I don’t think you need to completely change the methodology. How about just providing a way to estimate Brewhouse Efficiency (BE) from a user’s entered parameters? You already have equipment losses, so if you provided a place for Mash Efficiency, you could display the estimated BE and let the user choose that or enter their own value like they do now. If you wanted to get really fancy, you could average BE’s over the last ‘x’ brew days and give them an option to use that as well.
The problem with just using an empirically derived BE is that it takes a couple of iterations to get a stable number and that balance is upset if equipment is upgraded or changed as is often the case with homebrewers.
In general, it would be nice to be able to formulate a recipe, then on brew day take the measurements that you normally take, enter them in the program, and have it show you how things went. You do this really well with gravity readings and ABV. Could you extend that to efficiency as well?