The Mets Were a Bad Defensive Club

The 2015 season will forever go down as (literally) a banner year for the New York Metropolitans. A National League championship pennant will forever fly above Citi Field, and with youthful pitching studs Jacob deGrom, Matt Harvey, Noah Syndergaard and Zack Wheeler in place for the foreseeable future, a competitive future appears assured. As someone who has spent many years in a major league front office, I can assure you, however, that the page is turned from the present to the future at record speed in the game of baseball. The Mets, and all 29 other clubs, must quickly take stock of who they are and put the wheels in motion toward the club they which to become moving forward.

There is no such thing as a perfect ball club in this day and age. The champion Royals themselves lack offensive power, and their starting rotation was far from fearsome. One can easily make the argument that the Mets owned the starting pitching advantage in each and every World Series contest. Obviously, the Royals’ superior team defense, bullpen, position player durability and ability to make contact more than offset their shortcomings.

The Mets are a very curious case. On the last day of July, they stood 30th and last in the major leagues in runs scored. It’s pretty unique to travel from that low point to playing meaningful games in November. Their other chief weakness, which just happened to be the one that reared its ugly head when the stakes were the highest, was the Mets’ subpar team defense.

While no one would have placed the Mets’ overall defensive ability on par with that of their World Series opponents, both old-fashioned and new-age metrics agree that the National League champions profiled (at worst) as an average defensive club. You want old school? The Mets ranked fifth in the NL in fielding percentage, and made the fifth-fewest errors. You want a little more new-school? They ranked second in the NL in defensive efficiency, i.e., turning batted balls into outs. So far, they look positively above average.

Let’s get a little more cutting-edge, and move on to FanGraphs’ team defense page. Here we find the Mets 17th in overall defense, 13th in UZR/150, 19th in error runs, and 10th in range runs. Now we’re talking straight up major league average.

What if we were to take a look at the issue from a somewhat different perspective, both incorporating batted ball data into the equation, and evaluating team defense from a head-to-head perspective; how did clubs perform defensively as compared to their opponents over 162 games, in the same ball parks, with the same weather conditions?

First, let’s take a step back and take a look at the broad batted-ball picture, to develop some perspective and context for our analysis. I utilized a sample of over 40% of all batted balls in the 2015 season; basically, I downloaded all balls in play (BIP) allowed by pitchers who qualified for the ERA title as of either the All Star break or the end of the season. Overall, it’s a fairly representative sample of the population, though this group of pitchers managed contact a little better (.320 AVG-.505 SLG) than the whole (.323 AVG-.513 SLG).

MLB Production by BIP Type – 2015

	ALL AVG	ALL SLG	IN-PL AVG	IN-PL SLG
NULL	0.239	0.356	0.221	0.271
POP	0.037	0.049	0.037	0.049
FLY	0.183	0.546	0.092	0.159
LD	0.640	1.036	0.615	0.832
GB	0.257	0.281	0.257	0.281
ALL	0.320	0.505	0.294	0.372
———–	———–	———–	———–	———–
FLY + LD	0.445	0.827	0.396	0.551

A couple of notes about the above. First, as noted in some of my previous Statcast articles, 23.7% of all BIP did not register on the recording equipment; these are the null items, which include a disproportionate number of weak grounders and pop ups. Secondly, vertical angle information was not recorded on the data (downloaded from Baseball Savant), except for home runs. Therefore, BIP were categorized as pop ups, fly balls, liners or grounders subjectively (though apparently consistently), not by launch angle. This means that many batted balls classified as fly balls based on their launch angles in previous years, using the older, SportVision generated data, were classified as liners this time around. This is why I have added a combined “FLY + LD” line item above.

The first two columns list performance on all batted balls within each classification; the last two weed out the home runs, which we will obviously need to do if we are to measure team defense utilizing batted ball data. Now that we have established the above as the baseline against which all team-specific data must be measured, let’s now look at the batted balls hit and allowed by the Mets within my sample.

First, the Mets’ own production:

2015 Mets: Production by BIP Type

	ACT AVG	ACT SLG	PRJ AVG	PRJ SLG
NULL	0.203	0.323	0.203	0.323
POP	0.000	0.000	0.037	0.049
FLY	0.184	0.553	0.194	0.574
LD	0.582	0.895	0.644	1.055
GB	0.257	0.285	0.259	0.283
ALL	0.299	0.481	0.317	0.522
———–	———–	———–	———–	———–
FLY + LD	0.395	0.734	0.433	0.829

And then the production they allowed:

2015 Mets: Production Allowed by BIP Type

	ACT AVG	ACT SLG	PRJ AVG	PRJ SLG
NULL	0.227	0.328	0.227	0.328
POP	0.000	0.000	0.037	0.049
FLY	0.155	0.488	0.151	0.443
LD	0.629	0.996	0.639	1.028
GB	0.281	0.303	0.258	0.282
ALL	0.315	0.482	0.308	0.473
———–	———–	———–	———–	———–
FLY + LD	0.413	0.763	0.416	0.761

The first table above compares how Mets hitters performed on each BIP type compared to how they “should have” based on the MLB average of BIP hit at the same velocity. The second table does the same for all BIP allowed by Mets pitchers in my sample.

In the last two columns of both tables, which list the projected performance on all BIP hit and allowed by the Mets, you’ll notice that in almost all cases, the Mets outperformed their MLB brethren. Their hitters hit their fly balls, liners and grounders harder than the MLB averages listed in the first table above, while their pitchers allowed less authoritative fly balls and liners than the MLB average. Only the grounders allowed by their pitchers were hit harder — and ever so slightly — than the MLB average.

Next, pan your eyes left to the first two columns. You’ll note that despite the advantages enjoyed by Mets hitters and pitchers in terms of authority, that their opponents actually out-produced them overall, and specifically on most individual BIP types. Now, homers are included in these tables, and must be removed if we are to pass judgment on the respective team defenses of the Mets and their opponents.

Here are the defensive numbers for the Mets’ opponents:

Mets In-Play BIP Production – Opponent Defense

	ACT AVG	ACT SLG	PRJ AVG	PRJ SLG	% MLB AVG	INCL OPPO	DEF MULT
NULL	0.182	0.225	0.182	0.225
POP	0.000	0.000	0.037	0.049
FLY	0.088	0.147	0.097	0.165	81.4	74.9	108.7
LD	0.564	0.762	0.621	0.847	81.9	89.5	91.5
GB	0.257	0.285	0.259	0.283	99.6	110.8	89.9
ALL	0.273	0.348	0.289	0.371	88.6	96.1	92.2
———–	———–	———–	———–	———–	———–	———–	———–
FLY + LD	0.348	0.483	0.383	0.538	81.8	87.8	93.2

And here are the numbers for the Mets’ defense itself:

Opponents In-Play BIP Production – Mets Team Defense

	ACT AVG	ACT SLG	PRJ AVG	PRJ SLG	% MLB AVG	INCL OPPO	DEF MULT
NULL	0.210	0.248	0.210	0.248
POP	0.000	0.000	0.037	0.049
FLY	0.064	0.108	0.076	0.132	69.1	74.9	92.3
LD	0.605	0.800	0.614	0.829	95.4	89.5	106.6
GB	0.281	0.303	0.258	0.282	117.7	110.8	106.2
ALL	0.290	0.355	0.286	0.357	101.4	96.1	105.5
———–	———–	———–	———–	———–	———–	———–	———–
FLY + LD	0.363	0.491	0.373	0.517	92.6	87.8	105.5

The first four columns in both tables weed out the homers. The fifth column presents the actual production for either the Mets or their opposition (after multiplying by run values) and compares it to MLB average of 100. The sixth column (numbers identical on both tables) combines the actual production for both the Mets and their opponents, and compares it to MLB average of 100. The seventh and final column divides the 5th column by the 6th and derives a Defensive Multiplier for either the Mets (second table) or their opponents (first). The key overall number appears near the bottom of the second table: the Mets’ overall team Defensive Multiplier of 105.5.

The Mets were out-defended by their opponents by a significant amount on in-play fly balls and liners combined (105.5 multiplier) and on grounders (106.2). Almost completely across the board, Met hitters hit the ball harder, Met pitchers yielded less authoritative contact… but Met hitters were less productive.

The Mets ranked 25th among the 30 MLB clubs in this metric in 2015. The only teams with higher Defensive Multipliers? The Yankees (109.7), White Sox (109.0), Padres (108.9), Angels (107.3) and Braves (106.6). In looking at these six clubs, I stumbled upon the most basic of indicators of team athleticism/defense. How about triples for and against? These six clubs were cumulatively out-tripled by 189-138, with only the Angels, by 21-19, out-tripling their opponents.

Is the head-to-head Defensive Multiplier a perfect methodology? As with any major league club, or any metric, the answer is of course not. A 100% sample would make me feel more comfortable, for starters. This is more of a macro-type method — in contrast to the micro-type, fielder-specific methods publicly available here and elsewhere.

I’m a big fan of many of those methods, but they too are limited, by fielder positioning, park factors, etc. A method such as this tells you less about individual fielders, but its head-to-head nature weeds out park effects, and can give you a better feel for teams who position their fielders particularly well, or poorly.

Bottom line, the Mets were out-defended by their opponents by a significant margin in 2015. With the exception of Juan Lagares, when he was in center field, they lacked anything resembling a plus defender. Bad defense, modest offense — especially with the anticipated losses of Daniel Murphy and Yoenis Cespedes to free agency — and great starting pitching is probably not a pennant-winning formula going forward. Look for the Mets to seek aggressively athletic, two-way contributors to flesh out their lineup in 2016, in addition to the internal help, in the persons of Michael Conforto and Dilson Herrera, that is cost-effectively on its way.