Ten years ago I read Stinchcombe’s “Theoretical Methods in Social History”. I recently reread the parts that I had highlighted and I thought it useful to share some key passages.
One does not apply theory to history; rather ones uses history to develop theory. 
It is rather that the fashion in quantitative history has come to be that one must agree to be voluntarily ignorant of the any evidence other than numbers. 
As the argument develops, it will become clear why I am unenthusiastic about most quantitative history. Let me state the argument in capsule form.
For a number, say a count, to be theoretically interesting, it has to be a count of a comparable instance. What instances comparable for a scientist is that those instances have identical causal impact. Thus a count is more illuminating, the more theory and the more detailed examination of the facts went into making the instances counted comparable. But this ordinarily means that making a count should be the last stage of a scientific enterprise, a stage reached only after an extensive development of theory on what makes instances comparable. Is the proletarian in the Vyborg district of Petersburg or in the Baltic Sea Fleet equivalent in impact on the Russian Revolution to a proletarian in Moscow? Trotsky convinces me he was not (and if the proletarian was a she, in either place, she was not equivalent to a male proletarian either). Consequently, a count of proletarians in Russia in 1917 is fact of relatively little interest. 
A count should be the last stage of a theoretically oriented social research, after one has got to the point where one is willing to sacrifice theoretical advance. 
But if general concepts consist in the analogies between elements and if they are deeper if the analogies are deeper, then the basic investigatory task of concept formation is the deepening of analogies.
Far from it being the case that the most powerful general theorists ignore details, the precise opposite is true. Social theory without attention to detail is wind; the classes it invents are vacuous, and nothing interesting follows from the fact that A and B belong to the class;
“theoretical” research appears as a species of wordy scholasticism, arranging conceptual angels in sixteen fold ranks on the head of a purely conceptual pin.
But if conceptual profundity depends on the deep building of analogies from one case to another, we are likely to find good theory in exactly the opposite place from where we have been taught to expect it. For it is likely to be those scholars who attempt to give a causal interpretation of a particular case who will be led to penetrate deeper analogies between cases. 
Thus it is vain to seek sensible causal sentences about “nobilities” until we find out whether the nobilities were something like the same sort of thing in different countries. The only way to do that is to build up from and analogies and differences among acts. And what makes acts causally analogous is similarity in what people want and what they think they need to do to get it. 
Since a class is logically the same thing as a set of equivalences among members of a class, a causal generalization in historical materials is a set of analogies between particular causal assertions. A causal generalization is, “When royal governments have outlived their usefulness, weak kings or stars reign without governing and fall easily.” This is equivalent to asserting an analogy between the statement, “Tsar Alexander’s unwillingness or inability to decide what to do about the war caused events that led to revolution,” and “Louis XVI’s lack of clarity about the purposes and role of the Estates General caused events that led to revolution.” When the generalization appears in the history of the Russian Revolution, it appears as an “interpretation,” and has not particular claim to generality.
By the simple act of asserting that two instances are alike, however, a class, a concept, is created, a generalization about it is offered, some evidence is brought forth, and we are embarked on a scientific enterprise.
A detailed analysis of what the tsar wanted from his cabinet and the Duma, what Louis XVI wanted from the Estates General, and in what respects these were similar, set us on the path of a giving a realistic and useful definition of the waffle-word “weak” in the generalization.
The moral of this book is that great theorists descend to the level of such detailed analogies in the course of their work. Further, they become greater theorists down there among the details, for it is the details that theories in history have to grasp if they are to be any good. 
: Stinchcombe, A. L. (1978). Theoretical Methods in Social History. New York, Academic Press.
: p. 1
: p. 5
: p. 5-6
: p. 7
: p. 20-1
: p. 120
: p. 123-124