FanPost

Sandy, Where do Wins Come From? [With edits]

LET us go then, you and I,

When the empty upper deck is spread out against the sky

Like a patient etherized upon a table;

Let us go, through certain half-deserted games,

The muttering remains

Of restless nights in one-night Midwest roadtrips

And cracker-jacks and peanut-shells:

Seasons that follow like a tedious argument

Of insidious intent

To lead you to an overwhelming question….

"How far away are the Mets?"

I remember making a bold statement to a friend when the Mets traded for d'Arnaud, something along the lines of "Premium offense from premium defensive positions--that's how good teams win." I gave the 2006 Mets as an example, and a few other anecdotal examples. Not content just to say that, I wanted to test it. I compiled information on every playoff team between 2006-2012 to see how right I was. Turned out, there are tons of ways to make a playoff team. You can see the data I compiled here if you want. There were only a few claims I could make that about how to build a playoff team had any sort of statistical significance:

  • You probably need an AVERAGE of 3 WAR PER STARTER in order to feel confident about a playoff spot (in other words, 8 Byrds and 5 Nieses, or 4 Wrights and 2 Harveys + a bunch of replacement level dreck should = a playoff spot)
  • On average, teams got nearly 60% of their WAR from position players, and 40% from pitchers (the most pitching-heavy playoff team was the 2007 Dbacks, who got 62% of their WAR from pitching, while the most hitting-heavy, the 2011 Cardinals, got 80% of their WAR from hitting, though only 71% from position players in total, cause their defense sucked).
  • 25 out of 58 (40%) of playoff teams had below average offense by wRC+
  • The actual position breakdown of where playoff teams got their WAR is pretty even, as you can see:
POS WAR %
C 161 12 49
SS 160.2 11
2B 178.9 13
CF 185 13
3B 195.4 14 51
RF 157.7 11
LF 170.6 12
1B 160.9 12
DH 27 2

  • The average WAR for a playoff team is around 43.
  • The Yankees have 6 of the top 7 offensive seasons among playoff teams since 2006 by wRC+. Sigh.

However, I wasn't able to find some imbalance that all playoff teams exploited, like I thought I would find with up the middle players. I am able to make some conclusions about how NOT to build a playoff team, and draw some conclusions by comparing these teams to the three Mets teams under Sandy Alderson (FYI, all 2013 rate stats are extrapolated for a 162 game season):

  1. You cannot play terrible defense: Only the 2007 Yankees, who had the highest wRC+ of all playoff teams since 2006, gave up more runs on defense than the 2011 Mets, only two others gave up more runs on defense than the 2012 Mets ('06, '08 Dodgers).
  2. You cannot have a terrible bullpen: The three Mets teams since Sandy took over have had worse bullpens by ERA- than any playoff team (105 in '13, 117 in '11, and 123 in '12; the worse playoff team was at 103). They are also 3 out of the bottom 6 in WAR (although somehow their FIP has been middle of the pack).
  3. Don't suck (plus the exceptions that prove the rule): The 2013 Mets made major strides over the 2011 and 2012 vintages, and still only 3 teams with fewer WAR than 2013 Mets have made the playoffs ('07 Dbacks, '06 Cards, '12 Orioles).

To expand on that last point, the 2011 and 2012 Mets were both almost 3 standard deviations away from a playoff team in total WAR. That is a recipe for staying home in October. The 2013 Mets have made a .8 standard deviation improvement from the 2011-12 Mets. It is entirely due to pitching, baserunning, and defense. They are still 2 standard deviations away from being a playoff team. To put it another way, the 2011 and 2012 Mets had no chance of making the playoffs, and only 5% of the playoff teams in the last six years have had as little talent as the 2013 Mets.

YEAR TEAM TOT WAR %hWAR %pWAR %dWAR %brWAR ppWAR spWAR rpWAR pWAR Avg pp Avg sp ISO BABIP wRC+ Fld BsR spFIP rpFIP spERA- rpERA-
PLAYOFF AVERAGE 43.1 55 42 2 1 25.2 14.0 3.9 17.9 3.2 2.8 0.162 0.302 102 10.8 3.0 4.14 3.94 96 86
95th Percentile 55.6 74 61 17 6 36.6 21.1 8.0 26.7 4.6 4.2 0.193 0.325 116 74.4 23.8 4.89 4.64 109 105
5th Percentile 30.7 37 22 -13 -4 13.8 6.9 -0.1 9.1 1.7 1.4 0.131 0.278 88 -52.8 -17.8 3.39 3.23 82 67
2013 METS 30.8 55 41 -2 6 18.3 12.4 0.1 12.5 2.3 2.5 0.137 0.292 92 -6.4 19.2 3.53 4.04 102 105
Zscore -2.0 -1.2 -0.4 -1.9 -1.2 -1.2 -0.4 -1.6 -0.8 -1.5 -0.5 1.6 1.6 -0.3 -0.9 -2.0
2012 METS 26.3 69 46 -14 -1 14.2 12.6 -0.5 12.1 1.8 2.5 0.137 0.298 93 -36.9 -3 3.83 4.13 101 123
Zscore -2.7 -1.9 -0.4 -2.2 -1.3 -1.9 -0.4 -1.6 -0.3 -1.3 -1.5 -0.6 0.8 -0.5 -0.8 -3.9
2011 METS 26.7 94 29 -26 3 18.9 7.4 0.4 7.8 2.4 1.5 0.127 0.307 102 -70.6 7.5 4.07 3.86 111 117
Zscore -2.6 -1.1 -1.8 -1.7 -2.3 -1.1 -1.8 -2.2 0.5 -0.1 -2.6 0.4 0.2 0.2 -2.3 -3.3

The 2011 and 2012 Mets were not significantly better than the average playoff team in ANY category. Not one. The 2013 Mets are at least a standard deviation above an average playoff team in baserunning and Starting Pitcher FIP. That's it! When people say the team needs talent, not any one position or attribute, this is what they're talking about.

Conclusions

  • Where the wins come from seems not to matter too much, though most teams rely a bit more heavily on position players.
  • The 2013 Mets actually have made good strides. In a year where the emphasis was building for the future, the team managed to get almost an entire standard deviation closer to being a playoff team in total WAR. Also, which I believe is equally as important, they are no longer below the 5th percentile in any one category.
  • While the specific ways they improve from here on out seems to be less relevant than just scoring more runs and giving up fewer, they seem to have put behind them the black hole of defense and bullpen that was 2011 and 2012, when their bullpen and defense were so bad that almost no team in the last six years with similar numbers had overcome them to make the playoffs.
  • The magic number seems to be 13 WAR added to the 2013 Mets to have a realistic shot at making the playoffs. Mark that for your AAOPs.

***************************************************************************
EDIT: Thanks to some patient but critical stats advice from Robotoverlord and others, I have some updates. First off: the key problem with the maths:
all your data is relative to playoff teams — it doesn’t suggest anything other than the fact that the Mets are 13 wins worse than the average playoff team — not that 13 additional wins would bring the Mets into playoff contention an average amount of times.

Basically, because the '10-'12 Mets weren't playoff teams, it doesn't make sense to take a z-score calculated using a standard deviation taken from a data set including only playoff teams, and then imply that that is the deviation separating them from being a playoff team.
I didn't follow Robotoverlord's suggestion and figure out how many additional wins would bring the Mets into playoff contention an average amount of times, because that was 33 WAR, and that meant the 2013 Mets were only 3 WAR away from a 50% chance at making the playoffs, which may or may not be a statistically accurate statement, but didn't feel right. Plus, who wants a 50% chance at MAKING the playoffs? That's leaving too much to chance. I would say realistically the Mets can call themselves in playoff contention when their WAR total gives them a 75% chance at making the playoffs.
The new data set is thus:
AVERAGE 33.3 59 43 -2 0 19.0 11.5 2.8 14.3 2.4 2.3 0.152 0.299 96 -0.5 0.0 4.32 4.10 104 94
STDEV 9.8 16.4 13.7 11.9 3.1 7.2 4.4 2.3 5.6 0.9 0.9 0.0 0.0 8.0 32.2 9.3 0.4 0.4 11.2 12.6
75% Post 45.3 55 41 3 1 26.9 14.4 4.0 18.4 3.4 2.9 0.164 0.303 105 16.1 2.4 4.11 3.93 96 87
2013 X METS 30.8 55 41 -2 6 18.3 12.4 0.1 12.5 2.3 2.5 0.137 0.292 92 -6.4 19.2 3.53 4.04 102 105
Zscore -1.5 -1.2 -0.5 -1.7 -1.0 -1.2 -0.5 -1.4 -1.0 -1.6 -0.7 1.8 1.3 -0.3 -0.6 -1.4
2012 X METS 26.3 69 46 -14 -1 14.2 12.6 -0.5 12.1 1.8 2.5 0.137 0.298 93 -36.9 -3 3.83 4.13 101 123
Zscore -1.9 -1.8 -0.4 -1.9 -1.1 -1.8 -0.4 -1.4 -0.5 -1.4 -1.6 -0.6 0.6 -0.5 -0.5 -2.8
2011 X METS 26.7 94 29 -26 3 18.9 7.4 0.4 7.8 2.4 1.5 0.127 0.307 102 -70.6 7.5 4.07 3.86 111 117
Zscore -1.9 -1.1 -1.6 -1.5 -1.9 -1.1 -1.6 -2.0 0.3 -0.3 -2.7 0.5 0.1 0.2 -1.4 -2.4

This is good news and bad news for my understanding of the Mets chances next season. The bad news is that the WAR they need to reach is more realistically 45, not 43 as I had previously calculated. The good news is that the standard deviation is smaller, as is the z-score separating the 2013 Mets and a playoff-contending team. It seems that my previous calculations underestimated how many wins the Mets would need to add, but also overestimated how difficult of a jump that would be.
Other fun new conclusions:
  • The 2011 Red Sox were the third best team by WAR of the last six years, and did not make the playoffs.
  • The 2009 Mariners added 91 runs on defense!
  • The 2011 Mets were the second worst defensive team of the last six years. The Dude abides.


This FanPost was contributed by a member of the community and was not subject to any vetting or approval process.