter who does a good job and earns 1-2% on her money, they'll be 999 players who give it all back.
If card counting had the ability to destroy the casinos, they'd have been out of business a long time ago. Blackjack is profitable for the casino.
Not quite. The tracks will probably be a loss leader. The profit is going to come from two different places.
1) Dreamweaver webmaster
2) Keyboard cleaner (cheetos and pepsi and genetic splatter, oh my!)
3) Floating point wrangler
4) Monochrome wire detangler
5) Witnessing <body bgcolor="#FFFF00">
6) rpm dependency arbitrator
7) "Cowboy Neal option" writer
Just a note for job-seekers: the above position is yet to be filled. Please drop off resumes at /dev/null.
Unless it's the World Cup of Soccer-- in which case, your citizens will do it to themselves.
Someone downloading the free version != a non-purchasing customer
Though, I also have translucent diagonal lines that run across the screen that remind me of a CRT projector than needs its edges blanked. And the pincushioning has always been off. And on a cold day, I have to turn the contrast waaaaaaay down to keep it from shutting itself off. But aside from that, best $20 I ever spent.
I'll be seeing you in court, Java's Object
And why shouldn't some malicious attacker use every single exploit they've ever written/downloaded/stolen in their botnet payload? It's not like they're really concerned about bandwidth or CPU loads.
What? Dude! Everyone knows who rms is. He's the hacker character from Niven, Pournell and Barne's Fallen Angels
It's not in an abandoned nuclear bunker. It's in an "abandoned" nuclear bunker. Wikileaks goes chugging along-- but if anything ever hits their servers that's a bit too sensitive for the US government, someone presses a button, and the "abandoned" nuclear weapons in the bunker go "foom".
You mean emacs AND vi?
The grain of salt here is that the date was predicted by a 2008 AI which, as we all know, are not anywhere near as smart as a 2009 AI.
If I were an advertiser, I'd be concerned about the possibility that I run a good ad campaign, and got so many clicks my adserver got Slashdotted
If you can create features where I rank my social insurance number against celebrity's horoscopes, and then play hangman with my bank account information against friends to see who is better, then I'm totally in!
And according to a Microsoft press release, they feel confident that there are key indicators signaling IBM's adoption of Vista for the new supercomputer.