I asked Evan Jones and David Kahn to conduct an analysis of the ODS system. They did so and provided me with materials that I am using to apply a thorough investigation of the ODS mathematical model.
Right now I am going over the materials to make sure that I have fully verified the accuracy of the formulas so that I can review the recommendations. It is a slow and difficult process, but well worth the effort.
My preliminary investigation appears to reveal that the ODS system is exactly what I expected, only moreso: Stark. Since the numbers are small (Attack Levels and Armor Classes for example being only 1 through 6) it is the case that the effect of small differences in the numbers produces large variations in the results.
For example, two average fighters who are one level apart show a significant difference in percent chance of success against one another. If the calculations are correct, which is still in question at the moment, then it appears that the 1st Level Fighter can expect to defeat the 2nd Level fighter in 24 Melees, while the second Level Fighter can expect to defeat the 1st Level Fighter in only 6 Melees. Thus, the chance of victory for the 2nd Level Fighter is significantly greater than the 1st. Is this difference adequate? Does it work so far as the game is concerned, from a playability perspective? This remains to be seen. I am running a wide variety of scenarios, comparing differences in levels, weapons, armor, flanking, classes and range attacks including magic and missile weapons. It's quite a lot of work, but I plan to automate the process so that once I have the root formulas it should not be too difficult after that. But of course deriving the formulas *is* the difficult part.
In any case I am quite pleased with the work and am continuing to make progress. However, in light of my findings I must reconsider the August deadline for the final first draft of the ODS Rules Book. Given the difficulties I've had in validating the formulas so far, I'm not sure that August 1st can be achieved. However, nevertheless I will try.