Surely "streaky" is rigged!?!

[steadman]

Obviously I'm not a big poster. I don't have much to say which isn't already taken care of on here by some great contributors but I'm constantly reading about how games are not rigged but are "streaky" which doesnt make much sense.

How can a person believe that a game isn't rigged if they believe it is streaky? I'm talking here not about slots, keno, etc but the standard card games. Even it is still paying the same percentages in a streaked fashion then that is still manipulated thus "rigged".

I only mention it because I've had a streaky day on Pai Gow - I'm always amazed how many you can lose in a row in that game - especially as the chance of a loss is far less than on say a blackjack game!

Cheers guys.

Cheers Simmo.

 

[Simmo!]

I'm not a table games player but I'd guess (like slots etc) its down to "high variance" rather than any dubious activity causing the streaks, in particular when multiple decks are used. Just a hunch though. The audited payout percentages might also hold a clue to this too where published.

Cheers

Simmo!

 

[KasinoKing]

Hi Steadman,

What software do you play on?
I play quite a bit of PaiGow and know what you mean about the streaks. It's easier to think it's rigged than just having a bad run.
On Crypto's I have years of stats which shows me in the long term I get a fair game there.
But with MG & RTG I always seem to have terrible luck - MG is actually the worst for PaiGow I've ever played!
Boss & Playtech seem OK to me.

I guess the above is just my gut feel - it could all just be down to luck.

 

[steadman]

Hi guys,.


I'm not really suggesting that the games are - more commenting on a some of the posts in the forum - I notice on multiple posts there are the "not rigged just more streaky" comments which just doesnt make sense.

KK - I play at quite a few different casinos only if bonuses are involved. I'm personally just amazed at the amount of Pai-Gow hands you can lose in a row - even get a "feel" for whats happening before it happens sometimes but I'm not suggesting anything odd - I've had those feelings at a live table but Pai-Gow, unlike BJ, should not really contain many of these multiple loss streaks.

Cheers.

 

[bpb]

Hi guys,.


I'm not really suggesting that the games are - more commenting on a some of the posts in the forum - I notice on multiple posts there are the "not rigged just more streaky" comments which just doesnt make sense.

You are correct, streaky but not rigged comments make no sense. But people like to make up these kind of arguments to justify their gambling losses. It gives them hope that they can win in the long run if they just identify a streak early and up their bets.

 

[aka23]

You are correct, streaky but not rigged comments make no sense. But people like to make up these kind of arguments to justify their gambling losses. It gives them hope that they can win in the long run if they just identify a streak early and up their bets.

Gambling losses? I gain more than statisically expected almost every time I use progressive betting with SW I believe is streaky. The results for casinos where I have used progressive betting this month are below:

--Parlay--
Winward: +$950 on $500 bonus
Casino Vega: +$160 without betting or cashing bonus

--Real Time Gaming--
Shark: +$820 on $400 sticky bonus (or +$1220, if including bonus)

--Cryptologic--
Intercasino: +$200 on $100 bonus
Littlewoods: +$195 on 25GBP bonus
William Hill: +$75 on $40 bonus

--Grand Virtual--
Casino Lux: +$200 on $100 bonus (should have been more, lost nearly all bonus bets, as is typical with this SW)

--Random Logic--
888.com: +$165 on $100 bonus

 

[tencardcharlie]

I'm personally just amazed at the amount of Pai-Gow hands you can lose in a row - even get a "feel" for whats happening before it happens sometimes but I'm not suggesting anything odd - I've had those feelings at a live table but Pai-Gow, unlike BJ, should not really contain many of these multiple loss streaks.

Yeah it's easy to feel that way when the game seems to be against you
But as you understand, theories about 'streaky' software doesn't make sense.

Not only is it theoretically implausible that casinos should rigg their games in such a wierd way, also the lack of empirical evidence speaks for itself.

To my knowledge noone have ever shown that there should be higher risk to lose after a loss and higher chance to win after a win - which would be true if the SW was programmed to be 'streaky'. Such a test would not be hard to perform. Still we see nothing but anecdotes to back up those claims.

 

[aka23]

Yeah it's easy to feel that way when the game seems to be against you
But as you understand, theories about 'streaky' software doesn't make sense.

Not only is it theoretically implausible that casinos should rigg their games in such a wierd way, also the lack of empirical evidence speaks for itself..

There are several casino advantages to "streaky SW." Two of the big ones are

1. Increasing odds that a player will bust before completing wagering requirements

2. Increasing odds that a player who tries to make up losses by increasing bet size will bust (for example double bet with each loss)

It is also possible that the streakyness is more of a side effect than a desired effect.

To my knowledge noone have ever shown that there should be higher risk to lose after a loss and higher chance to win after a win - which would be true if the SW was programmed to be 'streaky'. Such a test would not be hard to perform. Still we see nothing but anecdotes to back up those claims.

Some of the anecdotes are extremely improbable situations. Yesterday I mentioned two roulette streaks that occurred minutes apart from each other , each of which is expected to occur once per 1.8 million spins. The chances of two such streaks within a few hundred spins of each other is for all practical purposes 0.

 

[frufrugirl]

AKA23, I noticed you list some of your bets in GBP. Arnt you a us player? If so, have you had any trouble from the casinos regarding using non resident currency?
And, back to the thread, I have to tell you I spent over $250 on a downward streak that wouldnt reverse itself, I cant believe I was such an idiot and continued hoping it would, the play was super fast -which, I had notice before when playing at this casino means a losing session...maybe its just in my mind but I had won on the previous deposit( no bonus),it was playing slow, cashed out $75 , came back the next day and deposited the $250 and pffft gone with variable betting without such as a howdy do! Now I pay attention to the speed in return play, if fast, it seems, I say "seems" to indicate a losing streak and I will stop. Saved my butt the other day!

 

[steadman]

Yeah it's easy to feel that way when the game seems to be against you
But as you understand, theories about 'streaky' software doesn't make sense.

.

Well it does make sense to an extent - streaking the game does appeal to the gambling instinct at it's most base level. A bad streak causes a faster bust and a good streak can ensure a player will play it out resulting in a longer session which can ultimately break those gamblers without a self-control reflex, so to speak. It's difficult for most people to leave when winning. Maybe not Casinomeister readers.

Take those scratchcard games you see on some sites - they are far from random - how many of them have 2 of the 3 required jackpot or other big win symbols? They are engineered to make people play more, so there are reasons to engineer games to be more appealing.

But basically my point was merely if players believe that a game like BJ/PG/CW is streaked in any way and not random - then they have to believe it is rigged. You cannot think one and not think the other.

 

[aka23]

AKA23, I noticed you list some of your bets in GBP. Arnt you a us player? If so, have you had any trouble from the casinos regarding using non resident currency?

The GPB bet mentioned above was at Littlewoods. Littlewoods requires all players to use GPB. Most of the other casinos that a play GBP or Euro at require this currency as well. Of the 2 or 3 where I selected GBP as an optional currency, none have given me any problems, including one where I have gained over $2500 over the past 2 months.

There are other casinos which have have special clauses about using foreign currency in their T&C. This might involve increasing wagering for bonuses, banning bonuses, or forbidding using a foreign currency alltogether (you can lose in a foreign currency, but you can't cash out winnings).

 

[tencardcharlie]

There are several casino advantages to "streaky SW." Two of the big ones are

1. Increasing odds that a player will bust before completing wagering requirements

2. Increasing odds that a player who tries to make up losses by increasing bet size will bust (for example double bet with each loss)

It is also possible that the streakyness is more of a side effect than a desired effect.

Actually, 1) increasing the odds for a player to bust before reaching the wagering requirements (by making the game streaky and not changing the HA) would benefit the player. And 2) would also benefit players who up their bets when they win, which would most probably make more people bet higher on winning streaks, giving them an advantage. I almost feel a bit sorry for casinos - if people have bad runs - it's rigged. If people have good runs - it's rigged. If there are runs that doesn't really affect the players results - it's definitely rigged.

Some of the anecdotes are extremely improbable situations. Yesterday I mentioned two roulette streaks that occurred minutes apart from each other , each of which is expected to occur once per 1.8 million spins. The chances of two such streaks within a few hundred spins of each other is for all practical purposes 0.

While this is extremely freaky, I would be more concerned if someone who made a proper experiment with the parameters decided before calculating probabilities and observed a 1 in 10.000 probability for a fair game. When thelawnet presented his test that showed how English harbour VP was unfair, I believed it immediately.

Freaky things will happen, and those will often be reported on boards. I've had my fair share of craziness to, like getting quads in holdem with both hole cards twice within three hands etc. etc. But looking at those events and then calculating the probabilities will always result in extremely significant results.

 

[aka23]

Actually, 1) increasing the odds for a player to bust before reaching the wagering requirements (by making the game streaky and not changing the HA) would benefit the player.

Not if it was a typical player using a reasonably constant bet size. Larger streaks increases variance. A higher variance increases the chance of busting before completing wagering requirements. The remaining players who do complete the wagering have an increased chance of continuing to play and lose their winnings.

While this is extremely freaky, I would be more concerned if someone who made a proper experiment with the parameters decided before calculating probabilities and observed a 1 in 10.000 probability for a fair game. When thelawnet presented his test that showed how English harbour VP was unfair, I believed it immediately.

How much of a "proper experiment" do you need? A player posted two screenshots of the same number being hit 5x in a row (occurred minutes apart from each other). Unless he played a billion spins before this without seeing this issue, it occurred far more often than expected.

As mentioned in that thread, I have also observed the same streakyness in that game several times. While I haven't seen a number 5x in a row, I have seen 5x in 6 spins multiple times, which has odds of 1 in ~37^4/5 = 1 in ~400,000.

 

[Simmo!]

Dunno about the "streaky" bit Steadman but 3 of the posters in this thread live on the 1066 line - spooky

 

[vinylweatherman]

Interesting.

Although the games are said to be random, there have been a number of posts from players who have won big with progressive betting and been denied the payout on grounds ranging from "robots" to "manipulating the game" (which takes place on the users PC, doesn't it ).
I have also seen "playing too fast".

Surely, if an operator is prepared to consider progressive betting as "abuse" they know something about the game software that is less than random and suspect the player has discovered how to ride the crest of the streaks.
Looking at audited payouts will NOT show whether the game streaks or not, it will simply show it pays out an amount that is consistent with its paytable and structure.
It is entirely possible to streak the output from an RNG by including feedback processes in the programming where the RNG output is resolved into game results. Only long term analysis of individual game outcomes will shed any light on this, and it is near impossible to extract data from most game logs in a machine readable form that can be bulk processed by an analytical program (unless the player is prepared to manually transcribe a blow by blow log into Excel, or cut & paste many short sessions).

 

[mgibson99]

But basically my point was merely if players believe that a game like BJ/PG/CW is streaked in any way and not random - then they have to believe it is rigged. You cannot think one and not think the other.

You are way off base here. Losing (and winning) streaks have absolutely nothing to do with a game being rigged. I have played blackjack in casinos all over the world and have experienced many odd streaks. I have had dealers get 5 blackjacks in a row. I have had streaks where everytime I hit 19 the dealer hits 20, and everytime I hit 20 the dealer hits 21. I have also had streaks where I couldn't lose for extended periods of time. That's just the nature of the game. I have never thought that the games were rigged.

Same goes for on line casinos. There are streaks, but that does not mean that the games are rigged. Now, I certainly can't say that no online casino is rigged. A particular casino very well could be. I have no idea. I just know that it is a completely false assumption to state that if you believe blackjack can be streaky (which it most certainly can), then you also must believe that the game is rigged. Absolutely no correlation between the two, unless you play an incredibly large number of hands and can empirically prove that the results are statistically unacheivable.

 

[steadman]

You are way off base here. .

You have missed the original point of my post.

I didn't say that if a game is streaky it is rigged!

I was making a comment on various posts which comment, for example, that "casino A BJ is streakier than casino B". My point was that if someone believes this to be the case then that person is making an assertion that either casino A or B is manipulating the game and thus that person is also asserting one of them is rigging the game.

I was just looking for discussion on this point.

 

[mgibson99]

You have missed the original point of my post.

I didn't say that if a game is streaky it is rigged!

I was making a comment on various posts which comment, for example, that "casino A BJ is streakier than casino B". My point was that if someone believes this to be the case then that person is making an assertion that either casino A or B is manipulating the game and thus that person is also asserting one of them is rigging the game.

I was just looking for discussion on this point.

Fair enough, but if that was your original point, you weren't very clear. This is how you started your post:

I'm constantly reading about how games are not rigged but are "streaky" which doesnt make much sense.

How can a person believe that a game isn't rigged if they believe it is streaky?

To me, the implication that you made initiallly is that any game that is streaky must be rigged. That is a different than saying that if one casino is more streaky than another, then it must be rigged (which I still say would be hard to prove without a large sample size and rigorous statistical analysis).

 

[steadman]

You are, of course, correct and I apologise.

All part of the learning process of using these boards.

Cheers

 

[KasinoKing]

You are way off base here. Losing (and winning) streaks have absolutely nothing to do with a game being rigged. I have played blackjack in casinos all over the world and have experienced many odd streaks. I have had dealers get 5 blackjacks in a row. I have had streaks where everytime I hit 19 the dealer hits 20, and everytime I hit 20 the dealer hits 21. I have also had streaks where I couldn't lose for extended periods of time. That's just the nature of the game. I have never thought that the games were rigged.

Same goes for on line casinos. There are streaks, but that does not mean that the games are rigged. Now, I certainly can't say that no online casino is rigged. A particular casino very well could be. I have no idea. I just know that it is a completely false assumption to state that if you believe blackjack can be streaky (which it most certainly can), then you also must believe that the game is rigged. Absolutely no correlation between the two, unless you play an incredibly large number of hands and can empirically prove that the results are statistically unacheivable.

100% agree; Streaky does not mean rigged.

Hey Simmo! - who's the 3rd 1066er here then?
Not you is it?

 

[tennis_balls]

whenever I spot more than two 1066 posters on one thread, I'm convinced that the thread is rigged.

You do not have permission to view link Log in or register now.

btw....does this mean there are Frenchy 1066ers as well?

 

[Simmo!]

100% agree; Streaky does not mean rigged.

Hey Simmo! - who's the 3rd 1066er here then?
Not you is it?

does this mean there are Frenchy 1066ers as well?

No...fortunately there's a big stream in the way

 

[henryVIII]

whenever I spot more than two 1066 posters on one thread, I'm convinced that the thread is rigged.

You do not have permission to view link Log in or register now.

btw....does this mean there are Frenchy 1066ers as well?

More posts like this PLEASE !!!

We need more of this crazy Tennis Balls humour.

 

[EasyRhino]

I think it's really a two-stage question:

1) Is the online casino more streaky than real life?

And there's real debate about this. Play at an online casino is so much faster than brick and mortar, and because people can do it in their pajamas, you're going to see a ton more hands online than you will in real life. And of course, the more you see of anything, you're going to see more real streaks. I mean, in aka's gambling log... imagine how many blackjack hands and roulette spins he's seen?

However, I personally think there is a decent chance that the random number generators used by online casinos are slightly defective. In that they may produce longer streaks than real life usual does. Heck, I've worked with a computer programmer who had to "improve" a random number generator he was coding because it seemed too streaky.

2) If the software is streakier, than was it made that way on purpose, or is it a "bug"?

For reputable casinos, this is a bug. Audited payout percentages are an example of this being the case. In most cases, the payouts you see for each family of games hover around the range that you'd expect.

 

[nafanny29]

Heck, I've worked with a computer programmer who had to "improve" a random number generator he was coding because it seemed too streaky.

Oh you mean the English Harbour guy right

Anyway my experiance of online BJ is that some casinos/software deal BJ that feels like its emulating real cards from a shuffled deck, and some deal BJ that feels like the RNG has decided the outcome and picks whatever cards it needs to achieve that result.

The "streakier" ones fall into the 2nd tpye.

Hope that makes sence.

 

[Zoozie]

First I must clear up a misunderstanding about RNG's.
I have fully confidence in RNG's used by ALL software providers. My ONLY worry about RNG's are that they are predictable(not random enough or RNG result can be pre-determined), but that would be a player advantage then! What can happen is cheating software. Ie. the software ask the RNG for a result and the software looks at the result and decide it didnt like it, and then it asks the RNG again and use that result instead. This scenario is indeed possible and is what has happened in the past with cheating software. The RNG has no idea what game(slot,BJ etc) the result is used for, therefore it is pretty obvious that the RNG can not cheat in any way. But in the intermediate SOFTWARE mapping the RNG result to game event, it is possible to cheat. This is another way to cheat instead of asking the RNG for another result, but again you have to blame the software. It is important to distinguish between the RNG and the software.

I have observed that BJ players are most likely to talk about suspicious winning/loosing streaks and blame the RNG/software. And when I play BJ - I often get the same feeling and therefore fully understand why a lot of people feels this, but I have a rational explanation for this. (It has happened at MG software which I am sure is not cheating). Because BJ is a kind of 50/50 win/lose game - players that lose 10 times in a row start shouting rigged. BJ is the best hand and people subconcience believe they won, but it is not really that rare it ends with a push instead and it often does and often several times in a row. Dealer hitting 21 starting with a 3 happens. Dealer having face card is common. Therefore there will be some sequences where dealer starts with a face card a lot more than expected. Also players does not seem to notice their winning streaks in the same way.
So basically because BJ is a low variance game with very few outcomes a lot of seemingly suspicous patters will inevitable emerge all the time. (*)

However getting 5 scatters/5 wild/5 retriggers in a slot game is much more rare but it does happen also and people at not complaining then. People blaming slots of long losing streaks should know better. Slots are designed this way and they are working as intended. This is the cost of being able to hit big win.(and facing a gigantic house egde at 5%+...)


(*) Most famous example is from 'Ramsey-theory' explained simple this way: Draw 6 dots (symmetric in a circle, to make it easy). Then connect ALL 6 dots with eigther a BLUE or RED line. When you are done there will ALWAYS be eigther a BLUE or RED 'triangle' Ie. there will exist 3 dots that all are connected with the same color.
Another way to express is that given ANY 6 people chosen, there will always be 3 that all know each other or 3 that all does not know each other. ('Know each other' is defined so the
both knows the other one). This example shows that given complexity enough, some patterns will always emerge.

Zoozie

 

[EasyRhino]

Incidentally, I just remembered one of the more famous exploits out there, Ron Harris, a computer programmer employed by the Nevada Gaming Control board, realized that the RNG used for some keno software could be predicted, and built a software program to determine the pattern. He and his accomplish were able to pick ALL of the numbers in a keno jackpot after just a few rounds.

Old / Expired Link

 

[Zoozie]

Incidentally, I just remembered one of the more famous exploits out there, Ron Harris, a computer programmer employed by the Nevada Gaming Control board, realized that the RNG used for some keno software could be predicted, and built a software program to determine the pattern. He and his accomplish were able to pick ALL of the numbers in a keno jackpot after just a few rounds.

This was also the case for some B&M VP machines. They could predict what cards would come on the draw. Not as usefull as the keno prediction though.

I know poker sites are now implemented so draw cards are decided 'on the fly'. (think of it as picked random from the deck instead of top cards, but the random factor now depends on the time you request the cards so you can not predict it)

So whenever any independant party approves a RNG for a software company, it really means NADA! They should test the software instead. English Harbour
probably had a perfect RNG, yet the software was rigged/flawed.

Zoozie

 

[Zoozie]

(*) Most famous example is from 'Ramsey-theory' explained simple this way: Draw 6 dots (symmetric in a circle, to make it easy). Then connect ALL 6 dots with eigther a BLUE or RED line. When you are done there will ALWAYS be eigther a BLUE or RED 'triangle' Ie. there will exist 3 dots that all are connected with the same color.
Zoozie

I was not precise enough it seems. You have to connect ANY two dots with each other, so you have to draw a total of 15 lines.

So the example shows that a predetermined pattern will always emerge. In BJ you play a sequence of hands and then you examine the data and find
patterns from the data. This is a much weaker requirement and therefore a lot more probable since you can try to find a pattern that match the data.

I have another more 'BJ' like example of patterns that will always emerge.

Write the sequence 1,2,3,4,5,6,7,8,9 where each number is eigther with a red marker or a blue marker.
Then there will always be a set of 3 numbers in the same color on a arithmetic progression. An arithmetic progression is a sequence of numbers that
differ by a constant amount. Ie 3,4,5 or 12,14,16 or 21, 26, 31, 46 etc.

Example: 1 2 3 4 5 6 7 8 9
Then you can find the arithmetic progression 1 5 9 in the same color.
Good luck trying to color the numbers 1-9 so there is no aritmetric progression of size 3
(if you only use the numbers 1 to 8, then it is possible to color them so there will not be an aritmetric progression of size 3).

So what does this have to do with BJ? If you play 9 hands of blackjack and consider them won or lost (push also lost etc.) then there will also ALWAYS be this artimetric progression of won or lost hands. ONLINE BJ MUST BE RIGGED - IT HAPPENS EVERY SINGLE TIME! Generally the theory says that an artimetric progression of any size (3 in this example) if the sample size is big enough (9 in this example).

Found this online article with more information and not too much math. (picture of nerd included)
Old / Expired Link

Zoozie

 

[happygobrokey]

(picture of nerd included)

 

[steadman]

As an aside, Ramsey Theory is accompanied by rigorous proving and does not, as far as I am aware, have any mention of any on-line casino's Blackjack algorithm no matter how many coloured dots I have joined up.

Interestingly, no online casino as far as I am aware has any rigourous proving therefore it would not be correct to have an acceptance of, for example, a Blackjack algorithm in any online casino as it would be correct to accept Ramsey Theory.

Furthermore, if we are talking mathetical proofs, surely anyone with a healthy questioning of any unproved system (i.e. An online casino BJ) are surely less delusional (in the scientific sense of the word you understand) than those that accept the systems and berate others for not doing so.

 

[AussieDave]

fortunately there's a big stream in the way

And I'm on the other side of that BIG pond

 

[Zoozie]

As an aside, Ramsey Theory is accompanied by rigorous proving and does not, as far as I am aware, have any mention of any on-line casino's Blackjack algorithm no matter how many coloured dots I have joined up.

Sorry, but I dont think you got the idea.

The whole point was when you play 9 hands of blackjack you can ALWAYS
find 3 hands that are all loss or wins that also form the arithmetic progression pattern described.

Instead of the red and blue numbers think of them as wins or loss.
Hand 1 won: make a red 1.
Hand 2 won: make a red 2
Hand 3 lost: make a blue 3

When you have 9 hands, there will always be the arithmetic progression of size 3 in the same color. That this pattern will emerge has nothing to do with blackjack, which was excactly what I wanted to state. Just a message to
those people that are looking for conspiracy everywhere.

Zoozie

 

[AussieDave]

Zoozie,

From my understanding there are two main types of processes used to generate a base seed.

In laymans terms:
1/ A base seed is extracted randomly from a frequency wave / atmospheric noise.
2/ A base seed is manufactured.

In regard to (1) it's my understanding that this type of seed produces a truly random outcome.

Where as (2) is know as a quasi RNG, where the outcomes are not truly random.

From what I've read RNG's just spit out gobble gook, it's the software that interprets the data and applies it (in this case) to cards, slots, dice or two flies crawling up a virtual wall.


A few Questions if you don't mind?

Which type of seed do casinos generally base their RNG's on? And which type of RNG do casino's use?



Cheers

 

[Zoozie]

Which type of seed do casinos generally base their RNG's on? And which type of RNG do casino's use?

Sorry, I do not know what kind of RNG the casino software does use.

But I have have heard of another type that uses radioactive decay to
generate the numbers and this is considered truely random also.

It is never the RNG I distrust, it is always the software.

Zoozie

 

[EasyRhino]

I've heard that a popular seed for software RNG's is the system clock measured down to a very small value, like milliseconds. That usually at least has the advantage of changing frequently.

 

[GrandMaster]

Read

You do not have permission to view link Log in or register now.

why the system clock is not a suitable source of randomness.

 

[steadman]

Just a message to
those people that are looking for conspiracy everywhere.

Zoozie

But surely the results of the Ramsey Theory would be applicable whether the RNG was random, non-random or the casino software was dealing a crooked game - so, in a sense, it's also a message for those looking to believe any software that is laid in front of them as being totally honest.

 

[frafi]

Rainbow ramsey theory

Hi Zoozie,

in the article in Old / Expired Link Jacob Licht gave this example:

"EXAMPLE (n = 35): 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
This example has a four-term monochromatic (blue) arithmetic progression: 6, 13, 20, 27, with a common difference of 7."

I cannot figure out why he chooses this 6, 13, 20, 27 progression because looking into the data I think the following arithmetic progression is much better here:
15, 18, 21, 24, 27, 30 with a common difference of 3.
It gives me 6 terms here.

Another would be 10, 15, 20, 25, 30 with a difference of 5 and 5 terms.

3, 9, 15, 21, 27 with a difference of 6 has also a better spanwidth.

Do you know why excactly this one?

Franz

 

[Zoozie]

Hi Zoozie,

in the article in Old / Expired Link Jacob Licht gave this example:

"EXAMPLE (n = 35): 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
This example has a four-term monochromatic (blue) arithmetic progression: 6, 13, 20, 27, with a common difference of 7."

I cannot figure out why he chooses this 6, 13, 20, 27 progression because looking into the data I think the following arithmetic progression is much better here:
15, 18, 21, 24, 27, 30 with a common difference of 3.
It gives me 6 terms here.

Another would be 10, 15, 20, 25, 30 with a difference of 5 and 5 terms.

3, 9, 15, 21, 27 with a difference of 6 has also a better spanwidth.

Do you know why excactly this one?
Franz

I must compliment you in your findings, you definately understood it.

First of all the number 35 is the minimum number such that there will always be an arithmetic progression of size 4. So take the challenge to find a 2-colouring of the numbers from 1-34 such that it does not happen, but a such does exist or 35 would not be the minimum number. It is not mentioned that 35 is the minimum number needed to find one of size 4 like 9 is the minimum number needed to find one of size 3. But it is so.

The 2-colouring in the article is not a good example since there even exist an arithmetic progression of size 5. Maybe he chose a random coloring and did not notice the one of size 5 or believed it was not important as this can also happen. The important fact was that there will always be one of size 4 (at least).

Generally these minimum numbers are damn hard to find and not many are known. To always find one of size 5 (for a 2-colouring) the minimum number is 178. Some years ago
it was not know what the mimum number was if you wanted an arithmetic progression of size 6, but maybe it has been found since.

So
N=3: 9 numbers is needed
N=4: 35 numers is needed
N=5: 178 numbers is needed.
N=6: ??? numbers is needed.


Zoozie

 

[ace4suited]

an interesting thread so far

firstly i would just like to thank those that have contributed to this thread so far - very interesting even if my brain is hurting a bit.

Much of this thread is about empirical evidence running against detailed RNG theory.

It is clear that it would be really hard to have any sort of satisfactory proof of whether a game has greater varience than what it should. It is also the case that "streaks" are hard to interpret by the player even if they did exist.

I am a BJ and Poker player primarily although i do dabble in VP and slots now and then but when it comes to streakiness I have only really got enough playing time to look at BJ.

If we take the software AND the RNG and take the combined effect of them then impirically there are some sites that seem "as a player" to be streaky.

I will say that 99% of talks of streakiness are about people complaining about losing. I do not see streakiness if it really exists as a bad thing but rather something that the player can exploit.

On some softwares myself and several other BJ players I speak to have experienced apparent streakiness on several sites. When we are in these negative streaks we bet less and in the positive streaks we bet more. Over the course of wagering 20,000 in bets on such a site between 1-20$ the "streakiness" factor does seem on some providers to have a positive player effect. This seems to be the strongest basis for supporting the idea that streakiness is real that some players (myself included) seem to be able to exploit in on certain sites for profit.

If it exists is it something that casinos do deliberately? I would say probably not but whilst casinos etc are often audited for the extent to which they are paying out appropriate amounts I do not think that much of the effort goes into checking that the varience is 100% as expected.

The combination of the RNG and the software that creates a BJ hand is not the same exactly from one software provider to another. If we say streakines does exist (it is highly debatable that there is sufficient proof) then it could be argued that it makes the software "imperfect" but this is akin to a roulette wheel not being 100% balanced. It is something that can only really affect a knowledgable player in a positive way.

Finally as a player i would say that streakiness is a bit like the "god gamble" i.e. if it doesnt exist then it has not done me any harm but if it does exist....

What this means is that if you sense the casino is really streaky then it does no harm to adjust your bets accordingly. It cannot effect your EV in a negative way. If however there is some merit in the notion of streakiness on a small number of sites then you have improved your EV.

 

[steadman]

Yes, it is an interesting discussion.

My point was in the original post (and explained further about three posts down) is that a player cannot say that one casino's set-% game is streakier than another casinos equivalent game - by default you are saying one of them is rigged. Doesnt matter whether you can take advantage of the streaky nature, doesnt even matter if one is actually streakier or it is just perception, doesnt even matter if one of the games is set to payback 200% - you are saying one of the games is rigged "in your opinion".

If I run a totally above board BJ simulator you can show Ramsey in the results, if I run a smart BJ simulator, say dealing smart seconds based on the player betting strategy, you will still find Ramsey in the results. This is what I understand!