So I don't want to dig up the posts but acoup.blog has long sections about medieval warfare. Basically the change can be explained by this. The primary reason a

So I don't want to dig up the posts but acoup.blog has long sections about medieval warfare. Basically the change can be explained by this. The primary reason a cavalry charge is so effective is not because of the force itself (despite what Total War makes it appear) but the psychological impact of a much larger beast charging into you at full speed, at a very deep instinctive level your instincts will scream at you to run from this to turn out of it's path. So when you have hundreds of the beasts bearing down on you heavily armored and men screaming at them it is almost impossible to not turn or flinch or flee. In antiquity especially in the Roman era this advantage was negated to an extent by professional armies, you could drill and train the soldiers over and over again to not break and not flee, thus muting the impact of the cavalry charge. However as society fragmented and the power of the state waned the infantry was more likely to be composed of conscripts and levies that were not professional soldiers and wouldn't be able to be psychologically withstand the impact of a cavalry charge. This in large part explains the "However, there is no historical consensus on when the stirrup became important in Europe. I’ve seen serious claims ranging from the late 300s to the late 700s." because it was a social rather than physical change that precipitated the rise of the heavy cavalry shock attacks. The proof of this is that as we see the rise of nation states and a resumption of the power of centralized administration we see cavalry again wane from an important shock force to a flanking and harrying force during the Renaissance. As evidence we see that even in to WW1 many of the generals in the early part of the war were hoping for a breakthrough in the line by the infantry which the cavalry could then exploit by turning on the enemies rear. Addenda: I just finished reading the article and the author is basically saying the same thing as I am now. Strong central state -> Professional military -> Cavalry charges less effective. Weak central state -> Warrior nobility class and conscript infantry -> Cavalry charges powerfully effective. I'm keeping my comment though because of the sunk cost fallacy.