Why WAR Shouldn't be Used in MVP Decisions

ESPN is bubbling with people pitching Mike Trout for the MVP over Miguel Cabrera because of his WAR. This is not an effecitve case for gauging MVP candidates for two reasons.

1.) There is no clearly established formula for WAR. Sites that provide the statistic, such as Baseball Prospectus, Fangraphs, and Baseball Reference, all calculate it differently; however, all of these sites calculate the value of WAR using these principles, and each site publicly acknowledges their methods for calculating their individual WAR values.

For instance, Baseball Prospectus has Mike Trout at 8.1 and Miguel Cabrera at 5.8. Fangraphs has Mike Trout at a 9.4 and Miguel Cabrera at 6.8. ESPN was quoting some extreme number with Trout at 10.3 and Cabrera at 6.3. 

2.)  The player that would replace the other is also completely subjective. We cannot know how a replacement player would do, or even who that player would be. Imagine the WAR of the various Angel center-fielders last year. Would the stat have picked up Mike Trout's unique talents? No. The stat is helpful for engaging conversations, but not for choosing MVP's. 

Let the standardized, time tested statistics tell the story, and let's not confuse people with subjective pontification.