# How robust is this 2020 measurement of the external field effect (EFE) in MOND?

There's a paper from a few months ago that claims to have measured this unique Modified Newtonian Dynamics (MOND) effect at high significance. The result seems very interesting:

• The effect is unique to MOND and not easily replicated by Lambda-CDM.
• The effect was measured in 100+ galaxies. In galaxies which were in strong external gravitational fields, the EFE was detected; in those that weren't, it was not detected.
• The analysis was conducted blind, so experimenter bias is not an issue.
• The authors analyzed possible systematics and didn't find anything.

Clearly if this result is robust then it would be very groundbreaking. How robust is this measurement? It's been a few months and the paper seems to have gotten only a few citations which don't directly address the measurement, leaving it unclear to me how the community is reacting to the paper.

• I haven't looked into it in detail, but the statistics don't seem to me to be robust at all. They claim to show that $e = e_{\text{env}} \neq 0$ as predicted by MOND. What actually shows up in their data (figure 5) is much messier. There is a cloud of data points with no apparent relationship between the two, along with a few outliers that have very high $e$, or very high $e_{\text{env}}$. They then ignore this and compare the means to show that $e = e_{\text{env}} \neq 0$, a conclusion that is totally driven by outliers, each of which don't remotely have $e = e_{\text{env}}$. – knzhou Jan 19 at 2:55
• It is as if I assembled a sample of 100 people, of which one individual was very rich, and another different individual was a redhead. Then I compute the rate of richness and redheadedness within the sample to be both 1%, and since these percentages are the same, I conclude that being redheaded ensures you will be rich. – knzhou Jan 19 at 3:00
• See related topic here: physics.stackexchange.com/questions/605621/… – MadMax Jan 19 at 19:35