aggregate on zero rows slow?

I am trying to figure out why given the same query:

SELECT MAX(fromoctets),MAX(tooctets)
FROM stats,trans
WHERE stats.id=stats_id
AND firsttimei=9224192
AND sourcepeeraddress='xxx.yyy.zzz.38'
AND flowindex=4701
AND timeslice<'15:20:00 Wed 1 May 2002';

Aggregate (cost=57.36..57.36 rows=1 width=16)
-> Nested Loop (cost=0.00..57.36 rows=1 width=16)
-> Index Scan using firsttimei_idx on trans (cost=0.00..54.33 rows=1 width=12)
-> Index Scan using stats_pkey on stats (cost=0.00..3.02 rows=1 width=4)

the select appears to take much longer to complete if the WHERE clause
matches zero rows, than if it matches some rows.

Is this possible, or just chance?

Cheers,

Patrick