Archaeological studies of early weaponry have relied for decades on ethnographic parallels—whether from ethnohistorical accounts, ethnographic literature, or from objects studied in museum collections. While such accounts and collected objects provided key data in the past, including of morphometrics and functionality, few studies have explored the quality of such data. In this paper, we critically assess a dominant theoretical paradigm, namely the utility of ethnographic collections to assess Pleistocene archaeological material. Our focus is how ethnographic spear morphometrics are used to propose delivery methods of archaeological weapons. We discuss the archaeological significance of early spears, and the role that ethnography has played in interpreting them. We provide new morphometric data of ethnographic wooden spears, which have been used analogically to assess the earliest archaeological hunting tools. We systematically collected data from ethnographic collections of wooden spears in five museums in the UK and Australia including mass, length, diameters and point of balance, alongside any recorded information on provenance and use. Older datasets, as well as the data in this paper, are limited due to collection bias and a lack of detailed museum records. By subjecting the new data to statistical analyses, we find that with a few exceptions morphometrics are not reliable predictors of delivery as thrusting or hand-thrown spears (javelins). Prevalent hypotheses linking variables such as mass, tip design, or maximum diameter with delivery are unsupported by our results. However, the descriptive statistics provided may remain useful as a means of comparative data for archaeological material. We conclude that using simple morphometrics to parse weapon delivery has had a drag effect on forming new and interesting hypotheses about early weapons.