[35] Regarding this allegation, it has not been shown that, by postponing elections in the named electoral units, IEBC acted in bad faith or was inuenced by irrelevant factors and considerations. From the explanation tendered, we are satised that the postponement was occasioned by a genuine mistake, which in our view, could have been avoided had the members and sta of the IEBC been more diligent when they went to inspect the templates in Athens, Greece where the printing of ballot papers was undertaken. [36] In the absence of any empirical data, we cannot nd a basis upon which to conclude, as a matter of fact or evidence that the postponement aected voter turnout as a consequence of which the 1st petitioner, alone, as a Presidential candidate suered a disadvantage. At any rate, the nature of the ballot being an individual decision and secret, there may be other variables to which the turnout in the named units can be attributed. From the evidence on record, however, it appears to us that this year’s General Election recorded one of the lowest turnout since the reintroduction of multi- party political system, some 30 years ago. If there was a low voter turnout, it aected all the six categories of candidates and its explanation, in our view lies elsewhere but certainly not a calculated suppression. [37] On the other hand, in rebuttal to these claims, the IEBC illustrated, with examples to our satisfaction that there was no nexus between the postponement of elections and voter turnout in the aected units. Far from the fact that this claim was undoubtedly just another red herring, it has nothing to do with the question under review, and accordingly we reject it and hold that there is no proof that the postponement resulted in voter suppression to the detriment of the 1st petitioner. (v) Whether there were unexplainable discrepancies between the votes cast for Presidential candidates and other elective positions [38] The 1st petitioner was categorical that there was systematic ballot stung in certain Counties mainly in the Rift Valley and central parts of Kenya, where, according to him a total of 33,208 votes were cast for President only without corresponding votes for the other elective positions. [39] On their part, IEBC and its Chairperson, while acknowledging that, indeed in some instances there was vote dierential between those cast for President and for other positions, maintained that they were insignicant; that the instances related to votes by prisoners and citizens in the diaspora who only voted for President but not for the other elective positions. The dierential also includes rejected ballot papers and stray votes which do not count as valid votes. [40] The well-established principle that the person who asserts a fact must prove it casts the burden upon the 1st petitioner to demonstrate that there were instances of ballot stung of such a magnitude as to justify the nullication of the Presidential Election. [41] Ballot stung, which is the illegal addition of extra ballots, is a type of electoral fraud aimed at swinging the results of an election towards a particular direction. Not a single document has been presented by the 1st petitioner to prove systematic ballot stung. A gure of 33,208 votes relied on in this claim is based on unproven hypothesis, that since the number of votes cast for President is higher than those for the other positions then, without more, it must follow that there was fraud. Fraud is a serious criminal oence and must be proved beyond reasonable doubt . Under section 5(n) of the Election offences Act, it is an oence for a person to vote more than once in any election. [42] IEBC has proered a plausible explanation for the vote dierential. There are categories of voters who only vote for the President, such as prisoners and Kenyans in the diaspora. There were an insignicant number of stray votes, whose combined eect cannot justify nullication of the election. kenyalaw.org/caselaw/cases/view/240578/ 20