Head and torso appear to have been the preferred targets. Head wounds are quite common.

A summary of wound distribution, from skeletal evidence, from Knüsel, Christopher , "Courteous Knights and Cruel Avengers" , in The Routledge Handbook of the Bioarchaeology of Human Conflict ed. Christopher Knüsel and Martin J. Smith (Abingdon: Routledge, 25 Nov 2013 )

Place	date	% head	% post-cranial
Fishergate (Battle of Stamford?)	11th century	18	82
Fishergate	13th/14th century	41	59
Wisby	1361	45	55
Towton	1461	72	28
Uppsala	1520	89	11

Since the head is approximately 10% of the frontal target by cross-sectional area, and about 15% by perimeter, the head is a preferred, and often the major, target.

Medieval fightbooks also favour torso and head. Knüsel summarises targeting in Talhoffer's fightbook as

target	%
torso	36
head	27
neck	18
weapon arm	6
other arm	2
legs	11

Written sources also describe the targeting of torso and head. Joinville's The life of St Louis describes attacks be lances to the torso (some stopped by armour, and some not), lance wounds to the face, and sword wounds to the face.

Wound distributions are also affected by armour. The torso is often the best protected part of the body, being protected by body armour and/or a shield. The head is often protected by a helmet, and can be protected by a shield. Legs are less likely to be armoured than torso or head, and are are less protected by shields.

Even without armour, not all targets are equally accessible. Arms, and to a lesser extent, legs, can be targets of opportunity, since they can be reached more readily, while the torso is a deep target, and harder to reach with weapons such as swords.

Armour and opportunity both contribute to the frequency of leg wounds. However, even with this, the number of head wounds was often greater than that of torso and limbs combined, and the head appears to have been much more the primary target than the legs.

The short answer is that the primary target was an area that was not defended. There is also a difference between a winning blow, and a killing blow.

Generally there are three parameters to think about. They are, 1: reach, 2: speed, and 3: ability to hit. One might add that preferably, you want an attack that will not put yourself out of position.

When it comes to hitting, a spear is a thrusting weapon, and so, it is reasonable to aim for something big, like torso and chest. You also want to keep a distance, so that you can use the spear again. Spears are probably often a weapon for mass combat, and much of the effect is in pinning the opponent, so that they do not have many choices in how to behave.

In modern times, daggers and knives are used in the same way, even though it is one on one fights. An experienced user will search for the centre line, and go for the fastest possible thrusts towards the centre of the opponent.

With a slashing weapon it is much easier to hit lower body parts, and with a sword, some of those attacks are also very fast, and what's more, one hit is probably enough to win the fight, since the opponent can't move.

Then, how did this turn out in reality? Well, we are talking here about many hundreds of years of history, with many battles. It is quite possible that armies did not target legs because, 1: they had little training, or 2: the opponents had leg armour, or training to protect the legs.

Fencing manuals talk about maneuvres to hit legs, and such slashes are shown in pictures, and it is well known that medieval soldiers wore leg protection.

But when the soldiers of an army had a chance to hit legs, they certainly did. We have evidence from the battle of Visby, where many of the fallen have slashing wounds to the legs. Osteologists estimate that some 70 % of the wounds are leg slashes. Many of these poor souls also have "killing blows", but it is believed that the leg slash was the deciding factor.