No problem, only curiosity

Hello community!

Does anyone know why I am getting a result less than 2 million rows in this
example?

postgres=# create table test_n_tup_live (id serial);
CREATE TABLE

postgres=# insert into test_n_tup_live select generate_series(1,1000000);
INSERT 0 1000000

postgres=# analyze test_n_tup_live ;
ANALYZE

postgres=# select n_live_tup, n_dead_tup from pg_stat_user_tables where
relname = 'test_n_tup_live';

n_live_tup | n_dead_tup
------------+------------
1000000 | 0
(1 fila)

postgres=# update test_n_tup_live set id = id + 1;
UPDATE 1000000

postgres=# insert into test_n_tup_live select generate_series(1,1000000);
INSERT 0 1000000

postgres=# select n_live_tup, n_dead_tup from pg_stat_user_tables where
relname = 'test_n_tup_live';
n_live_tup | n_dead_tup
------------+------------
2000000 | 1000000
(1 fila)

postgres=# vacuum test_n_tup_live;
VACUUM

postgres=# select n_live_tup, n_dead_tup from pg_stat_user_tables where
relname = 'test_n_tup_live';

n_live_tup | n_dead_tup
------------+------------
1999925 | 0
(1 fila)

Thanks!