In general, it's hard to have something like that for two reasons, one of them being like what Vudu said.
First, is that you would remove the incentive to recruit to fill your side. If you can win just as easily no matter how many players show, why put in the large amount of work required for successful recruiting? Also, design is yet more difficult when you can't plan for any particular numbers.
Second is that you could incentivize playing games with numbers in extreme cases. We do need to keep in mind extreme cases in scenarios, as there have been examples many years back with teams pushing extremes. What if, instead of having 40 players that average out to a typical skill level, you can get 15 awesome players and discourage participation by average and less-skilled players. We don't want any incentivization of things like that.
In this particular scenario, the design, when full, is 46 pilots on the allied side (38 fighters, 8 bombers) vs. 40 fighters on the axis side (40 fighters, 0 bombers). The design ratio of axis:allied pilots is thus 0.87 and not 1.0. It is not a symmetric design. The allies have bomber pilots that the axis doesn't, and scoring was designed to take that into account (which we discussed a lot on the design topic).
When I assign walkons, job 1 is achieving the design ratio. If I can do that and achieve side preference, I do that. But if more people want a side than works for balance, they don't get their choice. That's one of the disadvantages of not registering -- you are not guaranteed the side you want, and you are not guaranteed any particular plane type. Assigning walkons is complicated by people coming into and out of the arena (so numbers fluctuate some), by some walkons not following instructions, and by it being hectic to get everything assigned within about a 10-minute period. (You have to wait until a little after 12:00 for people to get into arena, then you lock, then you need to assign as fast as possible to give sides time to get people sorted and briefed -- combined with being pelted with "let me in" requests all the while.) Achieving the designed balance within a couple of people is at the upper level of what is possible as judged by about 100 past frames worth of assigning walkons.
In frame 1, the ratio was 0.86, so nearly spot on. In frame 2, when I was done assigning walkons, it was 37 to 32, a ratio also of 0.86. In frame 2, you can see that two people showed up late after I assigned walkons, making it 39 to 32 by the logs. So the axis was two pilots short of the ratio in frame 2.
In frame 3, if we get enough walkons, I will purposefully overassign to the axis side by a couple, to err on the side of achieving 0.87 for axis.
