Hey! I discovered your mod and tried it.
I tested pure Baltic warriors (plus enough shock to reach 1% archers) against a mix baltic-skirmishes (aimed at balancing skirmish phase damage).
My method:
I took two counts in the ERE, close from each other but not adjacent and gave them troups.
The defending count received a levy-like army of 12110 men (2*12k+2*2K). I moved his army in a county adjacent to the attacking count's county (terrain = plains). This forbids the defending count to fight with his personal levies since they are 3 counties away from the battle.
The attacking count received one or the two test armies for each battle: the pure baltic retinue or the mixed retinue (both 30K retinue cap). I granted a claim to the attacker. The armies fought in a plains terrain. There were 12 instances of the same battle for each test army (I saved just before the battle). all the troops were
in the same flanck. The same commander was used for each instance of the battle.
Attacking commander: 17 martial, genius and cruel
Defending commander: 17 martial, genius
Test armies (attacker):
pure Baltic: 32 baltic warriors (9600 B) + 3 shock retinues (600 HI + 150 A) = 10350 men total
mixed: 24 baltic warriors (7200 B) + 17 skirmishes retinues (850 HI + 3400 A) = 11450 men total
note on the mixed army: It has more than 20% archer but less than 30%. Therefore, it has some chances to trigger barrage tactics in melee phase which is not good for this composition (where I would prefer a heavy infantry tactics). I'll try a 19% archer mix later.
I did not find how to add the finno-ugric cultural building; this can bias the results in favour of the mixed army.
Results:
pure Baltic:
4 defeat out of 12 battles (+ one more defeat where I forgot to get the results).
Victories: losses = 12.6%; killed = 35.4%
Defeats: losses = 27.4%; killed = 22.1%
Mixed:
0 defeats out of 12 battles
Victories: losses = 16.2%; killed = 31.3 %
Observations:
- I did not calculate standard errors or confidence intervals but for the Pure Baltic, the variation in losses and kills across battles seemed very high. It was smoother for the Mixed. Given this variability, I don't think the differences between the two armies are significant. More battles still need to be done to get a good statistical distribution (I would say... around 30 battles given the defeats rate... but it is boring).
- Pure Baltic seemed completely all-in; it depended on whether the shieldwall tactic triggered and if the defending army triggered a strong tactic in the skirmish phase (which sometimes could even devastate the Pure Batlic under shieldwall). Even when winning, Pure Baltic finished with a very low moral most of the time (and in 2-3 occasions completely destroying the opponent. An example of the "all-in" characteristics of this composition was a defeat where the Pure Baltic lost 2229 men while killing 6169 men (this was the only battle like that)...
- On the other hand Mixed army obtained more constant results, finishing the battles with 1/3 to more than 1/2 their moral (except maybe on 2-3 occasions where it was lower). I noticed a few bad tactics (sometimes due to the commander's cruel trait) but never leading to a catastrophic turn of events (especially since it was negative for the archers, during the melee phase).
Dragatus, compared to your tests, maybe I used a more "borderline" situation due to the lack of cultural building. In this situation, reaching for a few damages in the skirmish phase was more powerful than only trying to hold and maximize the melee.phase. I should try it again with the cultural building.
Anyways, thanks for your mod... it is very useful ^^
