Is Troy Tulowitzki a leadoff hitter?

With managers getting the gift of new players at the deadline, it's important to decide where to place them in the lineup. John Gibbons of the Toronto Blue Jays, for instance, had one-fifth of his roster over-hauled. Terry Collins' Mets made some savvy deadline moves as well, solidifying a roster in need of some help. Both men have to determine where these new pieces fit in their everyday lineups.

Fans in both New York and Toronto have been in disagreement about how to manage their new batting lineups. What's that you say? Fans disagree with the manager of their favorite team all the time? Blasphemy!

Let's start with Terry Collins. Since the Yoenis Cespedes trade, the Mets have only played three games. The first two were against right-handers, and the lineup looked like this:

Granderson

Murphy

Cespedes

Duda

Flores/Uribe

Johnson

d'Arnaud/Plawecki

Pitcher

Tejada

Last night's game was against a left-hander though and looked significantly different:

Granderson

Cespedes

Uribe

Duda

d'Arnaud

Flores

Tejada

Niese

Lagares

Pretty straightforward. Michael Conforto will probably get more starts going forward, so that means Johnson's spot should be the one in jeopardy. We should also expect Lagares to get more playing time than a traditional defensive sub, but there is something of a logjam of actually talented hitters in the Mets outfield now. If we're going to look at lineup construction then, Lagares' name should only come up against left-handers. Maybe Conforto works in a platoon-like situation with Lagares in this way.

The tenets of lineup construction are well-documented in The Book, and quickly summarized here. Most importantly, Terry Collins should be building his lineup around the #1, #4, and #2 spots. Specifically, OBP in the #1 slot, SLG or ISO in the #4 spot, and overall hitting in the #2 spot. Let's see how the Mets' players rank in those categories in their last 700 PA and the rest-of-season projections (RoS) from Steamer:

	OBP	RoS	SLG	RoS	ISO	RoS	wOBA	RoS
Granderson	.332	.323	.412	.408	.169	.178	.328	.321
Murphy	.318	.323	.395	.398	.119	.119	.308	.312
Cespedes	.317	.305	.491	.452	.202	.192	.346	.325
Duda	.345	.332	.482	.430	.235	.194	.358	.332
Flores	.277	.283	.370	.395	.127	.154	.282	.293
Johnson	.299	.294	.387	.377	.154	.152	.303	.295
d'Arnaud*	.302	.308	.402	.431	.162	.179	.308	.319
Plawecki**	.289	.280	.307	.336	.070	.105	.260	.270
Tejada	.333	.312	.322	.322	.079	.079	.292	.281
Uribe	.338	.305	.432	.396	.135	.140	.336	.304
Lagares	.294	.289	.338	.354	.078	.102	.279	.280

* Travis d'Arnaud has only 623 plate appearances on record.

** Kevin Plawecki has only 219 PA on record. Considered adding 500 league average plate appearance but that would actually help his numbers and I'm not sure this exercise would benefit from that.

Let's start off with our most important batter, the batter that is going to come up the most often; #1. Among the leaders in OBP are Lucas Duda, Juan Uribe, Ruben Tejada, and Curtis Granderson. Duda, Uribe, and Granderson finish in the top 4 in slugging on the team though, so it might be worthwhile to place Tejada there for now.

For the #4 spot, Yoenis Cespedes leads the team in slugging, yet Duda leads in isolated power. Let's slot Cespedes #4 because the better overall hitter is Lucas Duda and we want him to bat more often in the #2 spot. That leaves Juan Uribe the best hitter left to bat third.

But wait, Tejada's rest-of-season projections for on-base percentage deduct his abilities by 20 points! And Uribe gets deducted 30 points! Uribe goes from being just a tick above league average to poor. Maybe these aren't the players we want seeing the most at-bats.

Perhaps Granderson is the leadoff hitter for the Mets, since his rest-of-season OBP projections are the only ones on the team that are expected to be above average. I do want to trust Tejada's 700 previous plate appearances, but in all likelihood, he's now our #9 hitter. That puts Cespedes #4, Duda #3 and we're finally making hay. This is starting to look familiar...

Filling out the remainder of the order, we'll put the best remaining hitter, Travis d'Arnaud, at #2. I would prefer #5 to be Daniel Murphy because of his ability to get on-base. That's a sentence I'm not sure I'd ever type. Numbers 6 through 9 are fairly subjective but I'd go with Juan Uribe, Kelly Johnson, the pitcher, and then Ruben Tejada. With not many speed threats on the Mets, having a better hitter than the pitcher here is still a benefit. Hopefully his on-base-percentage continues to beat projections down there. Here's how it looks:

Granderson	RF
d'Arnaud	C
Duda	1B
Cespedes	CF
Murphy	2B
Uribe	3B
Johnson	LF
Pitcher	P
Tejada	SS

What about the Blue Jays? According to many in the fanbase, acquiring Troy Tulowitzki has created a logjam of talented right-handed hitters. And, when he batted leadoff in his debut, many fans were upset at having such a good slugging player at the top of their lineup. Which meant when Ben Revere was acquired, the old school fans in the city rejoiced to have such a speedy player to finally fill that #1 spot. But Gibbons hasn't put Revere there. He's kept Tulowitzki there. Is he making a smart move?

Following the same logic, let's break down the Blue Jays roster:

Player	OBP	RoS	SLG	RoS	ISO	RoS	wOBA	RoS
Tulowitzki	.391	.342	.536	.465	.218	.197	.399	.347
Donaldson	.362	.343	.521	.492	.235	.221	.380	.360
Bautista	.374	.373	.524	.511	.272	.251	.385	.379
Travis*	.334	.310	.448	.402	.158	.142	.338	.311
Colabello**	.313	.313	.416	.445	.163	.176	.321	.327
Martin	.372	.337	.461	.417	.181	.173	.364	.331
Encarnacion	.344	.351	.468	.495	.215	.235	.352	.364
Pillar**	.293	.301	.377	.395	.120	.129	.294	.302
Revere	.330	.320	.366	.354	.063	.067	.308	.296
Smoak	.290	.315	.383	.411	.175	.177	.299	.317

* Devon Travis has only 238 career plate appearances. In this case, 462 league average plate appearances were added to adjust his scores for regression.

** Chris Colabello and Kevin Pillar have only 658 and 651 career plate appearances respectively. No adjustments were made.

Using the same methodology, on-base percentage reigns supreme for the #1 slot. That leaves us with Troy Tulowitzki. Incredibly, John Gibbons has accurately placed the player that's the best at not getting out at the top of the lineup, almost like he gets paid to make these decisions. Admittedly, Tulowitzki's rest-of-season projections take a hit and fall behind Bautista, Encarnacion, and Donaldson, but all of them have better ISOs. Similarly, all of Tulowitzki's numbers take a hit according to rest-of-season projections. This is likely a result of Tulowitzki's red-hot 2014 campaign though.

On a different team with different players, Tulowitzki might not be the best leadoff candidate, but on this roster, he seems to be. Other candidates could be Jose Bautista, Russell Martin, and Josh Donaldson. Because Bautista's and Donaldson's ISOs are higher, we want them batting with runners on a bit more. You could make a case for Russell Martin batting leadoff, but Tulowitzki does have 30 points of wOBA on him.

The next decisions are for the #2 and #4 spots. This one is extremely difficult between Jose Bautista and Josh Donaldson. Because their wOBAs have only a 5 point difference, I would rather see the higher ISO bat 4th. That makes Donaldson second and Bautista 4th. Bautista's rest-of-season wOBA would also support this decision.

This may really rattle some cages, but Russell Martin's ability to not make outs makes him the #3 hitter while Edwin Encarnacion slots nicely into the #5 spot because of his ability to drive the ball. By rest-of-season projections, Encarnacion is expected to reach base 15 points higher than Martin. Taking into account both numbers, it does still seem that Martin might be the better option here.

Slots 6 through 9 will then be occupied by Devon Travis, Chris Colabello, Kevin Pillar, and Ben Revere. Revere slips to ninth to become the 'second leadoff hitter,'  good batter that will complement the top of the lineup. Here's how it looks:

Tulowitzki	SS
Donaldson	3B
Martin	C
Bautista	RF
Encarnacion	DH
Travis	2B
Colabello	1B
Pillar	CF
Revere	LF

And there you have it--two perfect lineups. But it's never that simple. The takeaway from this exercise is threefold.

First, lineup optimization should be discussed in every major league clubhouse. Hopefully, it is done so beside the framed poster of the RE24 table. We can dream.

Second, is it really worth telling John Gibbons to move Russell Martin up two spots while sliding his other cornerstone sluggers down a spot each? What about calling up Collins and telling him to make any of the many changes shown above? Probably not. These are choices that, through the remaining 50+ games of the season, either team may lose anywhere between 1 and 15 total runs. Lineup optimization, while important, is only one facet of a manager's responsibilities. Sure, it's one he can control  -- it may even be his most important responsibility -- but random fluctuations or unpredictable hot and cold streaks could actually generate more runs. Just ask the Minnesota Twins. Lineup optimization is built on a premise that it is to be used every single possible time out for the most controllable, positive outcome over a long enough time period.

Which brings us to the third point. We have almost completely omitted defense from the equation. Wouldn't we rather see Juan Lagares in centerfield for the Mets? How about Justin Smoak at first base for the Jays? Never mind defense, what about platoon effect? Wilmer Flores at second base instead of Kelly Johnson against left-handed pitching? Ezequiel Carrera jammed into the Jays lineup somehow against right-handed pitching?

It's a difficult balance to make, but these are the choices and sacrifices we task with our major league managers. It just really shouldn't be Earth-shattering when a manager wants his best batter to get the most at-bats.

. . .

Michael Bradburn is a Featured Writer for Beyond the Box Score. You can follow him on Twitter at @mwbii, or reach him at michaelwbii@gmail.com