It is but obvious that the past voting behavior would have an influence on the future voting tendencies of the voters. The question is, to what extent does it influence and is there a mechanism by which the same could be measured?
As you say, quite obviously, past voting behavior has a strong influence on future voter behavior. You ask about a measure by which this could be measured. My recommendation would be to turn around your question: not looking at voters who's past behavior strongly influences their future behavior, but those voters for whom it does not.
Take a look at the literature on electoral volatility (i.e., voters switching party preference). As you will see, there is a lot of literature that tries to explain why some voters switch while others remain loyal to their party. In that way, by reading on voters switching parties, I am sure you will learn a lot with regard to the influence of past voter behavior on future voting as well!