Bonder-Farrell, darn it.
And the big hole in both, is that you cannot predict collapse.
Yes, of course...but Bonder-Farrell is a derivative of Lanchester's work, and it also requires modification (the assumptions aren't necessarily met in even an aerial meeting engagement, and heck, even Bonder and Farrell modified their own :-).
I was going back and giving original credit; even if Lanchester's differential equations might be superceded by current simulation-modeling techniques, his laws draw attention to some concerns that exist regardless of how they are quantified.
And the big hole in both, is that you cannot predict collapse.
That's one big hole. :-)
But regardless of the details, and regardless of how we bring in behavioral modeling or parameter estimation or the like, my point is that the purchase of the "best" platform is not always (and often isn't) the most cost-effective decision for a nation. And it is a decision that can't be made in isolation of considerations beyond direct combat effectiveness platform-to-platform (such as mission, crew availability, C4I burden/contribution, maintenance, TAT, collapse susceptibility, international relations, casualty sensitivity [strategic, political], etc.)