Here is a surprising graph of one hit wonder events. There is one hit in the ECal, but the trigger was pair1 and there is more than one ECal cluster.

It seems that is because I am not understanding the entries in the DST well enough. Note that I get the same result if I plot “ecal_hits.GetSize()”. I think I am looking at this incorrectly, because the file has 251823 events, but the produced histogram has 800602 entries. It seems thus that there is more than one “ecal_hits” entry per event, and indeed, for event_number == 1 I find 5 entries for n_ecal_hits. When I look at the code that creates these hits, this doesn’t quite make sense.