Reddit is a social platform that has become the meeting point for millions of people every day. Reddit allows its users to debate about any opinion or news, by posting comments and answers to previous comments of other users. In a previous work, we proposed a system to analyze the agreement between user comments in Reddit debates, that represents a debate as a two-sided debate graph. In this graph, comments are divided in two groups, the ones that agree with the root comment of the debate, and the ones that disagree with it. The edges of the graph represent disagreement between comments of the two groups. In this work, we propose an extension of the analysis performed by our system, by defining some graded measures that allow us to estimate different characteristics of the interaction between users in the debate. Our aim is to use such measures to identify different user profiles in the set of users in a debate, to finally help users to filter user profiles that they are interested to interact with. In this work, we focus on the particular profile of trolls, although we believe that the measures we define can be used to characterize other user profiles. To test our system, we analyze the top voted debate from the subreddit World News. Our results show that our measure values are indicators of a behavior pattern that can be linked with the generation of more conflict, that we consider as the main intention of troll users.