Recent literature has shown that achievement emotions, their regulation, and perceived competence play a compelling role in mathematics learning and achievement. Studies that have looked at these variables have, for the most part, adopted a person-centered approach, which examines relationships between variables found to a similar degree in all individuals of the group. Yet, scholars have outlined emotional inter-individual differences, in particular, in terms of gender and past performance. The present study examined differences among upper elementary students in how achievement emotions are related to each other. Cluster analysis revealed four distinct profiles based on a sample of upper elementary students (N = 354): those with high levels of positive emotions and low levels of negative emotions ("positive"); those with high levels of boredom and low levels of the other emotions ("bored"); those with high levels of nervousness, worry, and fear and low levels of positive emotions ("anxious"); and those with high levels of the six negative distinct emotions assessed and low levels of positive emotions ("resigned"). Analyses of variance showed that the first profile stood out advantageously from the last two regarding math performance and perceived competence. Findings regarding emotion regulation confirm the risky nature of the resigned profile. The bored profile ascribes no value, whether extrinsic or intrinsic, to problem-solving tasks. Practical implications for educational practices and possible avenues for further research are discussed.