The most common material metrics are strength-to-weight (Ftu/rho) and stiffness-to-weight (E/rho). If you consider E/rho of aluminum, steel, and titanium, they all have about the same value, so there would appear to be little to choose. However, there are other metrics that are often more important, but less often discussed. For example, for column buckling, the metric is (E)^1/2 / rho and for plate buckling it is (E)^1/3 / rho. When you check either of these metrics, you find that aluminum scores higher than steel or titanium.
For an entire airframe, how much each failure mode governs the structural sizing must be considered. If your tensile strength is 10% better, but only 10% of the structure is sized by tensile strength, then your net weight savings due to improved tensile strength is only 1%. A large proportion of airframe weight consists of thin walled structures, which are often buckling critical, therefore the (E)^1/3 / rho metric may be the most representative.
The 1982 paper by Ekvall (see below) considers 7 failure modes, shows metrics for each, and divides the airframe weight into 7 zones governed by each failure mode to estimate weight savings with improved materials.
Finally, the above only considers mechanical properties, and of course there are many other factors to consider. Can you get the material you want tomorrow, or must you special order it and wait 6 months? Would it be worth a cost of 10 times more to save 10% weight? It may have better metrics, but have poor corrosion characteristics, would you use it?
The following references are worth consulting...
PAPER
"Methodology for evaluating weight savings from basic material properties" Ekvall, Rhodes, Wald, ASTM, 1982
BOOKS
Materials Selection in Mechanical Design, M. F. Ashby, 2005
Optimum Structural Design, Leonard Spunt, 1971
Weight-Strength Analysis of Aircraft Structures, F. R. Shanley, 1960
Minimum Weight Analysis of Compression Structures, G. Gerard, 1956