## Monday, December 24, 2012

### Best Lineup - Los Angels of Anaheim

In my previous post I used my baseball simulator to look at the best lineup for the Cincinnati Reds vs both a LH/RH pitcher. Up at the plate this time are the Angels. Keep in mind that the input projections I am using are from Bill James and the input projections go a long ways in determining which lineup is the best. What you will see with the Angels, which may look odd to some people, is the placement of their catcher Chris Iannetta in the lineup. Iannetta is one of those players with unspectacular statistics but he has put up some pretty impressive OBP numbers. Let's take a look at the results and then analyze a little bit what is going on.

## Sunday, December 23, 2012

### Best Lineup - Cincinnati Reds

Moving on a little bit from my series on where to bat your best and worst hitters, I am taking a look at the best lineup for an actual team. In this post I will be looking at the best lineup for the Cincinnati Reds, when they are facing both a right handed (RHP) and left handed (LHP) pitcher.
I am using my baseball simulator that I wrote in C language, which takes pretty much everything that you can model in to consideration. Keep in mind I did not run every single permutation of lineups but it is very easy to narrow things down to the best lineup with some experience of doing this.

I used the lineups at MLBDepthCharts as my baseline as far as the lineup "to beat" and who the actual starters would be against both RH and LH pitchers. For input projections I used the Bill James projections for each player available at Fangraphs. I played the Reds as the visiting team, against an imaginary team of players that I kept constant and an imaginary set of starting and relief pitchers for both teams. This gives a WOWY (with or without you) sense to this exercise. All I did was run simulations (1 million of them) with different sets of lineups, keeping everything else constant. Below are the best lineups that I could find for the Reds against both a left handed and right handed pitcher.

I used the lineups at MLBDepthCharts as my baseline as far as the lineup "to beat" and who the actual starters would be against both RH and LH pitchers. For input projections I used the Bill James projections for each player available at Fangraphs. I played the Reds as the visiting team, against an imaginary team of players that I kept constant and an imaginary set of starting and relief pitchers for both teams. This gives a WOWY (with or without you) sense to this exercise. All I did was run simulations (1 million of them) with different sets of lineups, keeping everything else constant. Below are the best lineups that I could find for the Reds against both a left handed and right handed pitcher.

## Wednesday, December 19, 2012

### Where To Bat Your Best Hitter - Part V

Same exercise as before. I am using my baseball simulator to find out what it thinks are the best lineup constructions. The only twist, is that I am not inputting normal 9 person lineups in to it. I am for the most part entering two person lineups, with one person taking 8 lineup spots and the other person taking 1. I then switch roles between the two players with the other batting 8 times. I am running each game one million times in order to keep the noise down to a minimal level.

In the latest exercise I was asked to run some simulations using players with extreme OBP and SLG% numbers. So what I did was dig through the past two seasons of stats over at Fangraphs and came up with the following combinations of players that I am going to look at. There is a lot of data here, so I leave the analysis to the reader for the time being.

They are...

Low OBP/Low SLG: Paul Janish

Low OBP/High SLG: Mark Trumbo

High OBP/Low SLG: Jamey Carroll

High OBP/High SLG: Miguel Cabrera

High OBP/Reasonable SLG: Derek Jeter

From this list here are the combinations I ran.

Janish(8) + Trumbo(1) / Trumbo(8) + Janish(1)

Janish(8) + Cabrera(1) / Cabrera(8) + Janish(1)

Janish(8) + Carroll(1) / Carroll(8) + Janish(1)

Carroll(8) + Trumbo(1) / Trumbo(8) + Carroll(1)

Trumbo(8) + Cabrera(1) / Cabrera(8) + Trumbo(1)

Carroll(8) + Cabrera(1) / Cabrera(8) + Carroll(1)

Jeter(8) + Trumbo(1) / Trumbo(8) + Jeter(1)

Jeter(8) + Cabrera(1) / Cabrera(8) + Jeter(1)

And now for the results...

## Sunday, December 16, 2012

### Where To Bat Your Best Hitter - Part IV

Next up in my lineup construction exercise, I am going to look at two things. The first, will be a team with eight Albert Pujols' and one pitcher. The pitcher will be moved from the first spot in the lineup through each spot, all the way down to the last spot in the lineup. The second, will be a team with one Albert Pujols and eight pitchers. Once again, we are not worrying about defense, only about offense and how many games are won. Everything else is kept constant.

Let's take a look at the results...

### Where To Bat Your Worst Hitter - Part III

Now let's turn upside down what we did in Part I where we had 1 Pujols and 8 Bourns. This time we look at a lineup with 8 Pujols' and 1 Bourn. Again we have 9 possible permutations. Let's see which lineup wins the most games.

### Where To Bat Your Best Hitter - Part II

In Part I of this exercise I looked at the best lineup construction for a team that had one Albert Pujols and eight Michael Bourn's. I then plugged these lineups, using Bill James player projections into my baseball simulator. The lineup that won the most games had Albert Pujols batting third.

In the next step of this exercise I replaced one of the Michael Bourn's with another Albert Pujols. What lineup with two Albert Pujols' in it would win the most games? Logic would tell you that one of the Albert Pujols' should be batting third, right? If so, then where would you slot in the second Pujols? Let's take a look at the 36 total permutations of lineups and see which ones did best. Once again, one million games were simulated.

## Friday, December 14, 2012

### Where To Bat Your Best Hitter - Part I

In this exercise I use my simulator to determine where to bat your best hitter in the lineup... but with a twist. I am taking a team of nine hitters, eight of which are a typical speedy leadoff hitter (in this case I am using Michael Bourn) and the ninth is Albert Pujols. Throw defense out the window in this exercise, just look at which lineup construction with eight Bourn's and one Pujols' would win you the most games.

There are only nine different permutations. Pujols can bat anywhere from leadoff to ninth in these lineups. Which slot do you think would be the best place to put Pujols. My simulator takes in to consideration all offensive traits (hitting, speed) with the hitting projections coming from Bill James projections which are the only ones out at this point.

Results after the jump....

### Best Lineup vs RHP

So, the Angels just signed Josh Hamilton to a new five year deal. Now, it is time to take a look at what the most efficient lineup should be. Today I tackle that exercise in a look at how the Angels sluggers should be stacked up against your typical right handed pitcher.

To do this, I am whipping out my simulator which plays actual baseball games using sabermetric principles. Each worthy lineup will be simulated one million times in a game against an opponent that remains constant, with a right handed pitcher. The Angels will be the away team in all these simulations to give them at least nine innings of at bats in each game. I picked a group of opponents such that the game results are close to 50% wins and losses for the Angels.

I used the starting nine players as listed at the Angels MLB Depth Charts page. The measurement to determine the best lineup will be number of wins out of a million games. Now, I did not try every single permutation of possible lineups so the exercise is not totally complete but after you play around with switching the lineups around you can quickly narrow down the best ones. Bill James 2013 player projections were used for input.

Below is the best lineup that I found.

1) Mike Trout

2) Josh Hamilton

3) Albert Pujols

4) Kendrys Morales

5) Mark Trumbo

6) Erick Aybar

7) Howie Kendrick

8) Alberto Callaspo

9) Chris Iannetta

This lineup won 523636 out of one million games. For reference the lineup that MLBDepthCharts provides with Aybar hitting second, Hamilton 4th with Morales and Trumbo pushed down one slot wins a total of 522341 games. The question then becomes, how significant is this? Over a 162 game season, this would net the Angels an extra 0.21 wins. This is obviously an amount not worth fretting over. The big differences come when you totally screw around with the lineup which not even Scoscia would dare do.

Here is a breakdown of how the box score broke down for this lineup.

Name | AB | Hits | H1 | H2 | H3 | HR | RBI | RUNS | BB | SO | wOBA |
---|---|---|---|---|---|---|---|---|---|---|---|

Mike Trout | 4.344 | 1.144 | 0.724 | 0.195 | 0.047 | 0.179 | 0.475 | 0.285 | 0.410 | 1.246 | 0.339 |

Josh Hamilton | 4.201 | 1.035 | 0.563 | 0.217 | 0.013 | 0.242 | 0.633 | 0.189 | 0.466 | 1.499 | 0.344 |

Albert Pujols | 4.101 | 1.156 | 0.643 | 0.295 | 0.000 | 0.217 | 0.660 | 0.180 | 0.452 | 0.640 | 0.372 |

Kendrys Morales | 4.217 | 1.094 | 0.679 | 0.250 | 0.007 | 0.159 | 0.556 | 0.140 | 0.260 | 1.074 | 0.319 |

Mark Trumbo | 4.135 | 0.961 | 0.554 | 0.200 | 0.009 | 0.198 | 0.524 | 0.144 | 0.218 | 1.284 | 0.300 |

Erick Aybar | 4.036 | 1.059 | 0.771 | 0.215 | 0.031 | 0.043 | 0.289 | 0.201 | 0.203 | 0.640 | 0.292 |

Howie Kendrick | 3.956 | 0.982 | 0.653 | 0.249 | 0.012 | 0.068 | 0.364 | 0.159 | 0.161 | 1.045 | 0.283 |

Alberto Callaspo | 3.670 | 0.960 | 0.689 | 0.206 | 0.013 | 0.051 | 0.335 | 0.149 | 0.346 | 0.510 | 0.310 |

Chris Iannetta | 3.475 | 0.710 | 0.415 | 0.165 | 0.005 | 0.125 | 0.384 | 0.128 | 0.435 | 1.191 | 0.292 |

Subscribe to:
Posts (Atom)