Methodology:
The number of buildings are counted in a bar chart for multi-family building type (Fig. 1).  The data was plotted to view the distribution. A mean of 87 kbtu/SF is calculated and focus on the poorest performing in the fourth quartile where the EUI is over 100 kbtu/SF (Fig. 2). The analysis compares peer multi-family  buildings  aka "typology", to prioritize energy projects and quantify policy interventions. The analysis then merged the LL84 data and the PLUTO shape file using the BBL. The map is made using a geopandas dataframe (Fig. 3). The zip codes of concern (Fig. 4) are defined and used to filter the MF EUI within those problem zip codes (Fig. 5).  A table of descriptive statistics summarizes the mean EUI in the zip codes where there is LL84 data (Table 1). The mean for zip codes of 11216 and 11207 show EUI values around or above the 4th quartile.