Mystics & Statistics

I do state in War by Numbers that there were three validations of the QJM/TNDM although the first was not published. That is not entirely correct. It was not published as a stand alone validation, but significant parts of it was published. The actual engagements were all published as part of the Combat Data Subscription Service. It was eight volumes, with the first volume published in 1975. In there it listed all the engagements used by the QJM. For example (page 6):

Now, the actual results of these test runs was published in his 1977 book Numbers, Predictions and War (NPW).  As Trevor Dupuy specifically notes on page 58 of his book:

Appendix B contains a consolidated summary of HERO’s QJM Engagement Data Base. The first 8l examples in this consolidated statistical comparison show the theoretical results and actual results of these 81 World War II engagements (60 in the Development Data Base, 21 in the Validating Data Base; 61 in ltaly, 19 in northwest Europe, and 1 in Russia). In all of these the P/P value reflects an average German combat effectiveness superiority factor of about 23 percent.”

This was done back in the day when Data Base was two words and could exist on paper, vice a computer.

Anyhow, in Appendix B of NPW is “HERO’s QJM Data Base (as of May 1977)”.  The specific engagement in question is given as:

No.     Year & Date          Battle Designation         Force X Designation     Posture

1.         1943, Sep   9-11     Port of Salerno                B  46 ID                               A

                                                                                                                                 Air %:

Force Y Designation       Posture            Na            Nd               S/S          W   W     P/P

G   16 PzD                            PD                     12,917       4,250         1.83          0     22   0.73

                                                                % cas/day

Surp     P/P          PR/PR      CEV       x         y             I          I           SE         SE

1.5          1.10          0.87          0.79      3.51    0.94      7.4      2.3       1.02     3.85

I left out the subscripts. But one can see Appendix B (pages 234-235) for these details.

Now, this does not still actually do a direct validation in that it compares model results to actual historical combat results, but one can see the data they used for their inputs and what the outcomes were of these engagements.