Wednesday, October 8, 2014

Part I: How good were A.J. Hinch's D-Backs anyhow?

A bit more than a week ago, when A.J. Hinch's hiring was announced, I promised to take a look at the '09-'10 Diamondbacks to see what kind of talent he had.  At the time, I wrote:
"But it's not like A.J. Hinch inherited the '96 Yankees and turned them into the 2014 Red Sox.  The Diamondbacks have been, to understate the case somewhat, volatile in recent years.  It is difficult to determine their true talent level from the 2009 season onwards because of the volatility of their performance in subsequent seasons."
And the last sentence of that paragraph went like this:
"This warrants a post in it's own right, and I plan to investigate further later in the offseason."
And here we are!

Actually, this will be two posts.  This post will concentrate on the team effort only, and the next post will look at the production of the individual players that Hinch managed.

The main concern / criticism / negative comment made when Hinch was hired was that he was a horrible manager with a Diamondbacks team that was expected to contend, and therefore would be a horrible manager with an Astros team hoping to contend.  And yes, superficially, Hinch managed a team for parts of two seasons, and posted a horrible record (89-123), eventually getting fired and departing the organisation for a front-office scouting position in San Diego.

So let's look further into the performance of the Diamondbacks around that time with an eye to assessing their "talent level" (whatever that is!) in those seasons.  I am making the assumption that the assessment of the talent is greatly assisted by the four-plus full-seasons that have elapsed since Hinch's firing.  For example, if Hinch managed a hypothetical starting rotation of David Price, Felix Hernandez, Yu Darvish, Clayton Kershaw and Jon Lester to an ERA+ of 75, with a combined WAR of -0.3, and since then they have all posted an ERA+ of 150 and a combined WAR of 967, then you can be pretty sure that whatever happened on Hinch's watch was not good, and this may speak to his abilities as a manager.

But before looking at individual players, lets look at the collective effort.

And first up, a caveat:  I have no axe to grind in this analysis.  I am literally writing the article as I go.  I have literally no idea what I will find - if anything.  Whatever I find will probably be boring and unremarkable, and will make for a long, boring, offseason post.  But I also want to echo my previous statements in the last article when I identified how he has worked in baseball as a manager, in player development and in scouting, and blending these features seems to be what the Astros find desirable.  Those features, regardless of what I find, make Hinch an excellent managerial candidate, and I am certainly not trying to either sink him, or support him.  The decision has been made, and Hinch's hiring simply makes the 2015 season all the more interesting.

Right, lets look at the Diamondbacks' team statistics in terms of raw win-loss records since 2007:

  • 2007: 90-72 (first in NL West)
  • 2008: 82-80 (second)
  • 2009: 70-92 (last)
  • 2010: 65-97 (last)
  • 2011: 94-68 (first)
  • 2012: 81-81 (third)
  • 2013: 81-81 (second)
  • 2014: 64-98 (last)

Doesn't look good for the '09-'10 seasons, does it??  Two last place finishes, bookended by a second and a first. But the 2008 season - where the Diamondbacks finished second with a grand total of 82 wins is clearly an aberrant season.  Lets look deeper, see if we can find anything in the runs for and runs against, and the calculated Pythagorean W-L projected record:

  • 2007: Pythag 79-83; 712 runs for, 732 runs against; win difference +11
  • 2008: Pythag 82-80; 720 runs for, 706 runs against; win difference 0
  • 2009: Pythag 75-87; 720 runs for, 782 runs against; win difference -5
  • 2010: Pythag 69-93; 713 runs for, 836 runs against; win difference -4
  • 2011: Pythag 88-74; 731 runs for, 662 runs against; win difference +6
  • 2012: Pythag 86-76; 734 runs for, 688 runs against; win difference -5
  • 2013: Pythag 80-82; 685 runs for, 695 runs against; win difference -1
  • 2014: Pythag 67-95; 615 runs for, 742 runs against; win difference -3

This re-analysis does seem to shed some more light on the D-backs, and their inter-season volatility.  In both of their first placed seasons on this list, they have handsomely outperformed their Pythagorean win-loss projected record by 11 and 6 games - possibly because they sold their souls to the devil, and have paid in subsequent seasons.  This is probably not an uncommon trait amongst first place teams - if a team outperforms it's pythagorean win-loss projected record, then surely it is more likely to finish in first place, right?  The first place finishes, however, are the only years that the Diamondbacks outperformed their Pythagorean projected record.  Interesting.  Perhaps a manager could be given credit for this, although most statistically aligned baseball writers would normally put it down to plain luck.  The regression in the years around the first place finishes certainly suggests that luck is a major player in the inter-season volatility of win-loss records.

The D-backs were remarkably consistent in terms of runs scored between 2007 and 2012, before suffering large declines in 2013 and 2014.  If one wants to look at this period critically, you might say that there seems to be about 2 wins worth of runs scored between '09-'10, and '11-'12.  But the runs scored were certainly consistent with the two years before Hinch's reign of terror.  So perhaps Hinch played the Diamondbacks out of a few runs here, but I doubt it.

The Diamondbacks were less consistent when considering inter-year variability in runs allowed.  It is evident that the poor records in 2009 and 2010 were all about allowing the other team to score too often.  The 2009 and 2010 runs allowed totals are 40 and 94 runs are respectively more than the third highest year in this sample, which is their most recent last-place year (2014).

So the key to the slump in 2009 and 2010 is almost entirely to do with pitching, combined with whatever you want to believe about what the win difference between the Pythagorean predicted win-loss record, and the actual win-loss record.  It may pay to bear this in mind when looking at how individual players have performed under Hinch's watch.

There is one other way in which the Diamondbacks with and without Hinch at the helm can be compared.  Hinch took over from Bob Melvin on 8 May 2009, and was given his marching orders on 1 July 2010.  Hinch's record as a manager has been well publicised (89-123, 0.420 win percentage).  Bob Melvin managed the Diamondbacks to a 12-17 record as at 7 May 2009; after Kirk Gibson took over in 2010, the D-backs finished the season with a 34-49 record.  So in the '09-'10 seasons, in games not managed by Hinch, the Diamondbacks' other managers combined for a 46-66 record, or an 0.411 record.  No real difference either way there.

So I guess that we can assume from this that A.J. Hinch is not such a grotesquely clueless manager that he was vastly worse than a combination of Bob Melvin and Kirk Gibson.  If the Diamondbacks under performed around the time that Hinch was at the helm, it was most likely due to pitching.  I am certainly not seeing any big patterns here, which probably indicates that the "all I remember was that he was an awful manager" narratives are most likely overblown and/or inaccurate.

Regardless, there is no evidence that A.J. Hinch is entirely clueless, but we already knew that, because he was hired for a second manager gig.  We have to give the Front Office some credit, after all.  The Front Office also isn't entirely clueless, although given the public relations disasters over the last few years, sometimes it seems that they are.  And this article suggests that other Front Offices don't see eye-to-eye with the Astros.

Next up, lets look at how the players in the teams that he managed performed compared to the rest of their careers.