Optimizing the Mariners’ Lineup

One of the manager’s many jobs is to put out the best lineup that he possibly can every game. Most of the time, this is not the case, objectively. Managers have different ideologies and styles, and their lineups often reflect so.

In economics, there is something called the rational choice theory. What the rational choice theory assumes is than an individual has preferences among choices. These preferences are assumed to be made by surveying all possible options, taking in all information, and to make the most logical decision that is of greatest benefit to them. As stated previously, this isn’t always true.

To put out the best lineup possible, one should use all of the information given. In The Book: Playing the Percentages in Baseball by Tom Tango, Tango (as well as Mitchel Lichtman and Andrew Dolphin) attempt to quantify this using empirical data. In the book, there is an entire chapter devoted to sabermetric lineup construction. Luckily, this information is consolidated into two short Beyond the Box Score articles for you to digest if you would like, here and here. Allow me to introduce you to the conventional makeup of lineups versus the “optimized” lineup.

BTBS optimization

(Originally, I had the 4th and 5th, so the 3rd and 5th spot, switched around. Tom Tango himself let me know, and I have adjusted all three graphics to reflect so.)

Using a combination of conventional and sabermetric statistics, I’ve attempted to construct the best possible Mariners lineups against both left handed pitchers as well as right handed pitchers.

What I’ve come up with is this:

Versus RHP:

vs RHP


Versus LHP:

vs LHP

Right off the bat (ba dum tsssss), there’s a couple things that one could reasonably change. It’s early in the year, and not everyone has played the same amount. Not everyone has been healthy, and not everyone is good! What this means is that there is a decent amount of variance in the statistics of Luis Sardinas, Ketel Marte, and Nori Aoki that could change rather drastically. Several players have LHP/RHP splits that right now don’t look anything like their career splits. 58 games in, that’s not all that crazy. I expect Dae-Ho Lee’s numbers against righties to see some regression, while his numbers against lefties will likely improve.

I’ve always been a huge proponent of Robinson Cano batting in the two hole in the lineup, and against righties, he is! The nice thing about how the sabermetrically optimized lineups worked out in these scenarios is that no key players’ egos will be severely hurt. Kyle Seager, who has been absolutely god awful against left-handed pitchers thus far, might take offense. Over his career, he has a 95 wRC+, which is only 5% below league average. Thankfully, Seager is destroying right-handed pitching this year.

Cano and Cruz both would appear in either lineup in the five hole. If we’re worried about pissing off our superstars, some concessions might be made to appease them.

One thing to notice is, holy f*ck, we’re really good against right-handed pitching. Against left-handed pitching, we’re pretty damn solid too.

And with that, I would like to add a couple things that I have learned from Tom Tango, a pseudonym for a man who previously worked for the Seattle Mariners as a statistical analysis consultant. Tango now works for the advanced statistics-loving Chicago Cubs and, oh how odd, they’re in first place in the MLB, far and away. Tango is also the creator of FIP (fielding-independent pitching) and wOBA (weight on-base average).

When it comes to consulting a batter’s statistics against a pitcher in which he has supposedly owned:

The Book says:

Knowing a player will face a particular opponent, and given the choice between the player’s 1500 PA over the past three years against the rest of the league, or twenty-five PA against that particular opponent, look at the 1500 PA.

Sometimes managers make decisions because a player has “dominated” another player, and so they’ll substitute the player into the game by pinch-hitting them or bringing a reliever into the game for a specific batter. And sometimes it works! Unfortunately, just because it worked doesn’t mean it was a good decision, similar to just because it didn’t work doesn’t mean it was a bad decision.

When constructing a lineup and a player is deemed to be on a hot or cold streak:

The Book says:

Knowing that a hitter has been in or is in the midst of a hot or cold streak has little predictive value. Always assume that a player will hit at his projected norm (adjusted for the park, weather, and the pitcher he is facing), regardless of how he has performed in the very recent past. A player’s recent history may be used as a tiebreaker.

The Mariners, before his injury, have recently done this with Leonys Martin. There is one possible caveat, however. If the Mariners believe that the Leonys Martin before his DL stint is the Leonys Martin he will be going forward, then I support his move from the bottom of the lineup to leadoff. But only against right-handed pitching.

And so, due to injuries, matchups, fatigue/maintenance days, etc., these lineups aren’t possible every game. But in an ideal world, this is how I would currently construct the lineups. I’m not mad at anyone who vehemently disagrees with my way of thinking, or creating these lineups, but this is the most sound way that I have at my disposal.

The Mariners play tonight at 7:10 PM and their lineup looks drastically different from my lineup. One reason is that Leonys Martin literally isn’t allowed to play. Another is that Scott Servais likely believes Adam Lind is the better matchup tonight rather than Dae-Ho Lee and despite recent history, I don’t think his thinking is flawed.

Now let’s get excited about a Taijuan Walker start and taking the lead in this series.

Go M’s.

One thought on “Optimizing the Mariners’ Lineup

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s