The Mets are only two games out of a National League Wild Card spot, and they possess still-have-a-shot 24.2% odds of getting into at least one October game. It's not exactly the campaign they wanted, but they're still playing meaningful September baseball, with a chance for more. Of course, it wouldn't be a Mets season if it didn't come with a little controversy, and this time around, it's Noah Syndergaard's apparent preference to pitch to backup catchers Tomás Nido or René Rivera rather than starter Wilson Ramos.

This is about to come to a head again, since the Mets started Nido on Thursday afternoon against the D-backs and are almost certainly going to pair Syndergaard and Ramos on Friday evening against Clayton Kershaw and the Dodgers.

It's not hard to see Syndergaard's case. He has a 5.09 ERA throwing to Ramos and a 2.22 ERA throwing to Nido or Rivera, and a productive Syndergaard would be a boon to the team's postseason hopes. Just last year, Jacob deGrom was allowed to throw nearly exclusively to Devin Mesoraco down the stretch on his way to winning the NL Cy Young.

It's also not hard to see a reason for the Mets to hesitate, of course. Nido and Rivera have a combined .561 OPS this year and .639 over the last three years, while Ramos is hitting .295/.357/.430 (.787 OPS) this season, making him one of the 10 best semi-regular hitting catchers in baseball. Since Aug. 1, his red-hot .389/.424/.550 line makes him one of the 20 best hitters at any position. That's a valuable bat to have in the lineup.

“I’ve told [Syndergaard] that there are going to be lineups where different catchers are going to catch, and he knows that he’s got to go out there and compete when the lineup is posted,” manager Mickey Callaway told MLB.com's Anthony DiComo.

There's probably reason enough for this conversation to begin and end with "whatever makes your talented-but-inconsistent No. 2 starter" comfortable, but that's not satisfying. After all, Syndergaard's best start of the season (by Game Score) came on May 2, when he struck out 10 in a complete-game 1-0 win over the Reds -- with Ramos behind the plate.

And while that huge ERA gap between the backstops probably tells you something, there's a reason we don't use catcher ERA that often. It doesn't tell you if the Ramos games were against tougher teams or in tougher parks or happened to come when the Mets defense or bullpen or Syndergaard himself were at their worst. It doesn't tell you why, or what specifically is better or worse, at least in terms of what can be quantified outside of personal comfort.

Fortunately, we can do a little better than just ERA. We can go deeper and see what aspects of Syndergaard's game suffers -- or in some surprising cases, doesn't -- with Ramos behind the plate. (For simplicity's sake, we're ignoring the two April starts Syndergaard threw to long-departed former Met Travis d'Arnaud, allowing 11 runs in a pair of losses, as well as the lone Rivera game.)

Syndergaard has indeed been better with Nido ... but not bad with Ramos

The ERA numbers -- 5.09 to Ramos and 2.45 to Nido -- stand out, and without context would be read as "much worse than average" and "much better than average." Of course, we know ERA can be influenced by park, defense, and bullpen, and one bad game, like when Syndergaard and Ramos allowed nine earned runs in three innings to the Cubs on Aug. 28, can really skew things.

So let's start by looking just at Weighted On-Base Average, which is basically "on-base percentage if it gave more credit to extra base hits than singles," as raw OBP does.

Ramos -- .308 wOBA
Nido -- .264 wOBA

The NL starting pitcher average wOBA allowed is .321. So what this is saying is yes, Syndergaard/Nido have absolutely been better than Syndergaard/Ramos. But despite that 5.09 ERA, Syndergaard and Ramos have still been good. They've been above average. So how is the ERA so high?

The Mets bullpen has completely failed Syndergaard in Ramos games

Twice, Syndergaard has been pulled in the middle of an inning with Nido catching, and both times, things ended tidily. On June 30 against the Braves and July 24 against the Padres, the New York bullpen prevented any further damage.

Four times, Syndergaard has been pulled in the middle of an inning with Ramos catching. Please remove small children from the room. This is about to get unsafe for work.

April 10 (Twins) -- Jeurys Familia lets three reach, adding one earned run
May 24 (Tigers) -- Tyler Bashlor allows a sac fly, adding one earned run
June 4 (Giants) -- Seth Lugo allows two hits, adding one earned run
June 15 (Cardinals) -- Robert Gsellman lets two reach, adding two earned runs

That June 4 game, for what it's worth, is the one where Callaway said he regretted pulling Syndergaard when he did.

If we were to remove those five extra earned runs from Syndergaard's record, that drops his ERA with Ramos down to 4.60, which still isn't great, but is certainly better than 5.09 -- half a run better, in fact.

The Mets' infield defense has been worse with Ramos catching

If you were to just look at errors, you'd see that the Mets have made three behind Syndergaard with Ramos catching, and just one with Nido catching. Case closed, right? Not exactly. In addition to all we know about the flaws of errors, one of the three with Ramos was made by Syndergaard himself, and the other two came in games the Mets were already losing.

Instead, we can turn to Statcast data. On each batted ball, based on the exit velocity and launch angle, a value can be placed on it to express how often that ball turns into a hit (or not). So even though this Aug. 28 ball from Kris Bryant went down as "a double" because J.D. Davis and Amed Rosario failed to communicate, that's a ball that had a mere .160 expected batting average.

So, if we were to do this only on ground balls, to try to isolate the infield, we'd find this.

Ramos -- .239 expected average, .275 actual average (36-point deficit)
Nido -- .197 expected average, .202 actual average (5-point deficit)

With Nido, Syndergaard has more or less received what he's earned. With Ramos, the deficit is seven times as large. There's a lot of "plays not made" on weakly hit balls that didn't go down as errors -- and thus don't prevent an earned run -- like on this Mitch Garver dinker from April. That has nothing to do with Ramos specifically, of course. It's just a thing that's happening.

Syndergaard gets more strikeouts and fewer walks with Ramos than Nido

This isn't the most important thing, but it's a surprising thing, so let's surface it.

Ramos -- 24.3% strikeout rate
Nido -- 23.9% strikeout rate

Also, check out base on balls.

Ramos -- 5.2% walk rate
Nido -- 7.2% walk rate

These aren't large differences. For strikeout rate, it's almost something of a tie. But you wouldn't have expected that, would you?

Part of this may be explained by the fact that with Ramos, Syndergaard is throwing 73% of his pitches into the heart and edges of the zone, while with Nido, it's 71.6%. That may not be a good thing, as we'll get to. But there's more layers to this than you'd think.

Syndergaard has faced tougher offenses with Ramos ... probably

Let's preface how imperfect this view is. We're not accounting for home vs. road, or who was in the lineup that day, or that some teams were faced multiple times. A deeper look might turn this on its head. But at the highest level, Nido has caught Syndergaard 10 times, and nine of them came against offenses considered below-average this year. (That would be the Marlins twice, and Rockies, Phillies, Giants, Padres, White Sox, Pirates, and Royals once apiece. Only the Braves stand out.)

Meanwhile, of the 15 Ramos games, nine came against below-average offenses, and six came against average-or-better offenses, including four against the Nationals, plus the Dodgers, Twins, and Cubs. Again, this is not an in-depth look, but if you were so inclined, you would be able to look at the seasonal offense of the combined opponents and see something of a difference.

Nido -- .251/.316/.417 (.733 OPS) (89 wRC+)
Ramos -- .251/.322/.435 (.757 OPS) (95 wRC+)

This might not be a coincidence, if you consider the idea that better offenses may require the Mets to pair up better hitters in response.

We've just gone through and explained a whole bunch of reasons why Ramos being paired with Syndergaard isn't quite as bad as the ERA difference would make it look. (Again: That May 2 shutout of the Reds with Ramos wasn't just his best game of the year, it was the second-best of his career.) But, as we started out with: Syndergaard has indeed been better with Nido ... but not bad with Ramos.

Nido almost certainly is a better defensive catcher than Ramos. Nido almost certainly is a better defensive pairing for Syndergaard, even if Ramos isn't quite the problem that 5.09 ERA might have you believe. As we look at the things that Nido and Syndergaard pair up better on, we might just get to the root of the entire issue.

Syndergaard allows far less hard-hit contact with Nido

A "hard-hit" ball is one that has 95 MPH of exit velocity or more.

Ramos -- 34.8% of batted balls are hard hit
Nido -- 26.7%

Syndergaard allows a .526 average and a 1.072 slugging percentage on hard-hit balls, so the fewer the better. Now, the Major League average is 36.7%, so again, this isn't bad for Ramos, it's just better for Nido.

That's a good start, but let's do better.

Syndergaard is way better when ahead in the count with Nido

It's also a good thing to get ahead in the count, and Syndergaard does that more often (34.2%) with Ramos than he does with Nido (30.8%). But when he is ahead in the count, look what happens.

Ramos -- .234 wOBA
Nido -- .153 wOBA

The Major league average here is .226, so again, it's average-ish with Ramos, yet outstanding with Nido. Why?

Syndergaard throws different pitches ahead in the count with Nido

Overall, there's no meaningful difference in what types of pitches Syndergaard throws with either catcher. But when he's ahead in the count, look at this:

Ramos

• Fastballs -- 57%
• Change -- 19%
• Slider -- 13%
• Curve -- 10%

Nido

• Fastballs -- 49%
• Change -- 16%
• Slider -- 27%
• Curve -- 8%

Now we're onto something. When ahead in the count, Syndergaard throws twice as many sliders to Nido as he does to Ramos. He throws more of every other pitch type to Ramos, especially fastballs.

That seems like a problem for Syndergaard:

• Fastballs -- .340 wOBA
• Slider -- .234 wOBA

Now, why would a pitcher be hesitant to throw a good slider to put a batter away?

Ramos is weak at framing low pitches

At the top of the zone, Ramos is actually very good at framing pitches. Looking at the Statcast pitch framing leaderboards, right above the strike zone (Zone 12), you'll see that Ramos turns 58.1% of non-swings on the top edge into strikes -- third only behind Sandy Leon and Yadier Molina. That's great.

The problem is at the bottom of the zone (Zone 18), where there's a big gap between the two.

Ramos -- 32.5% of non-swings become strikes
Nido -- 47.2% of non-swings become strikes

For Nido, that's just slightly below the league average. For Ramos, that's the weakest in baseball; the next-weakest catcher there, Chris Iannetta, was released a month ago. Overall, Ramos is minus-3 runs in framing, and Nido is plus-3 runs.

Perhaps you can see where this is going. Ramos is weak framing low pitches, and Syndergaard either won't throw his slider as much when he's ahead, or he throws it nearer the zone -- to Nido, it comes in at 2.06 feet off the ground, while to Ramos, it's 2.22 feet. The outcomes are worse: Sliders to Ramos have a .463 slugging percentage. Sliders to Nido have a .292 slugging percentage.

"There's a certain it factor, there's a relationship -- a symbiotic relationship -- that [pitchers and catchers] can possess, and it's all about being comfortable out there," Syndergaard told the AP.

That's absolutely right. Ultimately, you'd probably want to defer to what brings out the best version of one of your most important pitchers, and there's little doubt that Nido brings out better performances. Then again, the offensive differences here are pretty enormous, too, and the defensive differences aren't that big -- or at least not as large as the large ERA gaps would have you believe.

Put another way: Nido is the better bet behind the plate. Ramos is the better bet next to it. It probably makes more sense to side with the pitcher, here. It's just not as clear-cut as you'd think.