Confusing contradictory data: many Snowflake users in Russia in Sept, or few?

I've been putting together a presentation for next week, and I've run into a contradiction that I don't yet know how to explain.

On the one hand, we have a graph that @hiro made with @dcf's sum/max scripts from #40012 (closed) of the number of Snowflake users in Russia from March until recently:

On the other hand, we have a graph of the number of "Bridge users by transport for Russia", https://metrics.torproject.org/userstats-bridge-combined.html?start=2022-03-01&end=2022-10-23&country=ru :

Now, we know from the tpo/anti-censorship/pluggable-transports/snowflake#40207 (closed) analysis that these Snowflake users in Russia in September are more plausible as Iranian Snowflake users who are mis-geoip'ed as being in Russia.

But ignoring why the data is how it is: what's going on? Does our data say there were a bunch of Snowflake users in Russia in September, or does it not?