This is the first post in a little series I'm slowly writing on how I see forecasting, particularly conditional forecasting; what it's good for; and whether we should expect people to agree if they just talk to each other enough.