FWIW, there is an interesting stat to measure batting runs produced that splits the difference between "players should get no credit for context/timing" measures like wOBA and "let's just pay attention to a piece of the puzzle and ignore the rest" stats like RBI: RE24.
In short, rather than just look at the production of the player, it counts the production in the context in which occurred. So, it actually counts the downside of Phillips low OBP and SLG in general, but gives him "extra" credit for coming through so often with men on base. Now, you can choose how much "credit" you want to give Phillips for the timing of his performances, but basically, this stat weights his performances by the leverage of the situation.RE24 (runs above average by the 24 base/out states): RE24 is the difference in run expectancy (RE) between the start of the play and the end of the play. That difference is then credited/debited to the batter and the pitcher. Over the course of the season, each playersí RE24 for individual plays is added up to get his season total RE24.
So let's look at a few numbers.
wOBA: Total measure of batting production using linear-weights (average value of a given outcome without consideration of context), scaled to OBP
wRAA: Batting runs produced relative to the average hitter, calculated based on wOBA
RE24: Batting runs produced relative the average expected outcomes give the base-out situations in which the player batted.
Diff: RE24 minus wRAA. A positive value means the player's "real" production in terms of affecting the team's chance to score runs was greater than we would assume if we just looked at total outcome stats irrespective of context.
What do we see? By wOBA, Phillips is almost exactly average. But when you look at his performances based on how they affected actual run expectancy (relative to what we would expect an average hitter to do), Phillips has actually produced 19 runs more than average. This difference between RE24 & wRAA is the largest in baseball.Code:wOBA wRAA RE24 Diff Choo .392 27.2 32.0 4.8 Cozart .268 -13.7 -22.1 -8.4 Votto .403 30.7 29.0 -1.7 Phillips .316 0.8 19.4 18.6 Bruce .349 11.7 15.6 3.9 Frazier .323 2.8 7.1 4.3
By contrast, Zack Cozart 's crappy performance has actually cost the Reds 8 runs MORE than you would think just looking at his wOBA. Ouch!
If you're a believer in giving players credit for clutch performances, for RBI, etc. -- essentially, giving players credit for not just how they perform but by giving more credit to performances in high leverage situation (for better or worse), RE24 is the stat for you.
Now, if you went in to Fangraphs' WAR and you replaced "Batting" with RE24 (which I wouldn't do exactly b/c "Batting" does an adjustment on wRAA to account for park effects), you'd find that Phillips, Votto and Choo are all virtually tied at around 3.6 WAR. Not that I would do this, but if you wanted to, I could see the Phillips for Reds' MVP case in a more reasonable light.
And on an interesting side note, I looked at the difference between wRAA at the team level. On average, team RE24 is ~7 runs lower than wRAA, so the systems don't match perfectly. But the relative comparisons are still very interesting. And what do we find? The Cardinals and A's have the highest difference at +39 runs. That is to say, their offense has produced 39 more expected runs than the mere sum of their individual batting lines would suggest. Meanwhile, the Rockies are 105 runs worse. OUCH!
In the FWIW department, wRAA appears to be more strongly correlated with actual run scoring at the team level than RE24 --- so you could consider it a better predictor of actual run scoring. But I thought the RE24 stuff was interesting nonetheless.