A comparative fact can be presented in two ways. Among white evangelical Christians, Obama had 40% fewer votes than McCain. or Among white evangelical Christians, McCain had 40% more votes than Obama. Focusing on why Obama had fewer votes than McCain may result in a different explanation from focusing on why McCain had more votes than Obama, although it is the same fact. Thus what determines whether we focus in our explanation on Obama or McCain? In two studies, we show that people generally focused more on the first part of the comparative fact. However, when the comparative fact is presented in a negative frame (less than) there was a shift in focus from the first to the second part of the fact. For neutral items this moderating effect did not occur. The Principle of Lexical Marking (Clark, 1969) and Loss Aversion (Kahneman & Tversky, 1979) are discussed as possible accounts for this shift in focus.