It's the difference between "number of specimens surviving" on the one hand, and "desirability to collectors" on the other.
Rarity ratings in the new RIC are based on "number of specimens survivng and known to the authors". Your Titus sestertius is indeed very rare, as confirmed by the die identity of your specimen with the Paris coin illustrated in RIC, and by the absence of that type in the BM collection up to 1976 (p. 140 in BMC).
Now the rarity ratings in the old RIC should theoretically have been based on the same criterion, but in fact they were heavily influenced by Cohen's franc prices. Market prices of course do not reflect absolute rarity: they are much more dependent on the historical or artistic interest of the type, determining how eager collectors will be to add it to their collections.
Titus' standing Roma type on COS II sestertii is very rare, but not interesting: so Cohen 181 correctly gave it the low price of 4 francs, and the old RIC unfortunately followed suit with rating "common"!
Many varieties of IVDAEA CAPTA
types on sestertii of Vespasian and Titus are quite common. They are of great interest to collectors, however, so Cohen priced them at 8, 12, or 20 francs. The old RIC again unfortunately followed suit, rating these types "scarce" or "rare"!