I suspect that part of the popularity of WAR is that it’s an all-in-one statistic. For everyday players, WAR attempts to take into account various aspects of both offense and defense, and give the statistician (and the baseball fan) a way to make a comparison across players at different positions, and across different eras.
Whether it does that well or not, and whether it’s wound up being used for purposes which its developers did not intend, are completely different questions.
Too late to edit: an example is early-career Ozzie Smith. During the first half-dozen years of his career, Smith wasn’t a particularly good hitter – his batting average was typically .230 or so, and his OBP was usually just above .300. But, at that time, he was already widely considered to be one of the (if not THE) best shortstops in the game.
Looking at his WAR for those years (roughly '78 to '83), his WAR rating (Baseball Reference) was typically in the 3.0 - 5.0 range, and primarily driven by his defensive WAR.
So, offensive numbers alone (like OPS) told you little about how valuable even a weak-hitting player at a defensive position actually was.
There’s a specific recent example - the MVP race between Judge and Altuve. While there were some things that Altuve was better at, Judge was clearly the more productive offensive player, but on the other hand Altuve was a very good hitter at a much more important defensive position. Without an all-in-one stat like WAR, you just end up squinting and wondering; as it stood, they came out very nearly equal in WAR, which left other factors (such as Altuve’s better clutch numbers) as the deciding edge.
Well, because no measureable stat tells you which player is better. that’s the whole point; baseball fans (fans of all sports) love to argue over who is better. Joe Carter hit more home runs than Rickey Henderson, but that doesn’t mean he was BETTER, does it? Dave Magadan had a higher on base percentage than Roberto Clemente. Who’s better; Lou Whitaker or Dwight Evans? Should the Blue Jays trade Marcus Stroman to get Jimmy and Ernesto the prospects? No one measurable stat will explain that.
Well, first of all, your basic point is right. There are no players in baseball history with really high WAR totals who aren’t great players. There is no great player with a career WAR total of 4. A list of the top WAR players is an obvious list of outstanding players. I don’t think it will come as much of a surprise to most people that Babe Ruth is the all-time WAR leader, and the top twenty doesn’t include anyone like Omar Moreno. You might be surprised at some of the rankings, though. I know it catches my eye that WAR feels Al Kaline is one of the fifty best players of all time and Vladimir Guerrero barely makes the top 200, or that Sammy Sosa and John Olerud are pretty much the same. Of course it’s not going to tell you Sammy Sosa is worse than Sam Horn; everyone around 60 WAR was a really good player. But it’s more accurate than just guessing.
WAR is just the latest attempt at a single adjusted estimate, that’s all, and has become the most popular one because it’s easy to grasp. If I tell you Bill was worth three wins more than a replacement player, that’s easy to comprehend and goes to what baseball is about. If I tell you Bill is worth 27.4 VORP, it’s not clear what the hell that means. If I say he’s worth 11 Win Shares, again, what the hell does that mean? Is it good? Bad?
Of course, but we aren’t talking about extremes. You don’t need advanced analytics to tell you Clayton Kershaw is a great pitcher. But it helps to truly understand value on the margins.