In a proceeding pursuant to subdivision 2 of section 330 of the Election Law to declare void the primary election held June 17, 1969 and to direct the holding of a new primary election of the Democratic Party for the office of Borough President of the Borough of Queens, the appeal is from a judgment of the Supreme Court, Queens County, dated August 22, 1969, which granted the application and directed a special election to be held on September 16, 1969. Judgment modified to provide that the special election be held September 30, 1969. As so modified, judgment affirmed, without costs, with the following memorandum: When the number of voters participating in an election is large, there will always be some irregularities. Nevertheless, it is fundamental to the democratic process and to our republican form of government that no suspicion attach to the results of any election. This policy is the foundation of the “ rational standard ” evolved in Matter of Ippolito v. Power (22 N Y 2d 594, 597-598) that “ if irregularities are sufficiently large in number to establish the probability that the result would be changed by a shift in, or invalidation of, the questioned votes, there should be a new election ”. The balancing of the factors affecting that probability must be, of course, determined by the exercise of a fair discretion. In the present case the number of irregularities far exceeds the margin of victory of the successful candidate. There are distinguishing features here from Ippolito, but they do not influence the application of the rule or the policy that is its underpinning. The fact that more persons voted in the election here than in Ippolito should not alter the rule; otherwise, whether a new election should be held would be decided by an arbitrary mechanical test of the size of the vote cast and not by the rational formula of whether the probabilities are that the result would be different, but for the irregularities. Nor should the rule be shifted because four candidates, rather than two candidates (as ■in Ippolito), were running in the election. In either situation, the margin of defeat lies in the balance, which the irregularities, if corrected, could affect in the ultimate outcome. Finally, that here the contest took place at the same time as a city-wide primary of national interest, rather than in a local primary, should not be weighted heavily against a new election, for, plainly, this is a circumstance which might arise in many eases, and does not affect the Ippolito rule and policy. Accordingly, we affirm the exercise of discretion by Special Term. 'Christ, Brennan and Hopkins, JJ,, concur; Beldoek, P. J., and *518Benjamin, J., dissent and vote to reverse the judgment and to dismiss the petition, with the following memorandum: In the primary elections held on June 17, 1969 there were city-wide contests in both the Democratic and Republican Parties for the offices of Mayor, Comptroller and President of the City Council, and a county-wide contest in the Democratic Party for the office, inter alia, of Borough President of the County of Queens. The total vote cast that day in the County of Queens for the candidates in all contested primaries was 276,614 and the total vote cast for the four candidates for the office of Borough President was 166,313. The official canvass recorded the election of appellant Leviss by 95 votes over his nearest rival, respondent Santucei. It was stipulated by the parties that 698 votes were suspect or invalid for some kind of irregularity without any evidence of fraud or intentional misconduct. In addition, respondent contended that according to the public counter on the voting machines, there were 617 more votes cast than there were qualified persons who signed the voter registration cards; appellant claimed that the difference was only 367. Special Term stated in its decision that in the absence of any stipulation between the parties as to the actual number, it would accept appellant’s figure of 367. Thus, there was a total of only 1,065 irregularities found (erroneously stated in the decision to be 1,085) out of an aggregate vote of 276,614, or, as considered by Special Term, a total of 640 irregularities out of the aggregate vote of 166,313 for the office of Borough President. This latter figure is reached by reducing the total number of irregular votes by the same percentage which the votes for this particular office bears to the total number of votes east in all the contested primary elections. The majority of this court is of the opinion that a new election is required because of the narrowness of the vote margin and the fact that the “ number of irregularities far exceeds the margin of victory of the successful candidate”. These, however, are but two elements “in the complex of factors that must be considered in determining whether the recorded result truly reflected who was rightfully elected ” (Matter of DeSapio v. Koch, 21 A D 2d 20, 22, revd. on other grounds, 14 N Y 2d 735). We must consider the totality of circumstances and the nature of the election as a whole in evaluating the effect of the claimed irregularities on the final result and in determining whether the irregularities resulted in thwarting the true expression of the will of the voters. “ Close elections usually leave in their wake nagging suspicions that perhaps the true choice of the electorate was not declared the winner. But all elections do not result in thumping pluralities that give reassuring evidence of the clear-cut mandate of the People; and there is no law in this State providing that elections of a specified closeness must be rerun. The margin of victory, no matter how narrow, in and of itself cannot justify upsetting an election (Matter of McGuiness v. DeSapio, 9 A D 2d 65). There must at least be a showing that would justify a reasonable belief that the challenged irregularity may have accounted for the vitctor’s plurality. Absent such a showing there must be a finality to closely fought elections just as in closely fought lawsuits.” (Matter of DeSapio v. Koch, supra, p. 22.) Of the 1,065 clajmed irregularities, 367 have been attributed to the fact that apparently there were many more votes cast, according to the public counter on the voting machines, than there were qualified persons who signed the voter registration cards. The voting machines used in the election were equipped with a single public counter, and reflected the aggregate of the votes cast in both the Democratic and Republican primaries. The use of a single voting machine for both parties constitutes a relatively new procedure and, as stated in the dissenting opinion in Matter of Ippolito v. Power (22 N Y 2d 594, 600), such procedure “would produce more mechanical and human errors than *519are usually associated with elections.” Moreover, although it has been conceded that 698 persons voted illegally in the primary elections, consisting, in part, of votes by registered Liberals and Conservatives, and blank, void or missing party enrollments, the record does not present a scintilla of evidence to support any inference that these invalid votes were cast for the winning candidate or are otherwise attributable to him. Matter of Ippolito v. Power (supra) decided by a closely divided Court of Appeals, and relied upon by the majority, does not compel a new election. In that ease, which involved a contested primary election for the local office of District Leader, the total number of votes cast was 2,827 and the number of invalid votes was 101. Thus, the number of votes east was small and the percentage of irregularities relatively high. In this proceeding, however, as previously indicated, there . were 276,614 votes cast in all the primary elections and 166,313 votes cast for •the office herein involved, and the percentage of irregularities (1,065 or 640) is significantly small, being less than one-half of one percent of the vote. We do not believe that the Court of Appeals intended in Ippolito to enunciate an inflexible rule that in every case where the number of irregularities exceeds the difference by which the winning candidate was elected a new election is required. Rather, in each case, the principal inquiry for the court’s determination is whether under all the circumstances, the claimed irregularities, in all probability, has resulted in frustrating the desired choice of the electorate. We have here considered the county-wide nature of the office involved, the total number of votes cast for that office, the total number of persons in Queens who voted in the vigorously contested primary elections, the exceedingly small percentage of suspect or invalid votes as compared with the number of votes cast and the complete absence of any fraud, misconduct or willful improprieties on behalf of the winning candidate, and conclude that there can be no rational basis for ordering a new election. In our opinion, it is unreasonable to assume that such a minute percentage’ of suspect votes had any impact on the election’s outcome so as to upset the results of this primary. Particularly, in the instant election, where there were four candidates running for office, it seems to us to be eminently unfair and unjust to speculate that the winning candidate “ probably ” received a number of the suspect votes sufficient to give him his margin of victory. Whatever justification there may be for indulging in such assumption where there are only two candidates, it certainly should not be extended to the present situation. It would be a strange anomaly if in a new election the third or fourth candidate should now emerge the winner, thereby completely frustrating the will of the electorate who had voted in the instant primary. Because of the human and mechanical factors involved in our elective process, no election can be completely free of irregularity. However, not every election should be overturned, despite the closeness of the vote, because of the presence of irregularities. It depends on the circumstances of each case. In our opinion, neither the dictate of fairness nor the protection of the voter’s franchise demands a new election in this case.