It uses a stat called RE24, but it isn't all that hard to understand. Statisiticiams look at all games played by all teams over a long period of time (at least a decade - I can't remember exactly how long), and measure how often teams score in each of the 24 different "base-out states." Man on first, two outs, how often on average does a MLB team score? And so on, for every single possible situation. Then, they measure how much the difference in the probability of a team scoring is from the time before a hitter's AB, and after it is over. The example used in the article is an AB where Cabrera came up with two outs and runners on 1st and 2nd - a situation in which, on average, a team is expected to score .33 runs.*
*[Contrary to what has been suggested earlier in this thread, this is not "speculative" either - it is based on real world averages over huge sample sizes. And these expected run values are at the very heart of all "linear weights" based sabermetric stats, which form the basis - in one form or another - of WAR, or at least the offensive part of it.]
Instead, Cabrera hit a three-run home run, so under the stat "RE24" Cabrea is credited with the difference between the runs actually socred and the runs they are expected to score (+2.67).
This stat is entirely based in the real world, and gives players credit for context (even outs) that advance the probability of scoring in an inning. Every single offensive play a player was involved in is logged - either as a positive, or a negative value. Offense only.
The article links to the leaderboard, but for the sake of simplicity, here's the Top Five:
1. Mike Trout: +56.52 runs
2. Edwin Encarnacion: +54.44 runs
3. Prince Fielder: +48.12 runs
4. Joe Mauer: +46.51 runs
5. Miguel Cabrera: +45.18 runs
The biggest single reason for the difference? Double plays. Cabrera had more "big" positive plays than Trout did, but Trout had more overall positive plays and far fewer negative ones (where the result of your AB is a reduction in the probability of your team scoring from what was expected before your AB).
Of course, this matters. Cabrera supporters can't point just to the big positives but ignore the big negatives. [For that matter, they shouldn't be allowed to ignore defense, or going first-to-third, or stealing almost 50 bases at a 90%+ clip, but that is another story.]
Just offense. Give credit for context when it is earned, but count the negatives when those are earned, too. As Cameron puts it,
quote:"You can go through each player’s play logs and see exactly where they earned and lost credit. There’s no replacement level here. We’re not dealing with defensive metrics that require some subjective inputs and can’t be easily replicated. This is just pure offense, and the total value of all the plays that both Trout and Cabrera were involved in.
And Trout still comes out on top. Ignore defense. Ignore things like going first to third on a single, or taking the extra base on a fly ball. Ignore WAR. Trout still wins. This is how amazing his season actually was. Even if you strip away the things that make Mike Trout special, he was still the best offensive performer in the American League this year, even while starting the season in the minors. This isn’t just the best performance of 2012 – it’s one of the best individual performances in the history of baseball."