Head acceleration events (HAEs) are acceleration responses of the head following external short-duration collisions. The potential risk of brain injury from a single high-magnitude HAE or repeated occurrences makes them a significant concern in sport. Instrumented mouthguards (iMGs) can approximate HAEs. The distinction between sensor acceleration events, the iMG datum for approximating HAEs and HAEs themselves, which have been defined as the in vivo event, is made to highlight limitations of approximating HAEs using iMGs. This article explores the technical limitations of iMGs that constrain the approximation of HAEs and discusses important conceptual considerations for stakeholders interpreting iMG data. The approximation of HAEs by sensor acceleration events is constrained by false positives and false negatives. False positives occur when a sensor acceleration event is recorded despite no (in vivo) HAE occurring, while false negatives occur when a sensor acceleration event is not recorded after an (in vivo) HAE has occurred. Various mechanisms contribute to false positives and false negatives. Video verification and post-processing algorithms offer effective means for eradicating most false positives, but mitigation for false negatives is less comprehensive. Consequently, current iMG research is likely to underestimate HAE exposures, especially at lower magnitudes. Future research should aim to mitigate false negatives, while current iMG datasets should be interpreted with consideration for false negatives when inferring athlete HAE exposure.