Problem with Fractional Stats IV Calculation

[StarBP]

Metalkid's post about the role of fractional stats in IV calculation relies on flawed mathematics.
Using Metalkid's example of:

10 10.67-10.99
12 12.34-12.66
14 14.00-14.33
15 15.67-10.99

Let's see if that always works. Try using the formula f(x)=1.7x+8.9. Since Fractional Stats (I call it Fractionalization) are not formula-dependent, the numerical coefficient doesn't matter (Translation: It should work with ANY linear increase, not just ones which are actually possible in real Pokémon stats). Using this hypothetical formula, you get

10.6
12.4
14.0
15.7

which rounds down to

10
12
14
15

That is the same as the sample stats Metalkid used in his post, but 10.6 isn't between 10.67 and 10.99!

Another example: Try Fractionalizing this set:

46
49
51
53
56
58
60
63
65

These are actual Pokémon stats where the Base+IV+EV/4 (rise per level) is 2.32! In one extreme case, Fractionalization predicts a value of 65.63-65.75 for the last stat, and the actual value is 65.32, which is off by more than 0.3, almost 2 1/2X the predicted range! I do know that the problem is with predicting the rise per level, and not with the Fractionalization itself. With a known rise per level, Fractionalization works perfectly. Then again, with a known rise per level, perfect IV's can be calculated without Fractionalization.

 

[Great Sage]

Your point is?

 

[Peanut-Lover]

He is trying to state that the laws of statistics and mathematics present a fundamental problem to metalkid's ideas of fractional stats within pokemon.


If there is indeed a more complicated algorithm, by all means, I'm sure that it can be fitted in.

Don't come to Smogon with just problems - as in pointing out an error. Post what it should be, by demonstrating a new algorithm for calculating it more precisely. And while you give us the theories, the imperfection can be noted if you go to the thread, and it will be counted in in the next version.


Just saying...

 

[MetalKid]

I wonder if my flaw is that when you get the theoretical range, you have to minus one... So if it was at 10 2/3, then the range would be 10 1/3 to 10 3/3 (11).

 

[StarBP]

I wonder if my flaw is that when you get the theoretical range, you have to minus one... So if it was at 10 2/3, then the range would be 10 1/3 to 10 3/3 (11).

Great minds think alike. I thought that, too, until I did the second example in my post. One of the numbers was off by almost 2 1/2X the predicted range, much more than the 1X that that specific correction would require.