8 situations, Previous, and last outcome: Machine picks.
AAA, A ,A : A
AAB, A, B : A or B
ABA, B, A : A or B
ABB, B, B : B
BAA, B, B : B
BAB, B, A : A or B
BBA, A, B : A or B
BBB, A, A : A
This way the machine is able to outguess you.
53% of the time instead of pure randomness (50%)
This is simple frequency of sequence patterns.
https://norvig.com/mayzner.html
Also we can see the letter "e" is the most likely letter.
Then if we have to guess a two letter sequence knowing "e" is the first,
We can then look for the most likely two letter sequence beginning with "e"
It's said :
1. er
2. en
3. es
4. ed
A quote from a famous guys say something like:
You won't need more than 4 variables, parameters.
The above machine has 3 variables,
Each variable has 2 states.
2^3 = 16 outcomes.
It's a lot already.
I used to think about markets this way,
2 variables: Order flow and liquidity,
2 States: Negative or Positive.
4 outcomes:
+ Order flow and + Liquidity -> Bullish
+ Order flow and - Liquidity -> Neutral
- Order flow and + Liquidity -> Neutral
- Order flow and - Liquidity -> Bearish
--> 50% of the time the market is not trending.
But one can take highs and lows for example:
2 Variables: High and Low
3 States: Inferior, Equal, Superior
8 outcomes :
Inferior High Inferior Low
Inferior High Equal Low
Inferior High Superior Low
Equal High Inferior Low
Equal High Equal Low
Equal High Superior Low
Superior High Inferior Low
Superior High Equal Low
Superior High Superior Low
AAA, A ,A : A
AAB, A, B : A or B
ABA, B, A : A or B
ABB, B, B : B
BAA, B, B : B
BAB, B, A : A or B
BBA, A, B : A or B
BBB, A, A : A
This way the machine is able to outguess you.
53% of the time instead of pure randomness (50%)
This is simple frequency of sequence patterns.
https://norvig.com/mayzner.html
Also we can see the letter "e" is the most likely letter.
Then if we have to guess a two letter sequence knowing "e" is the first,
We can then look for the most likely two letter sequence beginning with "e"
It's said :
1. er
2. en
3. es
4. ed
A quote from a famous guys say something like:
You won't need more than 4 variables, parameters.
The above machine has 3 variables,
Each variable has 2 states.
2^3 = 16 outcomes.
It's a lot already.
I used to think about markets this way,
2 variables: Order flow and liquidity,
2 States: Negative or Positive.
4 outcomes:
+ Order flow and + Liquidity -> Bullish
+ Order flow and - Liquidity -> Neutral
- Order flow and + Liquidity -> Neutral
- Order flow and - Liquidity -> Bearish
--> 50% of the time the market is not trending.
But one can take highs and lows for example:
2 Variables: High and Low
3 States: Inferior, Equal, Superior
8 outcomes :
Inferior High Inferior Low
Inferior High Equal Low
Inferior High Superior Low
Equal High Inferior Low
Equal High Equal Low
Equal High Superior Low
Superior High Inferior Low
Superior High Equal Low
Superior High Superior Low
Last edited:
