# a good mythology /distribution to use here

• Sep 29th 2011, 06:36 AM
subs
a good mythology /distribution to use here
Hi all,

i have some data that looks like this:

Code:

probability    margin of error
0                0.618656356        0.024338561
no data
no data
-1.5                0.585742772        0.018216632
-2                0.576441103        0.019498585
-2.5                0.564943253        0.019138716
-3                0.518688024        0.012458096
-3.5                0.468685832        0.024736913
-4                0.453384419        0.024670001
-4.5                0.440914867        0.024914331

the aim is to estimate probability from +14 to -14. can any1 suggest some helpful resources or recommend a possible method to explore.

some1 said to try the weibull distribution, but after reading up on it i'm not so sure...

thanks in advance - most appreciated.
• Sep 29th 2011, 10:59 AM
SpringFan25
Re: a good mythology /distribution to use here
I assume tha the first column is X and the second column are cumulative probabilities P(X<x).
If so, the weibull distribution cannot be fitted to that data since it is only defined for positive values of x.

What exactly are you trying to do? i am suprised to find either the webull distribution or any distribution fitting excercise in a pre-univiersity thread.
• Sep 29th 2011, 03:06 PM
subs
Re: a good mythology /distribution to use here
Hi SpringFan25,

this isn't really an exercise, it's really a real life thing and i'm certainly feeling well out of my depth with it :(

I am a professional gambler, i make money from sports, consistently, over now 10s of thousands of bets. (it's hard work BTW).

my numbers for a certain combination of handicap and total may look something like this:

Code:

MOV  #  push %
14        14        2.08%
13        15        2.23%
12        7        1.04%
11        10        1.49%
10        28        4.16%
9        8        1.19%
8        11        1.63%
7        35        5.20%
6        23        3.42%
5        7        1.04%
4        22        3.27%
3        64        9.51%
2        13        1.93%
1        19        2.82%
0        0        0.00%
-1        16        2.38%
-2        12        1.78%
-3        64        9.51%
-4        21        3.12%
-5        13        1.93%
-6        16        2.38%
-7        30        4.46%
-8        6        0.89%
-9        5        0.74%
-10        12        1.78%
-11        3        0.45%
-12        6        0.89%
-13        8        1.19%
-14        16        2.38%

the numbers i posted in post 1 are a reliable sports book's implied numbers. what i'm really trying to do is improve on my numbers (small sample sizes and "windowing" in places) by using the relevant numbers from the sports book.

so far i'm not getting much of a fit (cos i'm doing it wrong no doubt). because of the way scoring works in the NFL, the numbers jump up and down and a plot looks like a child's drawing of a crown.

would u be able to point me in the right direction please, it is very important to me to get pretty close. so i really really appreciate it.

thanks.