Clustering Analysis vs Clustering Illusion

[gizmotron]

nolinks://en.wikipedia.org/wiki/Data_clustering

Clustering Analysis

Cluster analysis or clustering is the assignment of objects into groups (called clusters) so that objects from the same cluster are more similar to each other than objects from different clusters. Often similarity is assessed according to a distance measure. Clustering is a common technique for statistical data analysis, which is used in many fields, including machine learning, data mining, pattern recognition, image analysis and bioinformatics.

nolinks://en.wikipedia.org/wiki/Clustering_illusion

Clustering Illusion

The clustering illusion refers to the tendency to erroneously perceive small samples from random distributions as having significant "streaks" or "clusters", caused by a human tendency to underpredict the amount of variability likely to appear in a small sample of random or semi-random data due to chance.

[bombus]

Are the clusters and parcels, etc, self manifesting, do they show themselves to the trained eye, or do you start with a breakdown of the game into a matrix of preconceived clusters & parcels, then slot each result into those boxes as they occur?

[gizmotron]

Are the clusters and parcels, etc, self manifesting, do they show themselves to the trained eye,

I'm not sure what "parcels" are. I have this to go on: "Something wrapped up or packaged; a package. A plot of land, usually a division of a larger area."

The clusters do show themselves to the trained eye. In fact that is the best way to detect them. That's how I do it before I rip the head off of cats.

[bombus]

Oh please don't tell me where you put those heads... ouch!

[gizmotron]

Oh please don't tell me where you put those heads... ouch!

Just kidding. But is it OK to sacrifice chickens when I play?

[bombus]

Just kidding. But is it OK to sacrifice chickens when I play?

Ah, the old blood splat method, ay... now that's random.

[gizmotron]

Ah, the old blood splat method, ay... now that's random.

I gave up on reading the entrails of a baloney.


oops! baloney is entrails, yuk!

[madupz4]

nolinks://en.wikipedia.org/wiki/Data_clustering

Clustering Analysis

nolinks://en.wikipedia.org/wiki/Clustering_illusion

Clustering Illusion

Is it not unreasonable to think of ALL streaks as clustering illusion b/c with roulette being a game of chance, the streak could end at any time? 

For example if 3 reds show in a row, the size of the sample is not sufficient to accurately predict or "guess" the next outcome, however if you have 7 reds in a row........ one could argue that this is a sufficient streak and could make an educated guess that reds would continue.  However, we all know that in roulette anything can happen at anytime, on the 8th spin a black could show just as easily.

How can this dilemma be overcome in regards to reading and using random to make a consistent profit?

[gizmotron]

How can this dilemma be overcome in regards to reading and using random to make a consistent profit?

By noticing if the trends are telegraphing their moves. I've described it briefly as the global effect. Now only a person that has a lot of experience can see how this happens sometimes. You've been told to look for it. Sometimes A generalized trend of sizes / lengths flows as a trend outside each trend. You can use that when it happens. Also, you can test streaks by getting in on them after three repeats and not seven. You are going to win/lose half of all your streak starts. In ECs that a break even. Anything after that is going to be great. Also, dominance is a winner too. If you get 20 reds and 2 blacks you still have a winner in the dominance game. Betting on randomness is only a game of hunting and testing the waters. There will be a trend where all your first bets are losers. Now can you think of a strategy that deals with that? I would change before letting trigger based betting kill me off with a deep draw down.

[potatochips]

There will be a trend where all your first bets are losers.

What the author forgot to mention, by the time you find a trend you will have lost your bankroll  

Gizmotron has a lot of toasts to eat before giving winning recommendations.

[gizmotron]

I recommend the even chance bets. You can use any 18 numbers to create them. The table offers 6 different groups. I would add at least six more created from any process that is easy to track. Only the even chance bets allow you to react one spin at a time. With two sets of dozens you need three spins to balance. With the EC's you only need two spins to have balance.

[gizmotron]

-