In his seminal work, Cleve [STOC '86] has proved that any r -round coin-flipping protocol can be efficiently biased by (1 r ) . The above lower bound was met for the two-party case by Moran, Naor and Segev [Journal of Cryptology '16], and the three-party case (up to a polylog factor) by Haitner and Tsfadia [SICOMP '17], and was approached for n -party protocols when nloglog r , however, the best bias for n -party coin-flipping protocols remains O ( n r ) achieved by the majority protocol of Awerbuch, Blum, Chor, Goldwasser, and Micali [Manuscript '85].

Our main result is a tighter lower bound on the bias of coin-flipping protocols, showing that, for every constant 0"> 0 , an r -party r -round coin-flipping protocol can be efficiently biased by (1 r ) . As far as we know, this is the first improvement of Cleve's bound that holds in the standard model, and is only n = r (multiplicative) far from the aforementioned upper bound of Awerbuch et al.

For proving the above lower bound we present two new results that we believe are of independent interest. The first result is that a sequence of (augmented) weak martingales have large gap: with constant probability there exists two adjacent variables whose gap is at least the ratio between the gap between the first and last variables and the square root of the number of variables. This generalizes the result of Cleve and Impagliazzo [Manuscript '93], who proved that the above holds for strong martingales. The second result is a new sampling algorithm that uses a differentially private mechanism to minimize the effect of data divergence.