Thanks, Pocus, I will reduce model VP values.
In this scenario, the Poles/Whites dominate for about a 2 month timeframe and can inflict substantial Red subunit losses. VPs for model subunits generally vary between 3-5 VPs. This type of situation produced about a 60NM difference between the sides in about a 7 turn timeframe. Basically the Reds have gone into such an early NM hole that I doubt they can recover even with arrival of substantial reinforcements. Yet the large subunit losses are more a reflection of initial weakness of the Red forces and the total numbers of units involved in combat rather than a truly significant degradation of the Red Force. And historically, this timeframe was more of a Pearl Harbor type situation which produced a united and determined Russia rather than a pessimistic Russia which is not reflected by the low NM produced in the scenario.
In this particular scenario, it appears modding model VPs is the best solution and really shouldn't have any negative side effects.
Just for future reference for scenario modders/designers, if you have one side dominant for a sustained timeframe and large VP values for models, NM can get out of control quickly if the stronger side is able to destroy enough subunits early. In addition, scenario objective values can also have an impact on NM. I have had to reduce both in this scenario.
Also maybe for the future modding wishlist, a universal factor for adjusting NM could be pretty useful vs adjusting model vp values/scenario objectives.