Wednesday, February 8, 2012

Project Context Neutral Runs and RBIs

Projecting a players Runs and RBI’s is a pain, and it’s largely considered contextual. So if you’ve got some good hitters hitting in front of you, you’re going to get more RBI opportunities, or good players hitting behind you, more Run opportunities.

The problem with this, is that context changes often. A team that previously stunk, may have a guy or two breaking out, and now suddenly another player is thrust into a situation where he can generate more runs. A key guy might get injured, traded, or simply moved around in the lineup.

For this reason, I’ve been working on a way to project a guys runs and rbi’s based on his skills alone. I’ve tweaked my process a bit over the years, and here’s what I’ve found the best method:

xRuns = HR + -.218 + .191 * (BB - .333 * CS) + .273 * (HBP + 1B - .666 CS) + .363 * 2B + 1.366 * 3B + .505 * SB

So just to summarize what this means, Extra base hits generate more runs, with triples generating bonus runs (because they indicate a speedy player, who’s going to score on more Singles then other guys), and net stolen base gains also improve your chance to score runs.

xRBIs = 2 * HR + .640 + .004 * (BB + HBP) + .234 * 1B + .427 * (2B + 3B)

With this one, again we see hits generate RBI’s, with extra bases generating more. Home runs generate bonus runs because you’re knocking yourself in, as well as anyone on base, and home run hitters generate more Sacrifice Flies.

Let’s look at some sample results from my 2012 projections:

name xRuns xRBI
Kemp 109 108
Ellsbury 103 97
Bautista 100 113
Bautista 100 113
Kemp 109 108
P Sandoval 89 105

Ellsbury is an interesting one on this list, Sitting in the #1 hole he traditionally had very few RBI’s historically, this system picked him for an RBI increase last year based on his budding power (and his lineup position changed to fit his new skillset).

Sandoval is also an interesting inclusion, he’s had budding HR and 2B power, and a change in his context could put him in line for a lot of RBI’s.

Obviously this method is not perfect, context does exist, I just find that it’s so fluid throughout the year, it’s fun to just ignore it, and project based on a batters skills. I find this particularly satisfying in fantasy baseball, because it’s a fun way to identify breakout players. A player with budding power (ellsbury, granderson), will eventually have their lineup position improved to take advantage of that power. These are two guys who I specifically drafted last year based on my projections, who both had their context improve, to match their skills.