Talk:Spring 2014 Event/@comment-60.50.242.46-20140501144645/@comment-861160-20140502024058

Saya's dataset is small and not controlled for a variety of factors, including engagement form. And he even says it himself that he cherrypicked 34 trials out of 74 to support his argument. It was also interpreted without knowledge of how the anti-air calculations worked back then, so the torpedo bombing numbers are very suspect. It is nowhere near conclusive.