Optimizing the Lineup | The Process Report

Optimizing the Lineup

It is no secret that the Rays have drastically altered their philosophical approach to offense in the last few years:

xBA/OBP/etc.. refers to indexed averages for each metric compared to that year’s league average production. During the first wave of glory the team had capable hitters that could do it all, but as that talent bled out the team had to find ways to make the engine hum without breaking the bank for the types of players that are capable of doing it all well. It started first when you saw the contact start to diminish. The power played up enough to deal with the lesser contact, and then in 2013 the team got back to doing it all at a high level.

The 2014 season saw a commitment to OBP, but it came with a tremendous fall off in power production. In theory, you would love to have guys on base more of the time, but when it comes primarily from singles and walks there are a lot of games where it can feel like if they just had a little more thump there wouldn’t be so many guys left on base. The front office seems to have made a concerted effort to not let that sort of thing happen again. In 2015 they tried to balance the lineup, and were right around league average, but still failed to translate a league average offense into league average run scoring. Then 2016 hit and they said, “No more half measures.” The team decided to go full bore on eschewing contact and walks to go get that delicious power. It didn’t work. Despite power production as good as the best offensive year they had ever assembled in 2009 the team still scored runs at a pace around 7% lower than average.

This begs the question of what you can expect in this coming year. Everyone would prefer to have players that hit for contact, take their walks and can drive the ball in the gaps and over the wall. Every fan in baseball wants that. Unfortunately, the supply for those type of players is incredibly limited leading to exorbitant costs to field a team of pure hitters. A similar concept in economics is the production of guns or butter:

Given a finite number of resources with which to work you can produce an array of goods that can focus on anywhere along the curve. You can travel anywhere you want on that curve, but the only way to produce more of both is to push the curve to the right via increases in technology. In the baseball world that means getting better, more expensive players (or growing your own). Failing that you’re left to pick your poison along the curve. We can somewhat replicate this into a Production Possibilities Curve that looks like the guns versus butter example above:

The population consists of all non-pitchers that have accrued at least 450 PA since 2014. I have plotted OBP on the x-axis, while the y-axis is reserved for SLG.The heavily white areas show that one-dimensional guys have been naturally removed from the sample long before they even get to the Show. There is no place for a guy with a .500 SLG unless he can get on base at a .300 OBP or better. The trendline is based on the interplay of these two components, and it is interesting to see a couple of kinks. Part of that is due to position. You’re always going to have guys that are so good on defense that you can deal with a subpar bat, but I believe there is more.

At the lower end you had at least be able to get on base if you can’t hit with much power. Teams will still take that. League average for both can be found where the black curve intersects with our trend, and you’ll notice that it starts to flatten a little bit until you get into the .335 area where it increasingly gets more steep. The flatter section is where you can get by being ok at both without great strengths or weakness. In the steeper area you can see where OBP becomes marginally less important as teams are willing to deal with a lesser OBP, because it comes with very real power.

Now to the curves. These are the examples of guns versus butter. The black curve is based on the average OBP/SLG for the 50th percentile wRC+. The red lines expand to include those players that were plus or minus one standard deviation by wRC+. The yellow demarcates where we are crossing into plus or minus two standard deviations, and the green line shows plus or minus three. Theoretically every guy on a curve, or reasonably within a similar band, should be considered fairly equal. You can choose the shape of that production, but only by moving to the right do you give yourself a chance at getting both.

You’ll notice that the majority of the Rays fit right in around that threshold going from the league average to one-plus curve. Other than Matt Duffy, the majority of guys are in the SLG over OBP section (above the trend). Rickie Weeks Jr. pushes closer to that double-plus section, and Evan Longoria is basically on that doorstep. Corey Dickerson is the one guy that pushes into the elite territory, though it must be said that this includes two years of production at Coors Field so I’d be fine dropping him to more firmly in that plus-two to plus-three section.

Knowing the shape of the production we can start to begin to formulate visions of ideal lineups. You probably want your better OBP guys at the top with your balanced, good hitters more toward the middle and then your worst hitters at the bottom. An enormous facet is the hand with which that day’s opposition starter uses to hurl baseballs. Using the matchup tool created by myself and Ian Malinowski of Draysbay fame in conjunction with Steamer projections (best we have so far) you can get an idea of how each of our batters would project against a league average version of that arm. Here’s a look at that with projected (non-concrete, eat me!) positions:

(Note that OBP/SLG relates to career rates versus that type of pitcher without regression or park adjustment)

While the lineup against righties is fairly cut and dry the reason for the ire above stems from the need to play at least three lefties against a same-side starter. Something that could have been avoided by signing another good righty stick, but whatever. I’m over it. I swear. As an aside, many things jump out, but perhaps the unsungest of never sang is that Michael McKenry should probably be given a legitimate shot to make this team. Currently, he is not on the 40-man roster, but we’ve got 45 days til first pitch so there is a lot that can still happen. While he has the same SLG against both types he throws in a bunch of OBP versus lefties. With Wilson Ramos not an option to start the year the team could easily slide Luke Maile back to Durham in order to let Casali and McKenry be the catching Janus. He has traditionally been a below average framer and thrower, not nothing, but Casali is fine at both of those things so this could be a good chance to mesh skillsets.

Now we could go back and forth arguing the merits of who should slot where, but smarter folks at Baseball Musings than any of us have built a tool that makes it easy to project run scoring based on these OBP/SLG figures. Here is what my projected lineup versus right-handers would look like:

(Open in a new tab if you like to be able to read things)

The Rays look like pretty solid against righties with their most optimal lineups putting up something like 4.9 runs per game, which would be one of the best marks in either league. Of course, this is optimal. Guys are going to get hurt, lesser players need to play to stay fresh, you’re never actually facing a league average pitcher, etc… Still, it feels like the team should be pretty fine against righties.


Let’s see the other side of the coin:

(again, open in new tab to make larger)

We see a significant step back here with the team putting up around a third less of a run, but the lineup does look better than I expected. People are sleeping on Rickie Weeks Jr., whom I think was a tremendous MiLB/NRI guy that isn’t even on the 40-man roster, yet. Weeks brutalizes lefties, and is still a good baserunner during the run of play, if no longer the stolen base demon he once was. The lefties end up buried in this lineup, which is mostly fine, as we don’t expect any of them to be big contributors against righties.



The 2017 Tampa Bay Rays will continue to feature a lineup that places priority on slugging over on base. Due to their more of a boom and bust nature the ceiling is just much higher for these players even if they lack consistency. For a team like the Rays that is coolly calculated in all things it makes sense to chase a strategy of higher variation. If it doesn’t work out, well, nobody thought much of these bums, anyway. In the event that there is over-performance the rewards push just so much higher. The Rays have assembled a team that is capable of getting to the Promised Land. Valhalla awaits a Rays club that looks poised to hit better than last year, if not quite back to the glory daze.


Leave a Reply

#layout { padding-left:20px; }