Decrease in Run Production Might Be Due to Defense

This is just a short little post today, but I think it raises a decent question in light of comments made my Mark Cuban about how there is no evidence that shows human growth hormone improves athletic performance.  In fact, evidence shows that it has no significant effect on performance and there is also some information developing that shows that therapies using this hormone are unlikely to produce tissue that is functional for an athlete.  In other words, hGH therapies are probably ideal for individuals who simply need to heal as opposed to those who need to heal for athletic competition as well as not for those who used it to reduce down time between cycling of other chemicals.  This really is not much of a surprise because gym rats and athletes have rarely been on the cutting edge for any therapeutical chemical use.  Conventional wisdom tends to think that everything that successful players do must work because the players are successful.  We should recognize this as being absurd.  However, we tend to believe, as a population, in fantastically simple explanations, such as magic pills and treatments devised by non-science majors who have no practical skills beyond taste testing performance supplements.

With this in mind, shouldn't we also consider the possibility that something that has been shown to increase muscle mass and improvement in some forms of athletic events, such as anabolic steroids, might also not have been all that impacting?  A few years back, I noted how we have all of a sudden been treated to an amazing improvement in fastball velocity since around 2007.  Additionally, there also appears to be a relationship between velocity and run environments.  I think that makes sense and should be readily acceptable.

Another consideration I wish to make is this: run environment and appreciation of defensive ability are also linked.

Consider the graph below showing numbers from the American League over the past decade and a half:



How does that relationship work with the run environment?



It actually fits pretty well, which is an impressive thing because errors are a very limited measurement of defensive ability.  In fact, errors are a pretty poor way to measure defensive ability as evidenced by our infamous Rubbermaid Trashcan notion.  That is, such a trashcan would have a 1.000 fielding percentage while also being the worst defender in baseball.

Anyway, I think that it is plausible to think that along with increased pitcher velocity that a greater appreciation (or measurement) of defense probably also contributed to the decline in the run environment.  This more and more marginalizes the conventional wisdom of drug testing impacting the game in a very significant way.  As teams began to realize the importance of preventing runs then they began to move away from the pure mashers who produced runs at a significant cost to defense.  At least, that would be the narrative explanation to carry forth this hypothesis.

Do the Orioles Hitters Have a Home Run Problem?

Yesterday, Steve Melewski of MASN wrote about the Orioles recent problems on offense.  These problems aren’t all that recent though.  Since the beginning of September they’ve only averaged 3.7 runs per game while hitting .228/.286/.372 (AVG/OBP/SLG) as a team.  One of the problems that Melewski states is that the Orioles have been struggling on offense because they’re too one-dimensional.  The problem according to Melewski is that, “the Orioles are too reliant on the home run and they need more variety from their offense”.  

He’s not ENTIRELY wrong, but scoring runs via home runs should never be a problem.  Every year, you’ll hear about a team who hits too many home runs, and this year it’s the Orioles.  This isn’t the first time this has been said about the 2013 Orioles, but the rhetoric has picked up as the Orioles continued to fight for a playoff spot, while their offense has endured its worst month of the season.  

Melewski goes through some offensive statistics, comparing this year’s team with the playoff team from 2012, and as he shows, the 2013 offense is basically better than last year’s version in essentially every category, except for on-base percentage.  And that is the real issue.  What follows certainly isn’t groundbreaking analysis.  I’ve highlighted where the Orioles are specifically located on each graph, along with the Red Sox and Cardinals.

As you can tell (and probably already knew), a high OBP correlates better to scoring runs than hitting home runs, as evidenced by the corresponding R² values (the closer to 1, the better the correlation).  The Orioles appear to be enough of an outlier in the second graph that they should consider themselves lucky to have scored the 5^th most runs in the majors. If Baltimore’s runs scored correlated approximately with their OBP (according to this model), they should have scored right around 600 runs, which would give them a Pythagorean win-loss record of 69-86.  These correlations change slightly from year to year, but the underlying fact remains the same, as you can see from a sample of correlations from previous years.

The fact that Orioles hit a lot of home runs isn’t the reason that their offense has gone cold.  The reason is that they don’t consistently get on base (.311 OBP compared to a league average OBP of .318).  If more of Baltimore’s home runs had been hit with men on base, you could bet that relying too much on the home run would not be considered a problem.