I’d say there’s still value in looking at behavior closely. Doing a formal study like this can end up being useful down the line in a field that seems unrelated. But I don’t really think the “evolutionary explanation” is supported by the evidence, at least as far as the abstract and publicly available data goes.
They’d be more credible just going from a pure behaviorism analysis than trying to cram this data into an evopsych mold, because I don’t think this experiment gives nearly enough new information to justify such wide generalizations about specific group dynamics at a species or gender level.
I’d like to see it done again with at least a control group doing something non-competitive, perhaps a focus group activity or something.
I’d also like to see some priming studies done in this way. I think the experiment they designed could be really useful to see how much priming plays a role during gaming interactions like this. Just out of personal curiosity.