Hayne J 25. from matters which I identified earlier in these reasons. The result declared was based on a scrutiny from which 1,370 ballot papers were excluded. The result depended upon who was excluded at the 50th exclusion point. The margin at that point was determined in the original and fresh scrutinies to be 14 votes one way but then (excluding scrutiny of the lost ballot papers) determined on the re-count to be 12 votes the other way. And the re-count yielded different tallies of votes and different decisions about rejection or acceptance of ballot papers from those reached in the original and fresh scrutinies, in numbers which cannot be dismissed as irrelevant or trivial. Those are reasons enough to conclude that it is more probable than not that the loss of ballot papers affected the result of the election which was declared. 105 If, as Mr Wang and Mr Mead allege, there were wrong decisions made in relation to reserved votes, the particular errors they allege could only reinforce the conclusion otherwise reached that the result declared was likely to be affected by illegal practices. 106 It is not necessary, in this case, to resolve any dispute about the meaning to be given to the word "likely" in the expression "likely to be affected" in s 362(3). It is, however, desirable to deal specifically with one submission made by Mr Dropulich. It was submitted, in effect, that if the Court could not take account of the records made in the original and fresh scrutinies about the lost ballot papers, the Court could not be satisfied that the result of the election was likely to be affected by the loss of the ballot papers. The Court could not be satisfied, the argument ran, because the Court could form no judgment at all. The Court could form no judgment because both outcomes (the result declared and the opposite result) were equally probable 67. This argument must be rejected. Wrongly, the argument treated the question of effect on the result of the election as requiring a petitioner to prove what the result would have been if the ballot papers had not been lost. The argument did not take account of all of the relevant facts that are known, including the closeness of the outcome, and the differences shown to exist between the original and fresh scrutinies and the re-count as to both tallies and rejection and acceptance of votes. It is more probable than not that the loss of the ballot papers affected the result of the election which was declared. Using the records about the lost ballot papers for other purposes 107 Subject to one possible caveat, those who submitted that the Court can and should have regard to the records which were made about the lost ballot papers all did so in aid of arguments that the Court should decide who would have been declared elected if the re-count had been conducted according to law. It is 67 cf Jones v Dunkel (1959) 101 CLR 298 at 304-305 per Dixon CJ; [1959] HCA 8.