# What should be done with cheats ? Part 5 :-(

Started Mar 15, 2013 | Discussions thread
 Like? 3
 Re: What should be done with cheats ? Part 5 :-( In reply to Deleted pending purge, May 28, 2013

Entries in a recent challenge have raised some eyebrows and the joint winner in this challenge raised mine too. Here are their recent challenge entries:

As a fairly random sample, I looked at the distribution of votes for their last 10 entries:

The first one looks normal enough:

But then the following nine all seem to have a bimodal distribution with an outlying peak at the high scores:

Now, as Conan points out, these kinds of distribution can happen by chance. Let's say that the probability of getting a distribution like that for any particular entry is 1 in 10 (i.e. 0.1). This is actually probably an overestimate, but that means that the calculation is conservative. If the probability for an individual entry is 0.1, then the probability of getting at least 9 out of 10 entries with that kind of distribution is:

(0.1 ^ 10) + (10 x 0.9 x (0.1 ^ 9)) = 91 x (0.1 ^ 10)

= 0.0000000091 or 0.00000091%

This is a very small probability. When the probability of something happening by chance is as small as this, it is generally accepted that the event didn't happen by chance. I'm fairly sure that a probability this small would withstand the 'beyond reasonable doubt' test in a court of law.

Complain
Post ()
Keyboard shortcuts: