When the number of ties is large, H will be a little smaller than it should be. There is a correction for ties that makes it easier to reject the null hypothesis. If you do not correct for ties, your rejection of the null hypothesis will be harder, and your test is more conservative. We shall not deal with this correction formula because it has little effect on the value of H unless the number of ties is extreme. If you are interested in the correction, consult Hays (1973) and Siegel (1956).