Re:How Not to Sort by Average Rating (2009)
by quantdev on Wednesday August 30, 2017 @11:52AM
(#15133168)
For a fixed confidence level, it looks computationally light weight: a dozen or so multiplications and divisions plus one square root, which could be approximated if needed. There is no inverse normal needed at run time.
Often you also want to give a configurable advantage or handicap to new entries.