An observation relating to the leading decimal digits (see decimal representation) of many data sets, in particular that the leading digit is more likely to be small. According to the law, 1 is the leading digit for around 30% of data and 9 only 5%. The law applies best to data that are distributed over several orders of magnitude, e.g. city populations, but would not apply to human heights. The law has been applied to demonstrate likely fraudulence in some data sets.