Roger, I need you to explain the scoring process again, step by step. I rewrote my code and I'm getting the same results.

Assuming the same jones probability (since the parameters never change) I am finding that no matter how many hits there are, the expected value, variance, the sums of each of these, and the standard deviation NEVER change.

Only the jones score sum changes, and the normalized score then becomes linear across the range of possible outcomes.

I rechecked against the description you posted and I'm following your directions. So maybe you miswrote something?

Brian

-----------------------------------------------------------------------------
reading file...
 3  H:/Geo/earthquakes/Catalogs/ANSS catalog/ANSS 6.5 and up 1980 thru 2012.txt
  Start Date: 2444239.50000000
    End Date: 2456293.50000000
   Time Span:  12055 days
total quakes:  1217
--------------------------------------------------
computing Jones probability...
      Window Size:  3 days
      No. Windows:  4018.3333333333
      Window Hits:  959
Jones Probability:  .23865615927001
====================================================================================================
JProb | hit | JonesScore | ExpectedValue |  Variance  |
-------------------------------------------------------
0.239 |  1  |  1.4327314 |   0.1343344   |  0.5284536 |  
0.239 |  1  |  1.4327314 |   0.1343344   |  0.5284536 |  
0.239 |  1  |  1.4327314 |   0.1343344   |  0.5284536 |  
0.239 |  0  | -0.2726702 |   0.1343344   |  0.5284536 |  
0.239 |  0  | -0.2726702 |   0.1343344   |  0.5284536 |  
-------------------------------------------------------
      Sums: |  3.7528539 |   0.6716720   |  2.6422678 | 
-------------------------------------------------------
       Standard Deviation:   1.6255054
-------------------------------------------------------
         Normalized Score:   1.8955224
====================================================================================================
new stuff....
jones score for a miss: -.27267019566049
jones score for a hit :  1.43273142706372
        expected value:  .13433440563172
              variance:  .52845356202693
predictions:  5
success | jones score sum |   expected sum  |   variance sum  |   standard dev  | normalized score | difference
----------------------------------------------------------------------------------------------------
     0  |   -1.3633509783 |    0.6716720282 |    2.6422678101 |    1.6255054014 |   -1.2519324788  | -1.25193247875338
     1  |    0.3420506444 |    0.6716720282 |    2.6422678101 |    1.6255054014 |   -0.2027808603  |  1.04915161844783
     2  |    2.0474522671 |    0.6716720282 |    2.6422678101 |    1.6255054014 |    0.8463707581  |  1.04915161844783
     3  |    3.7528538899 |    0.6716720282 |    2.6422678101 |    1.6255054014 |    1.8955223766  |  1.04915161844783
     4  |    5.4582555126 |    0.6716720282 |    2.6422678101 |    1.6255054014 |    2.9446739950  |  1.04915161844783
     5  |    7.1636571353 |    0.6716720282 |    2.6422678101 |    1.6255054014 |    3.9938256135  |  1.04915161844783
--------------------------------------------------
run time:   0.078
done.
--------------------------------------------------