The US Air Force, Navy and Marine Corps use the direction orientation task (DOT) to assess spatial ability of potential pilots. However, recent research has shown limitations with the current implementation of the DOT, as well as an earlier candidate replacement the DOT 2. Specifically, there is evidence to suggest that the use of a math strategy and math ability may be related to performance on both tasks. The focus of the current paper is on the evaluation of a revised version of the DOT task (DOT 2.5) designed to encourage the use of spatial solutions by removing the numerical values from the task. We compared the performance of DOT 2.5 with previously validated measures of spatial ability, namely the paper folding and block rotation tasks. We also asked participants how often they employed different strategies on the task (e.g., a spatial strategy or mathematical one). Our results show the DOT 2.5 is significantly related to spatial ability as measured with block rotation. However, while participants were three times more likely to use a spatial strategy on the task, only the use of a math strategy was related to improved performance.