This week wraps up a four week tour of how opponents and quality of defense affect hitters and pitchers. So far I’ve looked at how defense and pitchers affect batters, and last week I looked at how the quality of the opposing offense affects a pitcher. This week, I’ll be looking at how a pitcher is affected when he has varying levels of defense playing behind him.
The study I’m not doing
Only bother reading this if you’re interested in the gory details of the study. If you trust my judgment and only care about the results, feel free to skip to the next section.
This week’s study will be a little different than past weeks because there would be some selection bias if I were to do it the old way. That is, if I were to simply look and a pitcher’s stats with a good defense behind him and compare it to a pitcher’s stats with a bad defense behind him, the pitchers in each pool may not be randomly selected. It’s possible that pitchers with a good defense may be better overall than pitchers with a poor defense. This is because if a pitcher is good, he’s going to cost more for a team to acquire/keep. If a team has the money to keep said pitcher, that team is more likely to have the money to acquire a good defense as well. I can’t say for certain that this is the case, but since it’s a possibility, I’d rather not bias the study and will tackle this a different way.
The study I’m doing
Because of this, I’ll need to tackle each stat that we FanDuel players care about (IP, W, ERA, K) separately. To start things off, I’ll be comparing the difference between a pitcher’s ERA and his xFIP based on how good his team’s defense is (adjusted for differences in league average ERA and xFIP). To measure defense, I’ll be using Ultimate Zone Rating (in the form of UZR/150) with a high UZR being good and a low UZR being bad.
Here are the results of the first part of our study:
We see exactly what we’d expect. With the support of a great defense, a pitcher will outperform his xFIP by 0.21 runs. With a poor defense behind him, he’ll underperform his xFIP by 0.18 runs.
To determine defense’s impact on innings pitched, I decided to run a regression analysis using ERA as the independent variable and IP as the dependent variable. By doing this, we can see how many innings a pitcher will throw based upon how many runs he lets up. Using this, we can see how many innings a pitcher will throw based upon the change in his ERA that the defense imparts. Here are the results:
We don’t see very large differences here. There’s little more than an out separating the innings a pitcher throws with a terrific defense and with a terrible defense behind him.
To calculate the impact a defense has on a pitcher’s wins, I’ll be using Bill James’s Pythagorean Theorem, which calculates win percentage based upon his runs allowed and runs scored. I’ll be starting off with league average innings, starting pitcher ERA, bullpen ERA, and opposing runs scored and adjusting these based upon our results up to this point. Once everything is accounted for, here are the results for how much defense affects a pitcher’s win percentage:
So the difference between a great defense and a poor one is a 2% difference in win percentage. Over 32 starts (a full season), that’s roughly 0.6 wins. Hardly something to get excited over, but also not something we should completely ignore either.
It’s unlikely that defense affects strikeouts by very much aside from a pitcher throwing more innings and generating more strikeouts as a result, so I won’t go into too much detail here. The difference here is 0.12 K/9 between the best and worst defenses.
Adding all of this up, here’s how defense seems to affect a pitcher’s FanDuel value. Below is a chart that lists the overall points per start a pitcher would produce given each type of defense, followed by the approximate FanDuel dollars those points would translate to (using a basic regression equation I created).
While none of the underlying effects are enormous, when you add them all up, we do get some actionable results. A pitcher’s value will be roughly $250 higher when he has a good defense behind him. It’s not a lot compared to the $1500 difference we found for pitchers versus good and bad offenses last week, but it’s still a noteworthy data point.
Like I’ve said each of the past three weeks, this study was conducted knowing who the good and bad offenses would be for the year ahead of time. At the start of a season, we can’t be sure which defenses are good and which are bad, though we do have some idea. Padres pitchers are probably good plays (especially at home against poor offenses), while Pirates pitchers won’t be as good as their underlying peripherals suggest.
Derek Carty’s work can also be found at The Hardball Times Fantasy and DerekCarty.com. He has previously had his work published by NBC’s Rotoworld, Sports Illustrated, FOX Sports, and USA Today. He is the youngest champion in the history of LABR, the longest-running experts league in existence, and is a graduate of the MLB Scouting Bureau’s Scout Development Program (aka Scout School). He welcomes questions via e-mail, Facebook, or Twitter.