Bartolo Colon, Robert Gsellman, Seth Lugo and the Mets’ fifth starter options

The Mets hope to have their four young aces healthy for Opening Day and the entire 2017 season. Since all are expected healthy for Spring Training, let’s grant that at least the first part of the first sentence turns out to be true. Then the question becomes: Who’s the team’s fifth starter? Let’s compare the three leading contenders.

Bartolo Colon:
Pros – Has been extremely durable in his three seasons with the club, is a fan favorite, provides great intangibles
Cons – Age, salary

Robert Gsellman
Pros – Great combination of ground balls and strikeouts
Cons – Unsustainable HR rate, only 93.1 IP above the Double-A level

Seth Lugo
Pros – Mets as a team were 7-1 in games he started
Cons – Peripherals paint an entirely different picture than his actual results

Fans of each pitcher could likely put a few more in the “pros” column but as an overview of the candidates, these seem fair. Now let’s compare their actual performance in the majors in 2016. Obviously Colon will have a huge edge in counting stats, as he was with the team the entire season. He absolutely deserves credit for staying healthy all year. But we need to grant him that edge without allowing that to overshadow everything else that went on.

Player	IP	K/9	BB/9	HR/9	BABIP	LOB%	GB%	HR/FB	ERA	FIP	xFIP
Colon	191.2	6.01	1.5	1.13	0.291	76.50%	43.20%	11.50%	3.43	3.99	4.17
Gsellman	44.2	8.46	3.02	0.2	0.325	81.30%	54.20%	3.60%	2.42	2.63	3.38
Lugo	64	6.33	2.95	0.98	0.23	85.70%	42.80%	10.10%	2.67	4.33	4.71

IP – Huge edge for Colon
K/9 – Strong edge for Gsellman
BB/9 – Strong edge for Colon
HR/9 – Huge edge for Gsellman, with the understanding that this is not repeatable
BABIP – Huge edge for Lugo, with the understanding that this is not repeatable
LOB% – Edge for Lugo, although likely not repeatable
GB% – Strong edge for Gsellman
HR/FB – Huge edge for Gsellman, with the understanding that this is not repeatable
ERA – Edge for Gsellman
FIP – Strong edge for Gsellman
xFIP – Strong edge for Gsellman

At best this is a difficult comparison because Colon has performed at the highest level for 19 years while the other two made their major league debuts in 2016. If your primary tool for evaluation is experience, Colon wins this competition hands down.

Yet that has to be tempered with the fact that only 12 pitchers in MLB history have thrown 100 innings at age 44, which is where Colon will be in 2017. Five of those were knuckleball pitchers, four of them crafty lefties and the remaining three are Randy Johnson, Nolan Ryan and Cy Young. Only four of the 12 had an ERA+ of 100 or more.

Colon has defied the odds so far. But to exaggerate to make a point, betting on a pitcher this old to succeed is like playing Russian roulette. At some point the luck runs out. Only seven pitchers in MLB history have thrown 100 or more innings at age 45, four at age 46, two at age 47 and one at age 48. No one pitches forever and the wheels come off without much warning.

Much like the major prerequisite to living to age 100 is reaching age 99, Colon has passed the major requirement for pitching well at age 44 by throwing 191.2 innings at age 43. But it calls to mind the old Sports Illustrated story on tennis’ Bobby Riggs, that ran with the headline, “Never bet against this man,” after he beat Margaret Court. Of course, he went on to lose against Billie Jean King, which if memory serves, was his very next match.*

But even if you’re convinced that Colon has another year similar to 2016 left in his arm – how much are you willing to pay to find out for sure? If it’s a gamble that Colon can reproduce his 2016 season because of his age and a gamble that Gsellman/Lugo can repeat their 2016 season because of inexperience – do you want to gamble at $10 million or at 1/20 that price?

A lot of what’s causing people hesitancy about fully embracing either Gsellman or Lugo is how they did not come with the big reputations that a Matt Harvey or Noah Syndergaard did. But we don’t have to go too far back to see a guy not as heralded come up and make a big impact.

While Jacob deGrom came with more expectations than either Gsellman or Lugo, it would be revisionist history to claim that he was thought of as a future All-Star when he debuted back in 2014. Let’s run the same chart as before, this time substituting 2014 deGrom for Colon.

Player	IP	K/9	BB/9	HR/9	BABIP	LOB%	GB%	HR/FB	ERA	FIP	xFIP
deGrom	140.1	9.24	2.76	0.45	0.297	77.40%	45.40%	6.10%	2.69	2.67	3.03
Gsellman	44.2	8.46	3.02	0.2	0.325	81.30%	54.20%	3.60%	2.42	2.63	3.38
Lugo	64	6.33	2.95	0.98	0.23	85.70%	42.80%	10.10%	2.67	4.33	4.71

The FIP and xFIP numbers of deGrom and Gsellman are similar, although deGrom still had a huge edge in innings pitched. This is not to suggest that Gsellman is a deGrom waiting to happen. It’s just that we shouldn’t be shocked if Gsellman is able to come up and be successful without a big pedigree because we saw the very same thing happen just a few short years ago.

Finally, let’s add a few more recent homegrown Mets pitchers into our comparison. These are 2007 Mike Pelfrey, 2009 Jon Niese and 2010 Dillon Gee. Both Pelfrey and Niese had been up for very brief appearances the season before, so they weren’t in exactly the same situation. Also, both of those pitchers were more highly regarded than either Gsellman or Lugo, with Pelfrey being the ninth overall pick of the 2005 Draft. Gee is actually a much better comp but the others will still be good for comparison sake.

Player	IP	K/9	BB/9	HR/9	BABIP	LOB%	GB%	HR/FB	ERA	FIP	xFIP
deGrom	140.1	9.24	2.76	0.45	0.297	77.40%	45.40%	6.10%	2.69	2.67	3.03
Gee	33	4.64	4.09	0.55	0.225	80.70%	47.00%	4.70%	2.18	4.2	5
Gsellman	44.2	8.46	3.02	0.2	0.325	81.30%	54.20%	3.60%	2.42	2.63	3.38
Lugo	64	6.33	2.95	0.98	0.23	85.70%	42.80%	10.10%	2.67	4.33	4.71
Niese	25.2	6.31	3.16	0.35	0.317	69.40%	48.10%	3.70%	4.21	3.25	4.13
Pelfrey	72.2	5.57	4.83	0.74	0.325	69.00%	48.30%	8.80%	5.57	5.06	5.16

Just like the FIP and xFIP are similar for deGrom and Gsellman, those two metrics show Gee and Lugo as good matches. Recall that Gee has made 124 starts and counting in the majors. If Lugo could match Gee’s career, that would be an outstanding result.

Of course, there will be a lot more to determine if Lugo can repeat Gee’s career than their initial season in the majors FIP and xFIP. But the Mets have two guys who compare favorably (or better) to Gee at their disposal when choosing their fifth starter for 2016.

The 2011 Mets went dumpster diving, bringing in injury reclamation projects Chris Capuano and Chris Young rather than giving Gee a shot at the beginning of the season. Young couldn’t stay healthy, so Gee ended up making 27 starts in 2011, with a 4.43 ERA, which slotted between his FIP (4.20) and xFIP (5.00) marks of 2010.

The ERA estimators of FIP and xFIP did a fine job of predicting Gee’s following season, despite working with a brief 33-inning sample. Pelfrey did much better in 2008 than either of his previous season estimators, with a 3.72 ERA. Niese did worse, although xFIP (4.13) was almost a perfect predictor of his following season’s 4.20 ERA.

Of course three pitchers is hardly anything upon which to base an important decision. No doubt you could find three pitchers who did much worse than their initial ERA estimators. All this piece is suggesting is that it’s not ridiculous to consider giving the fifth starter’s job to guys we saw perform well down the stretch in 2016.

* – Just a gambling story so no need to mention Riggs’ alleged ties to mobsters and rumors of how he threw the match.