1. "Sample bias" - yes, I agree that the forum members are not representative but surely this is the most knowledgeable sample group from the stratum. I think this group know all model ranges including past models to make a comparison between materials.
2. "The product range - mainly Ti vs mainly CF". IMO such kind of generalization creates even greater bias - in time and cross-sectional terms and limited number of observations for CF. F.i. - how could you compare regular ti off-shore with CF models issued only as LE???, Also, Ti is relatively "old" material compared to CF which is used for recent models and thus more "fancy". Moreover, we have just few models to compare and thus, there's no need for generalization.
And now, ladies and gentlemen, in a blue corner is CF team: Team Alighni CF, BumbleBee, F-1. In a red corner is Ti team: Montoya, RBII, Survivor. What is the winner team??? Which is harder to find??? I really don't know. Looks like they're equally accessible or, more correct, hardly accessible.
So, my point is to compare the exact model between "as is" and "possible" material combination, in our case Grand Prix. I perceive this is just dreaming and may be somewhat destructive to the Grand Prix already released, but I'm digging for the truth...