Stats & Info: Doug Davis

BIS: Taking BABIP one step further

May, 25, 2010
You've probably seen stories on other websites citing Livan Hernandez as one of the luckiest pitchers in baseball in 2010, and Cole Hamels one of the unluckiest.

A look at their BABIP (Batting Average on Balls in Play) would tell you that they've probably been helped (Hernandez), and hurt (Hamels) by their defense this season. But looking solely at BABIP doesn't tell us that for sure.

But now our group has found a way to offer more concrete proof of how much a pitcher is being helped/hurt by his defense.

Here's how:

In our book, The Fielding Bible – Volume II we spent a lot of time trying to separate defense from the legendary “pitching and defense” that wins ball games. We took individual defensive performances and broke them down to provide accurate defensive assessments.

What I soon realized was that by isolating defensive performance we also managed to come up with the roots of a system to evaluate pitchers independent of their defensive environment.

Sometimes the defense makes a nice play and helps out the pitcher. Other times, it falls between two miscommunicating fielders for a hit.

Often, the fielder is standing in just the right spot at the right time. None of this is under the pitcher’s control, but he gets credit for it regardless of the outcome.

If life were fair, these anomalies would even out over the course of a season.

Based on the characteristics of each ball in play (location, trajectory, etc.), I calculated the approximate chance that the average defense converted the ball in play to an out.

I added up this probability for every ball in play over each game and over the full 2010 season to arrive at an expected number of hits allowed given the distribution of balls in play.

Without boring you with the details and adjustments I made (for now), I’ll present a list of the “luckiest” and “unluckiest” pitchers this season (please note that our BABIP calculation may differ from others due to differing treatment of bunts)

The 3 Luckiest
1- Livan Hernandez
Hits Allowed: 40
Expected Hits: 55

Conclusion: Hernandez has allowed 15 fewer hits than expected. Had he allowed 15 more hits, his BABIP would have risen from .183 to .267

2- Doug Fister

Hits Allowed: 40
Expected Hits: 52

Conclusion: Had Fister yielded 12 more hits, his BABIP would jump from .225 to .294

3- David Price
Hits Allowed: 45
Expected Hits: 55

Conclusion: Had Price allowed 10 more hits, his BABIP would rise from .241 to .300.

Hernandez, he of the 19 strikeouts against 18 walks in 55 innings pitched, rates as the most fortunate pitcher in baseball. Based on the locations and trajectories of Livan’s balls in play, we’d expect 15 extra hits to have fallen in. But instead, those plays became outs.

One rule of thumb is to expect every pitcher’s Batting Average on Balls In Play (BABIP) should move toward the league average (around .300) as the season progresses.

In each of the three cases, we find their Expected BABIP to be closer to the league average than their current BABIP.

The 3 Unuckiest

1- Brad Bergesen
Hits Allowed: 60
Expected Hits: 48

Conclusion: The Orioles have played poor defense behind Bergesen. An average defense would have turned 12 more outs on the balls he allowed into play, cutting his BABIP from .327 to .251.

2- Doug Davis
Hits allowed: 48
Expected Hits: 39

Conclusion: Had the Diamondbacks performance behind Davis matched that of an average defense, his BABIP would drop from .400 to a more reasonable .320.

3- Cole Hamels

Hits Allowed: 60
Expected Hits: 51

Conclusion: Hamels is bound to catch a break at some point. His BABIP of .316 would be 55 points lower if the normally-good Phillies defense had performed well for him.

In theory, utilizing hit locations and trajectories will lead us to better pitching evaluations. The next step is to refine the technique and evaluate its predictive power on historical data, and hopefully we'll get the chance to follow that up with further study in the future.

Dontrelle Willis' exasperating start

April, 13, 2010
Dontrelle Willis cleared 100 pitches over his five-inning stint against the Royals on Tuesday. This is the kind of start that you could call "The Exasperator."

It's the kind of start that is frustrating for fans and frustrating for managers. And it's been a trend over the last few years that they've increased in number. But by how much?

According to, there were 173 starts of more than 100 pitches and five or fewer innings pitched in 2000, accounting for about 3.5 percent of all major league starts.

In 2005, there were 193 such starts. The next year, there was a jump to 229 and that figure has held steady. Over the last two years, there were 222 and 226 respectively. So now we're looking at about one in every 20 starts, meaning we're likely to see at least one of these per day throughout the year.

Who are your most likely culprits when it comes to exasperating starts?

Milwaukee Brewers ace Yovani Gallardo actually led the majors in Exasperators last season with seven. Daisuke Matsuzaka topped the big leagues in 2008 with six, but amazingly went 4-0 in his exasperating starts. That doesn't provide much of an incentive to get better.

But here's one. Of the 10 teams that had the most exasperating starts in the majors last season, only one (the Red Sox) had a winning record.

Peterson makes difference for Brewers

April, 11, 2010
The Brewers likely had enough offense to be a playoff contender last season, ranking third in the NL in runs and OPS. But on the mound, they flat-out stunk. Their starting rotation had an ERA of 5.37 that was the worst in the NL and tied for worst in the majors. They also allowed the most homers and most walks in the National League.

In the offseason, they acquired both Randy Wolf and Doug Davis to bolster their pitching staff. But perhaps the most important transaction was the hiring of pitching coach Rick Peterson. Peterson is noted for his work with the Oakland Athletics in the late ‘90s and early ‘00s. Under his tutelage, the A’s staff finished in the top three in ERA from 1999-2003, including the AL’s best marks in 2002 and 2003.

With the Brewers, Peterson has stressed pitching to the bottom of the strike zone because "the average batting average… on every ball put into the play at the bottom of the strike zone is about .210 or .220."

The Brewers actually did this well last season, with 50% of their total pitches thrown in the bottom third of the zone according to Inside Edge, the 2nd-highest percentage in the NL. Their batting average allowed of .213 on those pitches down in the zone was also in-line with Peterson’s expectations.

Highest Pct of Pitches in Lower Third of Zone
National League, 2009 Season

Cardinals 53.3
Brewers 50.1
Braves 50.0

However, despite keeping the ball low in the zone, they were not able to generate a lot of groundballs. Only 44% of balls in play allowed were grounders according to Inside Edge, the 3rd-lowest percentage in the NL. That was likely one of the key reasons for their awful ERA, as they allowed a lot of line drive and flyballs, which resulted in a high number of extra-base hits and homers.

So how has the pitching staff performed this season, based on Peterson’s new philosophy of pitching down in the zone?

Through Saturday, the Brewers had kept 51% of their pitches in the lower third of the zone and allowed a batting average of .197 on those pitches – just what Dr. Peterson ordered! However, their groundball percentage was just 38.9% - not exactly the number you’d want to keep your ERA under five.

Randy Wolf – Sunday’s starter vs. the Cardinals – will need to improve on his first start performance if he’s going to avoid Peterson’s doghouse. Only 44% of his pitches were located down in the zone and he allowed a batting average of .250 on those pitches. His groundball percentage of 43% was also mediocre, and just 7 of his 20 outs were groundouts.

On the other side of the plate, the Cardinals hitters this season have not been impressive against pitches down in the zone, with a .182 batting average that is below the league average of .194. But – and Wolf will need to keep this in mind – 4 of their 9 homers hit this season have come on pitches in the lower third of the zone.

This should be an interesting trend to watch Sunday night and for the rest of the Brewers’ season. Can the Brew Crew pitchers continue to pound the lower part of the strike zone, and can they consistently generate enough groundballs to keep the runs off the scoreboard?

You can watch the Cardinals play the Brewers at 8 ET Sunday on ESPN.

Stat Week: Another look at "quality" starts

March, 23, 2010
Baseball Tonight continues its look at statistical analysis by looking at pitching evaluation methods. This piece takes a closer look at evaluating starting pitching

A few years ago, ESPN baseball columnist Rob Neyer wrote a piece about why the quality start is actually a quality statistic. The key argument in his article is that in 2005 there were 2,447 quality starts in the majors, and a team won 67.4% of those starts.

This got us thinking…what if we could somehow predict the winning percentage of ANY team given ANY starting pitching line. This seems like it would improve the quality start metric, as it would give a better indication of how a pitcher impacted his team’s chance of winning. Others, such as Bill James, have devised methods to do something similar (you may be familiar with "Game Score," listed in box scores). We tried another approach.

First, we compiled the starting pitching lines from every starter for the last five seasons. There are approximately 2430 MLB games per season, two starters per game, for five seasons (2005-2009). That's 24,300 observations.

We then used a statistical technique known as binary, or logistic, regression to predict a team's probability of winning based on the starter’s pitching line. Essentially, we plug basic stats from a box score (such as innings pitched, earned runs, walks, strikeouts) into a mathematical model and the model then spits out the team’s chance of winning the game based on those stats.

In order to simplify the model and keep it as close to the current criteria of a quality start, we used only innings pitched and earned runs as variable for the regression. It’s also worth noting that those two stats – IP and ER – also had the largest statistical impact on a team’s win probability.

Let’s get to the data: Here’s the predicted team winning percentage for a few different combinations of IP and ER by the starting pitcher:

Pitcher A: 6 IP, 3 ER Team win pct = 49.6%
Pitcher B: 7 IP, 3 ER Team win pct = 55.0%
Pitcher C: 9 IP, 4 ER Team win pct = 54.5%
Pitcher D: 6 IP, 2 ER Team win pct = 60.9%

What’s most interesting here is that we see at least one non-quality start stat line (9 IP, 4 ER) which gives the team a better chance of winning than the minimum quality start criteria of 6 IP and 3 ER.

As to the question of how we can possibly improve the existing quality start metric. Here is a quick example of how our model helps better judge which starting pitchers truly impacted their team’s chance of winning the game.

Using the current definition of quality starts, Roy Halladay and Doug Davis tied for 15th in the majors last season with 22 quality starts. We think most fans would agree that Halladay is arguably a better pitcher than Davis and likely helped his team win more games.

That’s where our new model can help.

If we set the our quality start threshold to any start where the starting pitcher gave his team at least a 75 percent chance of winning, Halladay had 13, nearly twice as many as the seven by Davis. Halladay’s 13 tied for sixth-most in the majors. Here’s a look at the top five from last season:
Now, let’s remember that this is just a start (no pun intended), as this regression model can certainly be improved by adding more variables and conducting further tests. Hopefully, though, this a good primer for a different way to judge starting pitching success, and will spark some interesting discussions in ballparks and bars across the country this season.