Forecasting Principles: This site provides the academic support that shows why we use role-play in our process.
In particular this site contains the following papers by Kesten C Green:-
Role thinking. A paper showing that just thinking yourself into a role (putting yourself in other people's shoes) does not increase your ability to forecast much above chance. In other words, you need a role-play.
Forecasting decisions in conflict situations: a comparison of game theory, role-playing, and unaided judgement. This is the paper that shows how role-play is the best method of predicting the outcomes of confrontations, and that judgement is little better than chance. Don't take my word for it - look at this paper.
Game theory, simulated interaction, and unaided judgement for forecasting decisions in conflicts: further evidence. Further evidence and more data supporting the above conclusions. The difference is that with more data, role playing stays in the lead, but game theory comes off worse and seems little better than judgement, which is itself little better than chance.
Value of expertise for forecasting decisions in conflicts: Please don't just ask experts what they think the outcome will be: Even the best will get it wrong.
Structured analogies and forecasting decisions in conflicts: This shows that the best way to use experts is to ask them for analogious past situations. By using this method much more accurate results can be obtained.
Brief one page article by J Scott Armstrong summarising the benefits of role play: