Sunday, August 17, 2014

Thoughts on Lineup Construction Part 1

In yesterday's Comments, we had a discussion and a bit of a dustup about lineup construction which started with a comment about where Joe Panik should be hitting in the lineup.  This is something I have given a lot of thought to over the years.  Back in the day, there was a long running discussion among Giants fans about whether Barry Bonds should be hitting #3 or #4 in the lineup.  Felipe Alou had announced he was going to try moving Barry to the 3-hole.  I wrote a post on sfgiants.com Message Board using a simple mathematical calculation to argue that Barry should bat 4'th because he would get more RBI opportunities in the first inning there than hitting 3'rd. Two days later, it was announced that Barry would be staying put in the cleanup spot because hitting 4'th gave him more RBI opportunities.  I like to think Barry or someone else read that post, but there are other places he could have gotten the same information from or he could have figured it out on his own, so I will never know.  I also had several discussions with someone over on the mlb.com Message Board about lineup construction in general, mostly about the relative value of OBP in different areas of the lineup arguing against the notion that OBP is the #1 priority throughout the lineup.

I'll start this discussion with a few general thoughts about lineups then talk about what types of hitters you look for in each of the lineup positions in subsequent posts.  First, I would like to say I have not read The Book or really any other sources on the subject.  My thoughts come from the above experiences and thoughts I have put together through years of watching the game:

1.  The biggest impact of ineup construction probably occurs in the first inning of each game because in all subsequent innings the leadoff hitter occurs at approximately random locations in the order.  It may also matter to a lesser degree at the turnaround between the bottom and top of the lineup that generally occurs 3-5 times per game.

2. The benefits gained from the first inning must be balanced by the fact that given the randomness with which the last out of each game occurs in the lineup, each spot you drop a batter in the lineup costs that batter approximately 18 PA's over the course of a season.  That is probably negligible from one slot to the next, but the 9'th spot in the order will have approximately 162 fewer PA's over the course of the season than the leadoff spot, which is huge.  The difference between the leadoff spot and the cleanup spot is about 54 PA's which is not negligible.  That's 3+ HR's for a hitter who averages a HR every 15 PA's.  Those fewer PA's have to be balanced against the increase RBI opportunities your best hitter will get by hitting lower in the lineup in the first inning and farther away from the 8-9 spots in subsequent innings.

3.  AL and NL lineup construction will be different due to the difference between having a DH in the AL instead of a close to automatic out(in most cases) in the 9-hole of in the NL.  Guys who draw a lot of walks in front of the pitcher are probably not helping their team a lot whereas the presence of a competent hitter in the 9-hole in AL lineups would tend to favor high OBP hitters throughout the lineup.  This is probably a factor in the relative prevalence of "sabermetrically oriented" GM's and managers in the AL vs the NL.  For this discussion we will assume the hypothetical team is in the NL and the pitchers will be hitting.

4.  For the purposes of this discussion we will stipulate that if you had 9 Barry Bonds on your team, you would be very happy to have one of them hitting leadoff and another hitting #2 and so on.   Ditto for Ricky Henderson.  We are assuming an average to above average payroll NL team where you are not going to have the perfect hitter at every or even multiple positions.

2 comments:

  1. "The biggest impact of ineup construction probably occurs in the first inning of each game because in all subsequent innings the leadoff hitter occurs at approximately random locations in the order. It may also matter to a lesser degree at the turnaround between the bottom and top of the lineup that generally occurs 3-5 times per game."

    This just made me want to bring up one thought: Your leadoff hitter might occur at what you're calling "random orders in the lineup", but he is always benefitting from having guys right behind him who are oft capable of knocking him in. This is why position in the order matters. Hence, very good hitters who are followed by lower average, lower OPS+ players are less likely to have their contributions cashed in on. However - someone has to suffer that fate.

    ReplyDelete
    Replies
    1. That is why you want to have as "deep" a lineup as possible with only the pitcher, or the at the very most the 8 and 9 holes being "automatic outs."

      It also explains why the Giants struggled so much to score runs when Pagan and Belt both went down at the same time and Morse went into a big slump. The effective lineup was shortened down to 3 hitters, Pence, Sandoval and Posey. You just aren't going to sustain many rallies with 3 good hitters in the lineup.

      Delete