If the goal is to alleviate the all-or-nothing behavior of the current missile-PD interaction and encourage 'regular' launchers and larger missiles, I think the launcher size idea is in a good direction. It makes creating larger salvos with larger missiles easier.
As I mentioned in my earlier post, larger salvos don't alleviate the all-or-nothing behavior. In fact, they create the problem. It's the law of large numbers working against us. Essentially, each missile-PD interaction is modeled by a dice roll with a probability p of a successful intercept, after all bonuses and penalties are applied. Adding/changing mechanisms such as ECM and tracking bonus only changes p, but not the dice roll itself.
As an example, assume on the PD side, there are n=2000 guns controlled by a BFC, each with a p=10% chance of shooting down a missile after all bonuses/penalties. Then, the number of successful interceptions is a binomial distribution B(n,p) = B(2000, 0.1). As N is large enough (there is a check for this), we can approximate this by a normal distribution N(np, np(1-p)) = N(200, 180), with mean 200 and variance 180. The mean part is easy, it means these many guns are expected to shoot down 200 missiles. The variance part can be used to determine the 3-sigma interval, where sigma^2 = 180, giving sigma~=13.4. Thus, there is a 99.8% chance that between ~160 to ~240 (200 - 3 * 13.4 to 200 + 3 * 13.4) shots are hit. That means, when the number of incoming missiles is outside 160 and 240, the result is almost determined: below 160 all will be shot down, and above 240 there must be leakers. When the number is between 160 and 240, the closer to 200 the better, there is a chance some will get through whenever you try. Given how flexible ship designs are, there is a much larger chance that a missile volley containing less than 160 or above 240 missiles, than just falls in between that interval.
There are a few ideas already came up before that, after some thought, can actually alleviate the all-or-nothing behavior, we just need to put them together so they break the law of large numbers.
The first is actually in VB6 Aurora and earlier versions of the C#, i.e., each BFC can only target one missile salvo (i.e., a group of several missiles launched by a single MFC). Of course, this does not stop a BFC from controlling 2000 guns, but it definitely discourages that since a second simultaneous salvo will completely bypass the PD guns.
On top of the first limit, I'd like to add a second constraint that each missile salvo can be only engaged by one BFC/MFC at a time.
The third is limiting how many missiles can be directed by a single MFC on the attacking side, i.e., the missile salvo size. This can be a separate tech line, going from maybe 4 up to 12, not too high so we don't trigger the law of large numbers. MFCs can be made cheaper to counterbalance this nerf, and the number of missiles guided can be chosen as a parameter when designing an MFC, where lower numbers make the MFC even cheaper. This limitation will make mass box launcher ships harder to make as more MFCs are required, but for fighters and bombers, nothing has changed.
With these changes, we are preventing the number of shots from going too high during one interception. This will significantly increase the effectiveness of the attacking missiles when the missile volley size is smaller than the defender's PD capability. Again, as ship design is flexible, there is nothing stopping people from designing massive ships with multiple BFCs each guiding 200 guns with a 10% chance to hit, trying to stop all missiles. But it is more reasonable, in the above example, to have 20 BFCs each controlling 100 guns. Of course, the attacker can always bring overwhelming firepower to saturate the PD, but let's check the cases when a smaller number of attacking missiles are used. Against 20 salvos of 4 missiles, totaling 80 missiles in a volley, each salvo actually has a 2.4% chance of some missiles getting through, whereas, in the above example, there is a 2.8E-24 chance of any missiles getting through. If those 80 missiles are grouped into 10 salvos of 8 missiles, based on the 2nd constraint above, only 10 BFC can engage. In this case, each missile salvo has a 20% chance of getting something through.
Combining this with the initial idea of missile launcher size changes (which also discourages box launchers, just we don't need to nerf the box launcher as hard), it should achieve the goal of alleviating the all-or-nothing behavior.