DoW Campaign Questions List: Rylian Questions Vol.II

Let me just start by saying I have greatly enjoyed the expanded axioms. There are so many useful, and amazing tables, and it is really helpful to me as far as making the world random yet consistent. However, after finishing the NPC generator, and taking a look at my political map, and refreshing myself on DoW Campaigns, I have several questions I would greatly appreciate some insight into.
 

Assuming I am making several kingdoms' worth of lords, what would their ability scores be? Above average? Top 20%? Normal? 

What about a feudal society, versus a "meritocracy" such as early mongolia or I guess the roman republic? I understand that noncombatants would be the subtype for those that do not fight, but I feel as though a land ruled by a monastic order of knights, like possibly the Teutonic Order, would have lords with increased stats at least in fields that they valued. How do you guys handle that?

Additionally, with that, how many officers tend to take proficiencies relating to war? Would the Teutonic Order pick military strategy more often, than, say, lords of the Papal States? How might that be handled?

On top of that, I saw that 25% of armies tended be veterans. Would it be reasonable to increase that in more wartorn areas in the same way that leveled characters would be found on the frontiers? How much of an increase might be too much?

Something that has been bothering me in DoW Battles and the unit conversion is for fast moving units, and whether or not they count as foot or as mounted. 
Dire wolves on goblins are cavalry. However, are hellhounds cavalry? Werebears? These are all 60 creature units, though. What about a unit of wolves or wardogs, that are faster than 120" but are much larger units than normal cavalry? How should I handle them, and should these units count as so far as scouting or screening? SoS please send help lol.

And lastly, I haven't been able to find anything on the forums or in the books pertaining to river crossings for armies. Would that have to be considered a construction project? Fords tend to be unpredictable, and I understand that Caesar would build his own bridges. But what if one wanted to build boats to help them cross? How would that be handled? Would you be able to cross as you built them?

Anyway, that's my flurry. Hopefully the spam wasn't too horrid.

To a few of your questions I can snag off the top of my head:

 

Occurence of veterans in wartorn regions: That's an interesting question. If I were to try and model that, I think what I'd end up doing is increasing the percentage by, say, 1d3*5% (so 5% to 15%, maybe), to ape the demand modifer rules - basically saying "this market has a demand for war". 
I'd assume at some point the veterans would disperse, and I might look at the timings from Domains At War on mercenary recruiting availability, reducing by 5% ever week for small settlements, every month for large, etc...

Stats: If there's a clear theme, like the papal states desiring clerics to rule, let's say, or the knights wanting fighters, maybe 4d6/droplow or something for the classes primary stat?

As far as "how many proficiencies would they take" - what might be interesting is rolling a 7th stat; if you get a +1 mod that's one level of MilStrat, +2 is 2, etc.

On cavalry vs fast-moving units - as far as I know, cavalry is only meant to be two creatures "together", one as a mount. Dire wolves on their own, or dogs, would be just regular troops.

Crossing a river: Absolutely a construction project, with all the soldiers available to work. What I'd do is figure out what's needed in labor and time to make one boat. I'd multiply that by the labor available to find out how many boats I can make at once.

Then, once the first rash of boats are done, you essentially have three groups of soldiers - Bank 1, Bank 2, and On Boat. Each bank side could continue to work on making boats while any working boat actually moves soldiers.

There's probably a point in this where making some sort of floating-bridge-flotilla makes more sense, but I'd have to spreadsheet it up, and it'd be wildly dependent on the properties of the river.