Board Thread:Questions and Answers/@comment-26571623-20151026085635/@comment-26486243-20151026122447

First source uses an entirely different hypothesis to the usual one. The second one kind of supports the existing numbers, but suggests that the cap is a soft cap rather than a hard one (though their st.dev is large due to small test sample size). Line 131 in source 3 agrees with source 2 about the soft nature of the cap, but suggests the effective softcap level might be 60 instead, while 272 claims the cap is both 60 AND a hardcap...

Still, opinions are worthless as proof, so only src2. is actually useful here. It seems gun cut-ins are not as bad as the previous 40 luck theory would indicate.