2020 EDIT: I've since revised my CCAF work completely in this regard, so don't use this post for research purposes (the 3057 total estimate is wrong and the growth rates are all wrong, as I confused absorbed regiments from the Xin Sheng era with new-build regiments). I'm leaving the post intact only to make everything else that follows understandable.First, the Av ratings, as requested.
https://www.dropbox.com/s/ecse3cz1oha1rpw/CapCon%203057%20AV%20ratings.docx?dl=0Now, on to a lot of the fun stuff in order to answer questions. This is also why these tables take so long to make.
The CCAF gets chopped down to an all-time low of 12 regiments following the 4th War and Adurien-Magistracy Invasion. By 3054 we know that they're back to 19 regiments (if you include the Big Mac). In order to justify their known 3063 total of 44 regiments, I estimate that they're up to 31 regiments by 3057. That means that less than 40% of their ranks are pre-4th War production. At the same time, they get dropped down to an all-time low in terms of factories as well, and those don't reappear anywhere near as fast as their regiment numbers do. They lose Corey, they lose Styk, they lose Nanking, they lose Tikonov, they lose St. Ives. Basically all they can build are Stingers, Wasps, Locusts, Vindicators, Ravens, and Cataphracts. Offhand, that's it (and remember, some of those pre-4th War survivals are going to be yet more Stingers, Wasps, Locusts, and Vindicators). That means a remarkably homogenous set of tables. Yes, when you fight the CapCon in this era, you may lose simply by being drowned in Vindicator corpses. Kind of boring, but that seems to be the established facts of the universe, and the whole point of these tables is to capture things like that. At least the AB tables get some Marik bribery machines to leaven things a bit.
The other thing is that when I completely redid the tables some years back to take into account regiment numbers and known weight distributions, and tied that into my Av numbers, it means I could say with some statistical certainty just how many of X machine actually existed in any one House. Now, because I have a record for each era I make tables for, I can be consistent in terms of these numbers. More importantly, I should, or the work was pointless. For instance, I start a set of 3057 tables by grabbing my excel sheet for 3050 for that faction and then begin making tweaks. So, in 3050 Liao statistically has 51 PXH-1 Phoenix Hawks in its ranks. Since they don't make those anymore, I have to make sure that the combined total of Phoenix Hawks in both the AB and CDF sheets for 3057 doesn't exceed 51. If I'm off by 1 or 2, no big deal, but that's the general constraint I'm working with.
So, take 3050, figure out how many regiments are in 3057, figure out how many of those are A or B-rated vs CDF rated, figure out new production and open-market purchases new to that era, and cross check the results constantly against each other and the 3050 total to make sure it all lines up both backward in time and across both sets of unit ratings. This took a while to set up.
Because of this combo -- the split between AB and CDF, combined with wild swings in regiment sizes for some factions -- Av numbers are often much rougher guidelines compared to what they used to be. All that matters to me is lining up the total number of old, fixed numbers of machines (the ones no longer in production) between the 3050 tables and the 3057 tables. The rest is gravy (we don't have production figures for newer machines, but we have the number of regiments added, so by process of elimination what is produced must fill those regiments).
Bear in mind, too, that what the Av ratings mean change depending on whether you're looking at AB or CDF tables. The AB tables are always dealing with a much smaller number of regiments (9, in the CCAF's case). For example, Av 2 in the AB tables means about a lance of machines scattered across 9 regiments; the same value for the CDF tables (22 regiments) means about two companies. Because the AB-rated regiments are fewer, I could add more weird Succession War-era obscurities and SLDF leftovers to their list, whereas in the larger pool of CDF regiments they're diluted to the point of statistical extinction.
So, on to specific points:
Number of Ravens: kind of addressed above (Liao doesn't build much else). I tried to keep the numbers of ECM ones down, though, focusing on the combat models. You've got about 30 ECM-equipped ones scattered throughout the 22 regiments of CDF (most being the primitive -1X).
Chance for good machines in CDF ranks: thought about it, but if the well-connected premier regiments don't have enough for their needs/desires I don't see how a F-rated unit is going to get enough to statistically matter, so I removed it. At the same time, I could allow the chance, but with a roll modifier based on unit quality, so that C has a chance and D and F do not. Might be worth it, probably be too fiddly to bother with. That all having been said, if a Field Manual RAT assigns something cool to a lower-tier set of units, and it's not too wacky, I'll probably run with it.
Salvage ratios: Yes, Liao fought the Davions more in recent history. But they didn't win a single one of those battles, and you have to control the field to salvage. I forgot to note that the salvage ratings should specify the 3025 tables only for that reason (I'll edit that in the PDF right now): it's all 3rd Succession War-era salvage.
Also, the green coloured band is part a suggestion made a while back by some clever reader asking for quick visual cues to help guide a guy flipping through these. Each of the House tables will have their colour.