N.B. this post has been edited in light of a comment received.
Is evidential decision theory silly? Of course it is. It's a theory. Professors write about it. Q.E.D.
True, Evolution too is a theory. Professors write about it. It must be silly. Sadly, the World is a terribly silly place and an inebriate acquiescence in taking stupid chances is what redeems its silliness from the point of view of Wisdom as the existential project of ludic agency.
Prof. Huw Price has a well written paper- easily intelligible to the intelligent lay man (or unintelligent unlaid man desperately trying to sublimate his libido by cruising philosophy blogs in the early hours of the morning so as to avoid boozily booty-calling fat chicks coz we all know how that ends)- in which he argues that a probabilistic evidential decision theory is defensible provided it incorporates notions of agency and effective strategy such that 'spurious correlations' disappear from the agent's perspective and thus they 'don't have to be dealt with by a possibly vicious restriction on the general principle that only causes increase probabilities'.
The problem here is that all causal factors we can name- including purely mental ones- are spurious because there is always some lurking variable which is too fine grained for us to measure or specify or have awareness off- and, more fundamentally, the notion of agency and effective strategy both ultimately hinge upon a teleology of backward causation (thinking about decisions which will affect your future means thinking differently, thinking as though you are someone entirely different, with different preferences and goals such that a person who will only exist in the future dictates what you do now- in other words your 'genidentity' has usurped actual identity.)
Prof. Price mentions Dummett's 'analysis of the conditions under which, without inconsistency, we might claim to be able to bring about past events. Dummett shows that we can accommodate a belief in backward influence, so long as we are prepared to give up the assumption that before we decide how to act, it is possible for us to find out whether the past event in question has already occurred.'
The fact is, it is never possible, on a sufficiently fine-grained phenomenology or theory of the world, to determine that any occurrence is truly 'Past'- which also means Gibbardian 'hyperstates' and judgments made by 'hyperagents' have no road to supervenience with respect to 'prosaic factual properties'; everything is always in a sort of 'mixed inference' or else a Frege-Geachian flux till Beenakker's boundary resolves Hempel's dilemma as the Cosmic cows come home. Thus any Agency and Intentionality-based 'inwardness' we can have knowledge off must be reverse mereological and Time arrow reversed as indeed is what we would expect if our minds evolved on a stochastic fitness landscape.
Prof. Price examines a specific medical Newcombe problem- the chocoholic who suffers from migraines. Choco (as Price calls her) may believe that craving chocs is a symptom of Migraine onset and that whether or not she succumbs to temptation is irrelevant to the outcome. However (this is the Newcombe aspect of her dilemma) refusal to eat the chocolate may be proof that she isn't in Pre-Migraine Syndrome and thus has a bearing on whether she suffers the migraine.
The problem with Price's analysis is that it doesn't look at experienced changes in the level of chocolate craving- i.e. it throws away relevant information- and thus is itself irrelevant. In fact, an actual migraine sufferer- or a guy trying to quit smoking, or ordering phall curries, or booty calling fat chicks- has to learn to monitor and manage changes in craving levels rather than how to perform some Bayesian type of cost benefit analysis or complicated Game Theoretic calculation.
Since laymen only interest themselves in Philosophy to the degree that they are struggling with the condition of being unlaid, it follows that Huw Price can't save evidential decision theory by an appeal to Agency- not even an Escort Agency- because fat chicks do too deserve love.
Mind it kindly.