Sometimes, candidates with high academic achievements perform poorly on aptitude tests. Conversely, candidates with poor academics sometimes score highly on aptitude tests. This is because academic tests and psychometric tests measure different abilities. Academic tests measure learned knowledge, whereas aptitude tests measure innate cognitive ability. Performance on one does not necessarily tally with performance on the other; therefore you cannot rely on academic achievements to predict aptitude.
Research shows aptitude tests to be considerably more predictive of job performance than academic achievements (Schmidt & Hunter, 1998). Therefore, aptitude tests scores present a more valid prediction of future job performance, much more so than academic achievements.
So don’t be too surprised if a candidate with a 1st class degree in maths does poorly on a numerical reasoning test, or if a candidate that failed English scores highly on a verbal reasoning test. Just remember that aptitude test scores are more predictive than the academic achievements, and must be given precedence in any selection process.
Schmidt, F. & Hunter, J. (1998). The validity and utility of selection methods in personnel psychology: Practical and Theoretical Implications of 85 years of research findings. Psychological Bulletin, 124(2), 262-274.