If you ask 20 different owners which pitching stat is most important, you'll get at least 25 different answers.
WHIP is (walks + hits)/(innings) - essentially how many baserunners a pitcher allowed. It is important; whether it is the MOST important stat is a matter of debate.
I would offer the following 3 pieces of advice:
(1) Certain stats have essentially no relevance in determining a pitcher's effectiveness: Wins, losses, saves, shutouts, complete games, and even really ERA, though the latter appears to play some role in how WIS estimates how many doubles and triples a pitcher allowed in real life.
(2) I would guess most owners here use some combination of the following: WHIP, OAV (opponent's batting average), BB/9, HR/9, and ERC.
(3) More important than the raw numbers are the + and # versions of these stats, which reflect normalization. Essentially the + stats tell you how much better or worse the pitcher was in that stat compared to his league average. 100 means he was exactly at the league average. > 100 means he was better, < 100 means he was worse. A pitcher with a 3.00 ERA in the 1908 NL will have a much lower (i.e. worse) ERA+ than a pitcher with a 3.00 ERA in the 1998 NL. The # stats are an estimate of how the pitcher would do in that stat against an "average" offense from across the WIS database. This is an oversimplification, and if you aren't familiar with normalization it's very much worth the time to get familiar with it. A good initial rule of thumb is that if you have two pitchers with the same "raw" stat - any of the ones I listed in (2) above - but pitcher A has a + stat of 110 and pitcher B has a + stat of 90, pitcher A will have a better # stat and will likely perform better on that metric than pitcher B.