Low-frequency impact sound insulation, down to 20 Hz, has a significant effect on humans' dissatisfaction due to noise in timber buildings. Today, the low-frequency procedure of the ISO 16283-2:2015 impact sound measurement standard covers the frequency range down to 50 Hz for the use of an ISO tapping machine, but does not yet cover the use of an ISO rubber ball. Here, microphone grid measurements were made in two small rooms that were excited by an ISO rubber ball from the rooms above. In each grid, 936 microphone positions were used to capture data representing the full spatial fields of impact sound pressures from 10 to 500 Hz for one excitation location for each room. The data show that the positions at the radiating ceiling surfaces have low maximum sound pressure levels compared to the pressure levels at the floors, especially in the floor corners. First, a measurement procedure to predict the maximum exposure of low-frequency sound in a room is proposed It is suggested that the maximum values for each frequency band in the corners opposite to the partition being excited (i.e., the floor corners) be used. Second, a procedure to predict the room average sound pressure level and the prediction's normal distribution is suggested. Iterative measurements with random microphone locations and random excitation locations are used. The advantage of this method is that the required precision and information about the sensitivity due to different excitation points are obtained.