flight_no by origin: a 8% spread

Breaking flight_no down by origin on flight-delays, because the headline average is 3,560 and no segment is actually there.

• SFO — mean 3,712, median 3,729 (551 rows)
• DTW — mean 3,648, median 3,703 (564 rows)
• DFW — mean 3,648, median 3,651 (612 rows)
• MSP — mean 3,633, median 3,739 (567 rows)
• JFK — mean 3,617, median 3,800 (548 rows)
• PHX — mean 3,594, median 3,595 (571 rows)

Top to bottom that is 3,712 against 3,431, a spread of 8.2%. The pooled average is 3,560.

select origin,
  count(*)                                               as rows,
  round(avg(flight_no)::numeric, 2)                      as mean,
  percentile_cont(0.5) within group (order by flight_no) as median
from flight_delays
group by 1
order by mean desc;

The groups are close enough that the pooled average is a fair summary. That is worth confirming rather than assuming: the check costs one query and the failure mode is invisible.

Notice the mean and median columns disagree only slightly here. Always compute both in the group-by. The comparison between them per segment is free and tells you whether you are looking at a level difference or a tail difference.

This is the setup for Simpson's paradox — the case where every segment moves one way and the total moves the other.