Home
So I'm kind of ...crazy. Sometimes. I get bizarre ideas like "How do I calculate the IQ of the IQ test?"
The answer actually came to me today when I was discussing with a friend how vague or specific the questions of a test should be to accurately depict IQ as much as possible. At the worst vagueness, it's useless and an inapplicable skill. At the worst specific, it's only useful for THAT test and nothing else. I thought there might be a sweet spot, but it had to lean to the vague side otherwise it's too subjective.
Plotting vagueness = 0, specific = 9 on a horizontal line like so:
0, 1, 2, 3, 4, 5, 6, 7, 8, 9
I estimated that a 3 would be the best. I tried finding the mean, median, and range and then meaning that and inversing its square from the length of the string. That's derpspeak for I averaged it, a lot. The answer? Motherfucking 3. I was spot-on. I realized there was an error if I started with 0 because I couldn't find the true median, so I went 1..10 (the .. means inclusive sets) and as expected, it was 4. Since the vagueness was the floor and the specific was the ceiling, I could replace the ceiling with ANY number and calculate its "sweet spot". Which I'm now just calling IQ. So I then calculated the IQ of the IQ test (1..300).
I guessed 128.It was 128 dead on. In other words, 128 is the most desirable outcome for practical usage in every day life. You are above the cut of "normal" people but not prone to the insanities of the geniuses. This "sweet spot" was an estimate that I had used many times before by eyeing number sets that needed a low-end bias average, and I finally figured out the ALGORITHM my brain has been using to do this.
I wrote a code and plotted all of the ceilings divisible by 10 on a graph that I wanted to. Hundreds from 100-1500, also 50 and 10 randomly. In the code "w" is a variable for the ceiling.
____________________________________________________________________
This algoithm has some bizarre properties. Firstly ONLY use numbers divisible by 10, it doesn't work any other way. Secondly, it appears to go up by 4's when ceiling is below 10. Then it goes up by 42's, then by 428's. You'll notice ANOTHER pattern. Ceilings that are 700 away (like 100 and 800) are exactly the lower value + 300. Which I find odd. The pattern continues to repeat. 200 ceiling is an IQ of 85 while 900 is IQ of 385.
Here's a data table I made:
Crazy as FUCK pattern. But note THIS about it: At 10, it's 40%. At 100 it's 42%. At 1000 it's 42.8%
It keeps getting more accurate in its percentage of the number as it climbs. So the IQ of any number, or better to say:
The "sweet spot" of a range is approximately ~42.8%
Most people end up in the 33.33~% range, or 1/3.
I guess 42 IS the answer to life, the universe, and everything. And I just calculated the question: What is the universe's sweet spot?
The best approximation I've gotten yet is 42.85714%
The answer actually came to me today when I was discussing with a friend how vague or specific the questions of a test should be to accurately depict IQ as much as possible. At the worst vagueness, it's useless and an inapplicable skill. At the worst specific, it's only useful for THAT test and nothing else. I thought there might be a sweet spot, but it had to lean to the vague side otherwise it's too subjective.
Plotting vagueness = 0, specific = 9 on a horizontal line like so:
0, 1, 2, 3, 4, 5, 6, 7, 8, 9
I estimated that a 3 would be the best. I tried finding the mean, median, and range and then meaning that and inversing its square from the length of the string. That's derpspeak for I averaged it, a lot. The answer? Motherfucking 3. I was spot-on. I realized there was an error if I started with 0 because I couldn't find the true median, so I went 1..10 (the .. means inclusive sets) and as expected, it was 4. Since the vagueness was the floor and the specific was the ceiling, I could replace the ceiling with ANY number and calculate its "sweet spot". Which I'm now just calling IQ. So I then calculated the IQ of the IQ test (1..300).
I guessed 128.It was 128 dead on. In other words, 128 is the most desirable outcome for practical usage in every day life. You are above the cut of "normal" people but not prone to the insanities of the geniuses. This "sweet spot" was an estimate that I had used many times before by eyeing number sets that needed a low-end bias average, and I finally figured out the ALGORITHM my brain has been using to do this.
I wrote a code and plotted all of the ceilings divisible by 10 on a graph that I wanted to. Hundreds from 100-1500, also 50 and 10 randomly. In the code "w" is a variable for the ceiling.
This algoithm has some bizarre properties. Firstly ONLY use numbers divisible by 10, it doesn't work any other way. Secondly, it appears to go up by 4's when ceiling is below 10. Then it goes up by 42's, then by 428's. You'll notice ANOTHER pattern. Ceilings that are 700 away (like 100 and 800) are exactly the lower value + 300. Which I find odd. The pattern continues to repeat. 200 ceiling is an IQ of 85 while 900 is IQ of 385.
Here's a data table I made:
- Code:
*NOTE: The "A" before a number stands for power of 10. I use it so 1000+ won't go beyond 3 digits. A50 would be 1500. A00 is 1000 etc
A50 => 642
A40 => 599
A30 => 557
A20 => 514
A10 => 471
A00 => 428
900 => 385
800 => 342
700 => 299
600 => 257
500 => 214
400 => 171
300 => 128
200 => 85
100 => 42
050 => 21
010 => 4
Crazy as FUCK pattern. But note THIS about it: At 10, it's 40%. At 100 it's 42%. At 1000 it's 42.8%
It keeps getting more accurate in its percentage of the number as it climbs. So the IQ of any number, or better to say:
The "sweet spot" of a range is approximately ~42.8%
Most people end up in the 33.33~% range, or 1/3.
I guess 42 IS the answer to life, the universe, and everything. And I just calculated the question: What is the universe's sweet spot?
The best approximation I've gotten yet is 42.85714%
