Comment on Understanding the $gamma$-ray emission from the globular cluster 47 Tuc: evidence for dark matter?

In a recent paper Brown et al. (2018) analyze the spectral properties of the globular cluster 47 Tucanae (47 Tuc) using 9 years of Fermi-LAT data. Brown et al. (2018) argue that the emission from 47 Tuc cannot be explained by millisecond pulsars (MSPs) alone because of a significant discrepancy between the MSP spectral properties and those of 47 Tuc. It is argued that there is a significant ($>5sigma$) preference for a two source scenario. The second component could be from the annihilation of dark matter in a density spike surrounding the intermediate-mass black hole candidate in 47 Tuc. In this paper we argue that the claimed discrepancy arises because Brown et al. (2017) use a stacked MSP spectrum to model the emission from MSPs in 47 Tuc which is insufficient to account for the uncertainties in the spectrum of the MSPs in 47 Tuc. Contrary to the claims by Brown et al. (2018), we show that the significance of an additional dark matter component is $lesssim 2sigma$ when sample variance in the spectrum of a population of MSPs is accounted for. The spectrum of 47 Tuc is compatible with that of a population of MSPs similar to the disk population.