A Substitute for War

Basketball philosophy

Stats Tuesday: A statistical case for Miguel Cabrera as American League MVP + Should a runs involved stat replace RBIs?

leave a comment »

English: Miguel Cabrera at Dodger Stadium.

English: Miguel Cabrera at Dodger Stadium. (Photo credit: Wikipedia)

The AL MVP race is the most interesting in years due to the debate of traditional stats vs new age sabermetrics, as many have pointed out. As of October 1st, 2012, Miguel Cabrera is in position to win the Triple Crown of leading the league in the traditional hitting stats Batting Average (BA), Home Runs (HR) and Runs Batted In (RBIs), but Mike Trout holds a significant lead in Wins Above Replacement (WAR), 10.7 to Cabrera’s 6.8 according to baseball-reference.com, a massive gap.

Trout’s gap in WAR comes from two places. One, he is given 2.3 points in defensive WAR due to excellent fielding in center-field, while Cabrera’s score is -0.2 at 3rd base. Secondly, Trout’s offensive WAR (8.6) beats out Cabrera’s score of 7.4 on baseball-reference. This is interesting because it is clear that Cabrera is the better hitter statistically, leading in OPS, batting average, slightly trailing in on-base percentage, and having a sizeable gap in home runs. A fascinating statistic is that Cabrera has 137 Runs Batted In (RBI) to Trout’s 83.

RBIs have of course lost favor in recent years for obvious reasons. They’re simply too context based. A player’s RBIs depends on the ones in front of him who get on base. In this case, Trout’s offensive value is more reliant on scoring from the bases than Cabrera’s is. For one, he hits lead-off, a position where where worse batters behind him mean there’s less chance of players to be on base – and the first bat of the game no players can be on base. Secondly, Trout is a fantastic base-runner and base stealer (Leading the AL in stolen bases at 48, amazingly only being caught 4 times). Third, by the fact that less of hits are home runs than Cabrera’s despite a higher on-base percentage, this is another reason why more of his hits end up with him on the bases, waiting for the batters in front of him to hit him in. RBIs of course can’t measure the value Trout brings by getting into scoring position. When a batter in front of him hits Trout in, they are credited with an RBI, while he is given a Run Scored, which is tracked but not given much weight in awards voting. Trout, unsurprisingly, leads the AL in Runs Scored by far with 129 – Cabrera coming in 2nd at 109. Trout leading Cabrera in offensive WAR comes down to favoring this “scoring from the bases” advantage Trout has over Cabrera, outweighing the extra damage Cabrera does at the plate.

We know Trout’s Run Scored nor Cabrera’s RBIs advantage over each other isn’t indicative of their value. Their roles are different, Trout’s favoring scoring off the bases and Cabrera with the bat. This is concerning because fundamentally, we should want to measure how much runs a player actually scored in a game, in a same way we want to know how many points a basketball player actually scored.

So what I came up with a little stat to try and incorporate both runs batted in and runs scored off the bases for a player. I can’t be the only one who’s tried this, but nonetheless here’s what I did: I took a player’s Runs Scored and subtracted Home Runs, as when a player scores a Run off a homer, it’s counted as an RBI. A good term for the total Runs Scored not counting Home Runs, is “Baserunning Runs Scored”. I then added this number to RBIs. The stat thus is simply (Runs Scored – HRS) + RBIs, or Baserunning Runs Scored + RBIs.

Thus this counts the “Total Runs Involved In” for a player, accounting for the runs a player scores either with his bats or from baserunning – which he deserves because it was his at bat appearance to put himself into scoring position. In the case of a player like Trout, importantly this also accounts beautifully for his baserunning and stolen bases. His strengths in this area put himself in scoring position more often, increasing the chance that he would be driven home – and he is rewarded for it statistically by exactly as much as it increased his team’s runs on those plays.

Furthermore, while Cabrera batting 3rd allows him to get RBIs by driving in 1-2 in the order more than Trout does batting leadoff, Trout batting lead-off allows him to get more Runs Scored by the Angels 2nd and 3rd best hitters Albert Pujols and Torii Hunter waiting to bring him home with their bats. This cancels out a lot of the effect of Cabrera’s position in the lineup boosting his RBI stats.

So what are Trout and Cabrera’s numbers? As of October 1st, 2012:

Mike Trout –

129 Runs Scored – 30 Home Runs = 99 Baserunning Runs Scored.

83 RBIs

99 Baserunning Runs Scored + 83 RBIs = 182 Total Runs Involved In

Miguel Cabrera

109 Runs Scored – 44 Home Runs = 65 Baserunning Runs Scored

137 RBIs

65 Baserunning Runs Scored + 137 RBIs = 202 Total Runs Involved In

Despite a massive gap between Trout and Cabrera in runs scored off the bases in Trout’s favor, it is not enough to make up for the RBIs gap Cabrera has and thus Cabrera is involved in 20 more Tigers’ runs via either the base or off the bases. In the WAR calculation 10 runs is worth about 1 W, so Cabrera’s 20 runs advantage is worth about 2 wins. Because Trout is credited with +1.4 in OWAR over Cabrera, a move to +2.0 for Cabrera makes the the swing +3.4 for Cabrera. Trout leads by +3.9 in overall WAR, a swing of +3.4 in Cabrera’s favor drops Trout’s lead to +0.5. A reasonable estimate of the “new” total WARs (splitting their total as a guideline) is about 9.0 for Trout and 8.5 for Cabrera. Trout maintains an overall lead because of the value of his defense adding +2.5 to his WAR, assuming the DWAR numbers are trustworthy, while Cabrera’s offensive advantage is only +2.0.

These stats indicating an advantage for Cabrera on the offensive end and a defensive one favoring Trout, with the overall number supporting Trout, seems very reasonable to me. A few other things that could plausibly could be argued for in Cabrera’s case:

With identical plate appearances as of October 1st of 6049 usefully for this comparison, the Angels have scored about 6% more total runs than the Tigers. A better offense increases both a player’s chance to get an RBI (by players getting on base more), his chance to get a Baserunning Run (by players hitting him home more), while also tiring out opposing pitchers more often. The difference between the Angels’ and Tigers’ offenses is not incredibly significant, nonetheless one could try to adjust for it. A decrease of 6% to Trout’s runs involved in to match this team advantage, would drop his total runs involved in from 182 to 171, worth about 1.1 W. In the previous calculation Trout’s lead in WAR was trimmed to +0.5, so even less than half of that team adjustment would bridge that gap between the two MVP frontrunners.

Secondly, team offense can also be used to argue Cabrera meant more to HIS offense than Trout, as Cabrera’s Total Runs Involved In number is 27.9% of the Tigers’ total runs, to Trout’s 23.7% of the Angels’, which is a more sizeable gap % wise than the Total Runs Involved In difference of 202 vs 182 indicates. It’s always a debate in the MVP races of every sport whether a player should be judged on how indispensable he is to his team or whether they should be judged in a vacuum, but Cabrera clearly holds a bigger “share” of his offense.

Third is introducing the concept of veteran leadership in Cabrera’s favor, which is an angle I’ve seen thrown out by a few writers and fellow players. While I wouldn’t use this as a significant reason to decide on the MVP race in a sport as individualistic as baseball, I don’t think it’s unreasonable that Cabrera at his age could be having more of a positive psychological effect on teammates’ performance than Mike Trout.

Overall, I don’t think there’s a necessarily wrong answer in this MVP race. If I had to choose, I’d likely go with Cabrera, but it’s a tight decision.

From a larger baseball perspective, I wonder this concept of Total Runs Involved In is the future of the RBIs stat. RBI is simply too flawed to last much longer in a numbers game like baseball. However including the runs the player contributes to by getting on base, as well as by bat, seems like a much stronger statistic, removing a lot of the context concerns of RBIs. It’s hard to imagine an MLB that doesn’t take seriously the concept of “Who’s scoring the runs” in some form another, but perhaps the move is to evolve to meet the higher theoretical standards that the Moneyball MLB era has brought in. Runs production doesn’t have to go away, but perhaps it needs to adapt. Measuring how many runs a player is involved in via the bases or bat isn’t a perfect measure of a player’s value to his team’s runs, but I believe it’s much closer to it than RBIs.


Written by jr.

October 2, 2012 at 1:35 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: