Many individuals have flipped cash however few have stopped to ponder the statistical and bodily intricacies of the method. In a preregistered research we collected 350,757 coin flips to check the counterintuitive prediction from a physics mannequin of human coin tossing developed by Diaconis, Holmes, and Montgomery (D-H-M; 2007). The mannequin asserts that when individuals flip an peculiar coin, it tends to land on the identical aspect it began — D-H-M estimated the likelihood of a same-side final result to be about 51%. Our knowledge lend sturdy assist to this exact prediction: the cash landed on the identical aspect most of the time, Pr(similar aspect)=0.508, 95% credible interval (CI) [0.506, 0.509], BFsame-side bias=2364. Moreover, the information revealed appreciable between-people variation within the diploma of this same-side bias. Our knowledge additionally confirmed the generic prediction that when individuals flip an peculiar coin — with the preliminary side-up randomly decided — it’s equally more likely to land heads or tails: Pr(heads)=0.500, 95% CI [0.498, 0.502], BFheads-tails bias=0.183. Moreover, this lack of heads-tails bias doesn’t seem to range throughout cash. Our knowledge due to this fact present sturdy proof that when some (however not all) individuals flip a good coin, it tends to land on the identical aspect it began. Our knowledge present compelling statistical assist for D-H-M physics mannequin of coin tossing.
By František Bartoš, et.al., that may be a paper from late final 12 months.
