Some surprising observations on mass combat

 I read A Practical Guide to Medieval Warfare by Richard Brooks and John Curry.

Really enjoyed it, found it enlightening and surprising in many respects. You need to buy it on amazon in paperback or via kindle, which I intensely dislike -- I'd prefer to pirate books, or buy through alternate retailers.

Sometimes the book gets lost in the weeds of (I think) irrelevant historical debates, important only to specialists. It also wastes time justifying the relevance of wargaming and reenactment to historical research. Putting those caveats aside, it was a worthwhile read, and I'd recommend it, supplemented by other readings in pre-modern warfare (I've also been reading War in Human Civilization by Azar Gat, the blog ACOUP , and the recommended books list on the Ask Historians reddit. Additional caveat: very little about war makes sense divorced from its historical context. You'll want to bone up on your regular history too!)

Anyway, here are some takeaways from the book (and a few other sources I've consulted):

Wearing armor doesn't affect your walking speed. In combat, you'll mostly be walking.
It *does* affect your running speed and stamina, slowing you by about half. https://www.youtube.com/watch?v=pAzI1UvlQqw
It also affects your speed over rough ground.
Training also affects speed. When you've drilled together, you can move faster while staying in formation. (The formation moves at the speed of its slowest member on the front. You want to have allies on either side, so if the guy next to you slows down, you'll slow down too. On the other hand, you're not likely to slow down very much if the people around you are familiar to you, because you don't want to look like a coward or betray your friends. (Keegan, The Face of Battle.))
Armor affects morale. Soldiers in heavy armor advanced faster, as they were more confident that they could withstand blows.

People want to bunch up for safety, but if they get too close, they can't use their weapons effectively. You really want like 2'6" between people.

Infantry doesn't charge. They advance, slowly, all together. This is actually really easy because nobody wants to get ahead or behind their peers.
(Charging is probably either a Presence check (-4) or a technique from a +2 martial art.)
Absent a charge, infantry wants to stop at the maximum range of their weapons, poking at each other.
When infantry does charge, they do it from a distance of 90 feet or less. 30 to 45 feet is more likely. Any greater distance and you're likely to spread out, have stragglers and leaders -- the last thing you want!

Morale is crucial.
You can rout the enemy as soon as they collapse, which happens quickly, and doesn't necessarily correspond to the overall state of the battle.
How does routing work? Two ways for an army to collapse:
1. The front line gets too many men taken out. Individuals don't want to face multiple opponents at once, so they back up. The people behind them back up to give them room. Soon the whole line is moving back, away from the enemy, losing formation.
2. As guys in the front get wounded, they fall back and get replaced with guys from the back lines. But if things look grim, the back line doesn't want to move forward. It'll stop advancing, or even start fleeing. This is actually how most routs happen in reenactments. The back line could see that the battle was going poorly, and evaporated, leaving the front line in the lurch, where all they could do is retreat.

Reenactors said the following was most important to their morale:
• Heavy armor
• Shields
• Being uphill of the enemy
• Being behind fortifications or obstacles
• Having confidence in the unit leader
The very worst thing is to know that the enemy has penetrated behind your lines.
We can probably add "belief in a righteous cause" or similar. Of 360 pre-battle speeches by real medieval commanders, 109 made reference to a just war. This is the single largest commonality in such speeches, followed by assistance from God (108), and simple orders (107).

Sieges and street fights. The vast majority of battles between armies were sieges.
Towns and castles had small garrisons, around 100 guys plus a civilian militia.
The average siege lasted about a month. Soldiers only owed 40 days of service a year, so it's important to end the siege quickly!
Most recorded sieges (75%) were successful. This makes sense, you don't want to bother with a siege if you can't win!

It really does take like 10 minutes to bash through a normal medieval door.
In a fight on limited ground (a street, inside a building, a dungeon corridor) the attackers want to outnumber the defenders at least 2:1. There's no benefit to a numerical advantage greater than 4:1.
If the defenders are outnumbered that badly, they usually want to spread out into a bunch of different buildings, forcing the attackers to spread out to take the buildings one by one or suffer constant harrying and attacks from behind. Meanwhile the defenders should post archers on top of the buildings one row back from the contested area. If the defenders aren't outnumbered as badly, they should try to hold alleys between buildings, while archers stand on top of those buildings.