I have seen those studies pop up from time to time on HN with conclusions altering between "it's reachin worrying proportions", "there is an increase", "there is no increase", "they are so rare as to not worth considering".<p>From what I understand, the decision of what kinds of shootings are included has a large influence on the conclusion - e.g., the "there have been 18 shootings in 2018 so far" articles from a while ago used a comparatively low threshold for inclusion.<p>The threshold for this study seems to be "4 or more victims", which I think is similar to the threshold official publications used at the beginning of the Obama administration. I believe there were complaints that the threshold is unreasonably high which caused it to be adjusted - however, I don't have any sources for that ready, so if anyone knows more, please correct me.<p>In any case, it's important to look at the criteria if one wants to compare those studies.