In the background I always think of a 'performance factor' as an unseen stat. This can be made up of a number of things producing a value(s) that can be applied to (relevant) actions. It might include: 'motivation'; morale; team co-efficient; a player stat multiplier of weight/height/age; suitability for role; fit with team; player 'event' aptitude; etc etc.
so, we have an equation, trying to find A, where A=x + (3-y) + (2Xz)

I don't claim any or all of these are correct but when I look at the numbers we are given and try to think of game mechanics that could be extracted from them then its some of the tings I come up with to provide the 'reason' I would not otherwise have. eg why is X better than Y at FK when Y has SA and better shooting crossing? = better aptitude/performance/voodoo factor on X than Y >.<
an explanation maybe is that if those factors give a +1% or a +3% and the game/program has a roll dice of +- 20% you can't get a clue.
As for injuries for example, if moral/condition/grass quality/resting hour/M&P give a +1%, 2 or 3 bars give a -3%, but overall there is a roll dice of 10 or 20% (and hit max), all above becoming minor.