Mets' September screw-up

Dillon Gee certainly turned heads with his outing Tuesday and hopefully the Mets will give him regular starts in September. With the playoffs out of reach, the club needs to use the final month of the season to give players a chance to show what they can do. Which is why Bobby Parnell should be the closer and Hisanori Takahashi should be in the rotation.

Instead, Parnell is being used as a righty setup guy who manager Jerry Manuel tries to limit his exposure to LHB and Takahashi is employed as the closer. This alignment may ensure an additional victory or two this season, but the difference between 82 or 83 wins is negligible. It makes no sense because Parnell learns very little not facing lefties and there is virtually no chance that Takahashi closes for the Mets in 2011.

Takahashi is a free agent at the end of the season. Is it worthwhile to sign him for 2011? Is he just a bullpen guy in the majors or can he be a reliable starter? Conventional wisdom says that Takahashi is better suited to be a reliever because his numbers are so much worse the 2nd and 3rd time through the order. But the 35-year old also enjoyed some success as a starter. Was that success merely a fluke because teams did not have a good scouting report on him? A half dozen more September starts could have helped provide a better answer.

Here are Takahashi’s OPS by times through the order (numbers through Tuesday)

1st PA – .598
2nd PA – .919
3rd PA – .990

But can we take those numbers at face value? Are there any mitigating circumstances that we should factor in before using them to draw any conclusions? How do those numbers stack up to other starters on the staff? Let’s examine that last question first.

1^st PA

	PA	AB	AVG	OBP	SLG	OPS
Johan Santana	279	246	.252	.330	.402	.732
Jonathon Niese	248	223	.238	.300	.323	.623
R.A. Dickey	205	186	.237	.294	.312	.606
Mike Pelfrey	249	226	.305	.364	.428	.792
Hisanori Takahashi	108	95	.232	.288	.316	.604

2^nd PA

	PA	AB	AVG	OBP	SLG	OPS
Johan Santana	261	236	.242	.289	.314	.603
Jonathon Niese	221	200	.265	.324	.460	.784
R.A. Dickey	188	173	.289	.326	.393	.719
Mike Pelfrey	241	210	.319	.385	.481	.866
Hisanori Takahashi	106	94	.293	.353	.565	.918

3^rd PA

	PA	AB	AVG	OBP	SLG	OPS
Johan Santana	217	207	.217	.247	.329	.576
Jonathon Niese	186	162	.333	.405	.549	.954
R.A. Dickey	148	140	.200	.243	.271	.514
Mike Pelfrey	189	174	.276	.328	.391	.719
Hisanori Takahashi	66	61	.371	.409	.581	.990

We see immediately that Takahashi has the worst numbers the second and third time through the order but we also see that Pelfrey struggles in the second PA and Niese in the third PA. The fascinating thing to me about these numbers are how Santana’s OPS declines the second and third time through the order and how Dickey is nearly unhittable if hitters face him a third time.

So, since Takahashi’s numbers are so much worse than the other Mets’ starters, doesn’t this justify thinking he’s best suited for relieving? In a word, no.

First, Takahashi has between one half to one third of the PA that the other pitchers have. This means his numbers are much more susceptible to a small sample size fluke. Second, Takahashi did not get to pitch on a regular five-day rotation. Twice he was moved to the bullpen and once he went 13 days between starts.

Finally, Takahashi pitched a game in Puerto Rico in a stadium that is really not up to Major League standards. The Mets and Marlins combined for 37 runs in the three-game series. The three games before that, the Mets and Twins combined for 19 runs and the three games after Puerto Rico the Mets and Nationals combined for 22 runs.

If we remove the start in Puerto Rico, Takahashi’s numbers the second time through the order become:

.256/.305/.465 for a .770 OPS

This would be the third-best mark on the staff, behind Santana and Dickey but ahead of Niese and Pelfrey. Just that one start is really skewing the numbers. How about if we also remove the game when he went 13 days between starts? Then his numbers the second time through the order look like this:

.235/.253/.370 for a .623 OPS

Of course, anyone’s numbers would look better if you removed the two worst starts from their overall record. But Takahashi is suffering from a double whammy here. First, he does not have enough starts to compensate for these bad outings and second, both of his bad outings have a logical explanation (Hiram Bithorn, 13 days between starts).

I don’t think a .623 OPS the second time through the order is indicative of Takahashi, just like I do not think a .918 OPS is what his true talent level is, either. The true number is most likely somewhere in between. But we have no idea where. And that’s why he should be starting right now – to give us more information about what kind of pitcher he really is.

The bottom line is that no one can say if Takahashi has the ability to be an effective Major League starter. He simply does not have enough innings for anyone to make an informed opinion. Those who cite the numbers for the second and third time through the order are making judgments on small samples that they would not do for other pitchers.

In 12 games as a starter, Takahashi is 4-4 with a 5.01 ERA. In his last 12 games, Pelfrey is 3-7 with a 5.63 ERA. There is just as much evidence that Pelfrey cannot succeed as a starter as there is for Takahashi, more if you count what Pelfrey did in 2009.

This is yet another indication of why I believe that Jerry Manuel is not a good manager. Instead of doing what is best for the players and the long-term health of the franchise, Manuel is trying to squeeze every last win out of a non-playoff club. Because of this, we have no idea if Takahashi can start and if Parnell can handle pitching the ninth inning.