The month of April is now well in the rear-view mirror, and with it, over a month's worth of statistics with which to dive into. While many statistics have yet to stabilize enough to knock off the warning of “small sample size”, others can give us a decent idea of where teams might be headed. I wanted to take a look at something that could very realistically affect the Rangers moving forward: Regression.
One particular area of interest in recent weeks has been hitting with runners in scoring position (RISP). The Astros and Yankees, for example, faced each other in the 2015 American League Wild Card Game. Yet, after Sunday’s games, the Astros sit at 8-17, the Yankees at 8-15, both teams last in their respective divisions. One major reason for that? They’ve struggled to hit with RISP.
Back in 2013, the St. Louis Cardinals tore through the National League by shattering the record for batting average with RISP at .330. The previous record had been .311 by the Detroit Tigers in 2007. In an effort to look at a more all-encompassing view, wOBA is perhaps a better measure, as it more properly weights all offensive contributions. A wOBA of .372 with RISP in 2013 was nearly 30 percentage points higher than the next-best team in baseball, and 52 percentage points better than the next-best team in the National League.
Although it’s often hotly debated among fans, hitting with RISP hasn’t exactly proven to be a repeatable skill. That is, a team that excels in RISP scenarios in one season won’t necessarily replicate that performance in the next. And indeed, the St. Louis Cardinals of 2014 ended up putting up a wOBA of .307 with RISP in 2014 en route to scoring 174 fewer runs in the process.
Knowing that, I wanted to take a look at MLB team stats this season and see which teams might be due for some offensive improvement -- or in some cases, the opposite.
Fortunately, Tom Tango helped us out quite a bit in the appendix to The Book. He provided us a road map with which to perhaps “predict” any regression, if any is to be expected at all. First, I compiled a list of wOBA statistics through May 10 for both RISP scenarios and non-RISP scenarios. Using that, we can come up with the statistical uncertainty in those numbers using the formula "σX=√(wOBA(1.1-wOBA)/(N))", where N is the number of plate appearances in each scenario.
Having that, we can use Tango’s appendices to help us come up with an “expected” wOBA -- or “RISP skill”, if you will. Using the data we come up with, we can compare the difference between each team’s current RISP wOBA and the “expected” wOBA, as you can see in the “DIFF” column in the table below.
It’s important to note that, for the purposes of this discussion, regression to the mean isn’t necessarily a bad thing. In the case of teams that have a positive difference, those are teams who can expect their regression to improve hitting with RISP.
As you can see, a significant number of teams can likely expect some increase in RISP offense, while a few near the bottom can actually expect to get worse. Sorry Braves fans, things don’t figure to get much brighter in 2016.
However, we can see that teams like the Astros and Yankees, the two I talked about previously, can both expect increases in wOBA with RISP. In the case of the Astros, that improvement looks to be fairly significant, which would lend credence to the idea that, although they started the season slow after being a World Series pick by some, they might have what it takes to recover and still make a run at the postseason.
Teams like the Boston Red Sox who find themselves with negligible differences can mostly take their team performances at face value. That is, they are performing at or around their expected skill level, offensively.
Now, there is no real timetable for how long these stats might take to normalize a bit more, if at all. In the case of the 2013 Cardinals, they were expected to regress to the mean for nearly an entire season, yet defied the odds. That tends to be the exception, rather than the rule. We do know, however, that with enough time -- in this case plate appearances -- these numbers are likely to regress. And although the Houston Astros and New York Yankees both find themselves in last place today, I don’t expect that to be the case two months from now, all other things remaining the same.
Even the Chicago Cubs, the team with a whopping +103 run differential, can expect some upward regression. Yes, the best team in baseball should actually count on their offense becoming even more potent in run-scoring scenarios. Interestingly enough, the Cubs have scored 9 more runs on the season than the next-closest team -- the Boston Red Sox -- while having the highest number of plate appearances with RISP. That alone would seem to tell us what we need to know about RISP stats: It's far more important to be able to hit and get on base in all situations than to be good at situational hitting. In the case of the Cubs, this means getting runners on base en masse and counting on the law of averages to come into play.
So, how does this affect the Texas Rangers? As you can see in the table, they're the team that would predict, based on the numbers, to regress the fourth-most in MLB in RISP situations. So while last night's win might have been all fun and games regarding that big 7-run 8th inning, that's not likely to be the norm moving forward.
In being actually below-average in terms of overall wOBA, Texas will need to improve on its overall hitting in order to remain in the hunt for a second straight AL West crown. We could argue that Prince Fielder is dragging that down a decent amount -- a fair argument, to be sure -- but it's also worth noting that Shin-Soo Choo figures to help the offense in the aggregate upon his return.
Whatever ends up being the case, the hope would be that regression doesn't bring the Texas Rangers offense crashing to earth before we hit the high-temperature days of summer. In allowing the 5th-most runs in the American League -- as well as a pitching staff that isn't exactly impressing with peripherals -- the offense will need to remain effective going forward.