Why, sometimes I've believed as many as six impossible things before breakfast. 1 Speculations, conjectures, suppositions, opinionswe encounter them all in history and everyday life. Yet do speculations land us chiefly in the realm of Carr's 'parlour games' and 'mighthave-been's'? To what extent does our historical reasoning admit multiple possibilities of what might have occurred at a point in time? When offering explanations and interpretations of events, actions, and processes, we make such claims through the use of selective evidence, choosing to pursue one angle over another, one alternative over other ones that might not fit as harmoniously with what we consider plausible possibilities. Our willingness to accept one explanation over another or our decision to accept several possible explanations at the same time are linked to our 'ability to accept the rationality of the unobserved' (Okasha 2000, 693) and emphasise our bounded rationality. From inference to the best explanation may seem like child's play but it also grounds much of what we do as historians (Gelfert 2010;van der Dussen 2016).The question of the validity or goodness of an explanation relates to broader questions of historical judgement and the philosophy of historiography. The concept of a best explanation can be interpreted usefully in terms of Lipton's concept of loveliness versus likeliness of an explanation (Lipton [1991(Lipton [ ] 2004Bird 2010;van der Dussen 2016). When choosing between potential explanations, Lipton outlines two stages. First, we filter out explanations that are less plausible from all others that we have imagined or developed. Next, we examine the remaining potential explanations and rank them according to our evaluation of explanatory goodness.