Abstract The application of sequence stratigraphic concepts, first proposed during the mid-1970s, has led to both successes and failures in petroleum exploration. Although a broad spectrum of reservoir occurrences are predicted in siliciclastic sequence stratigraphic models, the most successful applications have been in incised valley fills (lowstand and transgressive systems tracts) and turbidite systems (lowstand systems tracts). Notable exploration failures have also occurred in these settings. We analyze one example from each of these depositional settings to illustrate the successes and limitations of sequence stratigraphy in exploration. The first example is from the Pennsylvanian Morrow Formation of southeastern Colorado. This is an incised valley fill trend that produces from several different valley fill systems. Of the 181 wells drilled in the Clifford to Siaana fields trend, 41.4% of the wells penetrated the valley (75 wells), 19.3% of the wells found reservoir-quality sandstone (35 wells), and nine new fields were discovered (5% rate of success). Since 1985, when the play was first interpreted as an incised valley fill, drilling statistics indicate no significant increase in the success rate of discoveries, even though a greater percentage of wells penetrated valley fill facies. This case study indicates that the major problems in valley fill exploration were (1) locating the valley, (2) delineating reservoir sandstone within the valley, and (3) finding a trap for the reservoir. Exploration by Soekor (Ltd.) Pty in the Lower Cretaceous turbidite systems of the Bredasdorp Basin in offshore South Africa is ideal for demonstrating the frontier exploration applications of sequence