Length of a minimum as predictor of next solar cycle's strength

Motivated by a prevailing view that a long minimum leads to a weak sunspot cycle, we estimate the correlation coefficients between the length of a cycle minimum and (i) the following cycle's peak, (ii) the preceding cycle's peak, (iii) following peak minus preceding peak and (iv) depth of minimum. Using both sunspot number and spot area data, we find that a long minimum is both followed and preceded by weak cycles. Similarly short minima are followed and preceded by strong cycles. Consistent with these results, we find no correlation between the length of a cycle minimum and the difference in peaks of the following and preceding cycles. From sunspot number data, for longer-than-average minima, five following cycle peaks were lower than that of the preceding cycles' peaks, while four were higher. Following shorter-than-average minima, seven cycle peaks were higher than the preceding peaks and seven were lower. Therefore one cannot predict from the length of a minimum whether the next cycle will be stronger or weaker than the preceding cycle. Thus we cannot predict whether cycle 24 will be stronger or weaker than 23. We also find that there is a strong anticorrelation between the length of a solar cycle minimum and the depth of that minimum. We define the depth as the least spot number or spot area (13-rotation averaged) within the span of a cycle minimum. We speculate that this anticorrelation is due to the longer time available for annihilation of late cycle toroidal flux across the equator in the case of a longer minimum.