Roulette Ask Me Anything About Betting the EC's (Even Chances)

Yes, being in a hurry or drinking while posting are two big mistakes which I've done plenty myself too.

 

Good luck with that.

You are looking at the overall probability of "X heads in a row" in a set of all possible outcomes BEFORE THE SEQUENCE STARTS.

You are not updating your probabilities as the actual events unfold. Which is why I explained the NASA mission success probability. Once takeoff is successful that probability is now 1 and no longer contributes to overall mission failure.

Before you start a sequence the probability of getting 4 heads is:
0.5 × 0.5 × 0.5 × 0.5 = 0.0625

After the first flip, let's say you got a head, your remaining probabilty to complete the 4 head sequence is now:
1.0 × 0.5 × 0.5 × 0.5 = 0.125

But let's consider what happens if you get a tail on the first flip. The probability of getting four heads is now zero as it's now impossible to get four heads in a row in the remaining 3 spins. When we update the probablity to the known fact we see that it agrees:
0.0 × 0.5 × 0.5 × 0.5 = 0.0

Let's assume you got the head on the first flip and now after the second flip, let's say you got another head, your remaining probabilty to complete a four head sequence is now
1.0 × 1.0 × 0.5 × 0.5 = 0.25

Or do you say it is less likely? If you say it's still 0.0625 how would you resolve this position if someone came up and asked you whom had not seen the first flip what is the probability of the next two flips being heads? And what about the hundreds of flips before you turned up? What would that guy say to you? Do you think all the past flips affect the future flips? If you think that then your sequence probabilty calculation should never be using 0.5 for each flip.

Logical checkmate, but let's continue..

At the third flip, let's say you got another head, your remaining probabilty to complete the 4 head sequence is now
1.0 × 1.0 × 1.0 × 0.5 = 0.5

So we see the last remaining flip the probability is indeed 0.5 which is what you used in your original calculation to get the overall probability of four heads in a row.

You can't keep saying your probability is at the start before the first flip, you must update your probabilty as each event unfolds.

If you're still not convinced with the logical and mathematical argument the test it.

Flip a coin 3 times and see if the forth flip after 3 heads (or 3 tails) makes a head (or tail) less probable. Tally up the results of the forth flip at least 100 times getting to three heads (or tails) in a row. You will see that every time you get to 3 flips the forth flip will be very close to 50/50. The coin does not know the past.

 

Because the jizzatron hasn't taken his meds and he is on a tilted rampage.

 

Luckily we play in the extreme short-term and not in the long term. In the short-term anything can happen. I've told this story many times. Years ago at the MGM there were 12 or 13 Reds showing on the Marquee and here comes this guy running up dragging a suitcase behind him. And bets $100 on black. To make a long story short he drops $1,500 with a Marty and lost every time. He looked like a balloon somebody let the air out of. This is back in the day when I was still trying to wise people up, which I don't do anymore. I told him if you follow the streak you can only lose once but if you bet against it like he did you can lose many times. He looked at me like I was the most insane person he'd ever seen. He said that was his entire bankroll for the weekend and the casino just cheated him out of it because 18 reds in a row is impossible. True story.

 

Probability doesn't require the short term to be even, fair, nice to you, balanced or ever catch up, and also doesn't require the long term to be fair or perfectly even handed either, unlike what you might have been told or imagine to be true, there is no hidden force or hand keeping things tit-for-tat even or fair.

The long term is also cruel to expectations of fairness, and we will witness a much larger difference in absolute counts as outcomes actually DIVERGE further from even. It is only the AVERAGE or ratios that converge as a percentage of the total trials.

The standard deviation for a coin flip is 0.5 × √n. You can work out how far it can drift from fair yourself.

After a million flips/spins you may have up to a thousand less heads vs tails and still be within the expectations of 1 standard deviation. The universe doesn't care about squaring up the counts. 1000 parts per million is 99.9% fair.

Now do a billion flips/spins and we can expect the difference in counts of heads vs tails can be in excess of 30,000 and still be within 1 standard deviation. So where's that long term fairness now?

Clearly the universe isn't making up for the past outliers and it's actually looking worse as things go on.

As a percentage ratio it's now converged to 99.99% fair but the difference in absolute counts has diverged further and will continue to do so.

I hope the above is useful and enlightening to understand the actual reality of the math vs the assumptions and wrong intuitions most people have of random.

 

@Georgie, think about this from another perspective. If your calculation is correct, it should apply to shorter streaks too, right? Example: the chance of 2 HEADS in a row is 0.5^2 = 25% or 1/4. Following your logic, it must mean that if you see just one HEAD, the chance of the next outcome being TAILS is 75%, since the chance for HEADS to continue the streak is 25% (the chance for either one or the other outcome must be 100%).

This also applies to TAILS, so if you see one TAIL you should bet HEADS, because your chance of winning is again 75%.

Now apply this to roulette even chances, where you actually get paid for making the right call. Ignore the zeros for simplicity and it means merely by betting the opposite of the last outcome, you will win 75% of bets, for which you are paid even odds.

Is it possible that this strategy has been overlooked for 100s of years, if it worked? Of course not, that's absurd. Not to mention the countless other strategies that would also work just by waiting for "virtual losses". And that's the point; if your calculation or argument results in absurd conclusions that tells you that it must be invalid.

 

Yes, but apparently not for you, just everyone else.

 

Anybody who's concerned about the long-term should never be allowed near a casino. Everything in the casino happens in the extreme short-term which has nothing at all to do with the long-term. The casino does not allow you to make long-term bets so why even discuss it. The long-term has nothing to do with anything.

 

Nope. You should be following the casino's example and be very concerned about the long-term. That's what they do and they're very successful. They don't care about the transient ups and downs, because they know the edge is on their side and in the long run they'll be in profit. Players should likewise forget about short-term variance because it can't be avoided or exploited. Just focus on getting an edge (hint: you can't get it from just looking at past spins).

 

This is not exactly the full story as the casinos result is over a large number of players individual random walks. The ArcSine law plays a part in expectations for the player variance which is different to what the casino experiences on average over all players.

William Feller a legend in probability theory wrote in his book An Introduction to Probability Theory and Its Applications:

We shall encounter theoretical conclusions which not only are unexpected but actually come as a shock to intuition and common sense. They will reveal that commonly accepted notions concerning chance fluctuations are without foundation and that the implications of the law of large numbers numbers are widely misconstrued. For example, in various applications it is assumed that observations on an individual coin-tossing game during a long time interval will yield the same statistical characteristics as the observation of the results of a huge number of independent games at one given instant. This is not so...

Contrary to popular opinion, it is quite likely that in a long coin-tossing game one of the players remains practically the whole time on the winning side, the other on the losing side.

In a long sequence of coin flips, the probability that heads leads 85% of the time is about 25.32%. This means that in over half of long, evenly matched games, one side will lead for at least 85% of the time.

Obviously the corrosive effect of house edge is ever present.

 

The casino plays in the long term so long term math applies to them. Single players play in the extreme short-term they won't live long enough to even approach the long-term. It's essential to understand this if you ever want to win in the casino. The long-term has absolutely nothing to do with your play. Steve Wynn has talked about this many times about how the long-term math is always in his favor. But short-term players can really screw up his profits in any one quarter, because with them anything can happen.

Hint: you can only beat by looking at past outcomes. I know this for an absolute ironclad fact because I do it every single day at multiple online casinos. When people like you tell me I'm not doing this or it can't be done, I simply dismiss you as ignorant. Because you obviously are. It's probably not your fault, you did not come up with this on your own. That's what you were taught. Because if you really researched it you would see how wrong you are. But because you're a math person being wrong is the last thing you want. You would rather look like a fool then be wrong.

 

Once you have your own edge over the casino, the house edge is about as bothersome as a tick on the ass of an elephant. It means nothing.

 

This is the lie that I expose as a lie in the RR thread. There is a reason for learning the truth. Meanwhile people still pass on the lie as truth. It's nothing more than a myth. The proof is one of the funniest of all proofs too. It's mathematical.

 

Wrong again. I used not to believe what the math said. I thought that there must be some way to pick more winners than losers, or at least reduce the losses. After extensive testing I realized the math is correct. The are objectively no sequences which are followed by more wins than losses when you take variance into account. The only explanation for your success, if it's true, is precognition. In which case, I have nothing more to add. I'm one of the mathboyz and you're one of the mythboyz. Let's leave it at that.

 

I've heard of this phenomenon but never tested it. Have you? 85% is a lot. Does this mean one side will be in the lead continuously for 85% of outcomes? That seems implausible. What seems more likely is that there will "crossovers" at times.

 

Which makes you wonder why mathZombies are here? It can't be that they think so highly of their intelligence that they need to wear the Spandex and cape and save poor gamblers.

Here's one for them. I was just kidding. Reading Randomness is a baloney festival and there is no way that it could ever work because of long term odds can never be overcome. Anyone spending time on it and winning so much that these believed odds are worthless, having almost doubled their take, are just living in a unicorn rainbow delusion. TwoUp Your Butt is right. That's his Indian name. He is so smart. Even little children admire him. Thank you for coming here and putting the light of day on our fantasies'. I will have to quit winning because some day I must lose it all flat betting. TwoUp Your Butt is my new North Star.

 

That's sort of revealing. It makes you think that possibly taking variance into account might be a mistake.

 

Here is a long ago stated truth about RR. You can lose all the bets that math says you should lose in a session and still come out ahead every time that you play to win using RR. If I play 100 spins and lose 53 of them, only winning 47 of them, by my knowledge and skill alone, I will win every one of those 100 spin sessions. The secret was shared in the Baccarat forum when Lungyeh shared it as strategic and necessary in his thread on playing Baccarat as a business. Several people have used the method in their own versions of beating the casinos. This is not a mindless progression. So don't get a bunch in your loin cloth TwoUp Your Butt.

 

They already look like fools and they are wrong. It's that magic mirror mirror on the wall. It makes them look like the emperor has clothes.

 

Watch this Spike and others that know what I mean. Have you seen any Elegant Patterns or the Global Effect in the past month? I doubt that Spike would deal with it because he goes after one net win and he is done. But he might have seen one of these anyway.