08 April 2016

Expanded Roster: Last Year The Orioles Had A LOB Problem

Camden Depot is proud to offer opportunities for writers to share their thoughts on our ESPN-affiliated site. Articles are often one-time things or are an entry into becoming a full-time writer here. Feel free to contact us at CamdenDepot@gmail.com if you wish to inquire on the process.

Last Year The Orioles Had A LOB Problem
by Joe Wantz

photo by Keith Allison
Last week’s release of Miguel Gonzalez marked the end of his mostly successful career as a Baltimore Oriole. From 2012-2015, Gonzalez was third amongst Orioles starters in ERA and inning pitched and came into 2016 as a presumptive member of the starting five. Those solid numbers, however, masked frightening peripheral stats. Gonzalez posted the worst FIP and xFIP of any pitcher who started more than 20 games for the Orioles in that time period, and the wheels began to totally fall off in the second half of 2015.

How, then, was Gonzalez able to be a steady, if unspectacular, starting pitcher for three and a half years? While there are a few possible answers, one of the most glaring issues was Gonzalez’s left on base percentage (LOB%). When runners got on base against Gonzalez, especially in 2014, they almost never scored. That season, Gonzalez posted a LOB mark of 85.5%, which would have led the major leagues if he pitched enough innings to qualify. For comparison’s sake, Clayton Kershaw stranded 81% runners in 2014. Clayton Kershaw was worse at stranding baserunners than Miguel Gonzalez, and not by a small margin.

While this is striking enough, considering the massive skills gap between Gonzalez and someone like Kershaw, it is not an outlier in the Orioles rotation. From 2012-2014, the starters had the seventh best LOB% in all of baseball despite having a generally underwhelming set of arms. They were able to accomplish this despite ranking in the bottom 10 in the league in line drive rate, ground ball rate (dead last), fly ball rate (also the worst in baseball), K/9, BB/9, and HR/9 (last yet again). They also had the third largest gap between staff ERA and FIP of any team in baseball.

Needless to say, this is not the profile of a good starting staff, much less the starting rotation for the team with the most wins in the American League during that time period. Yet, that was exactly what happened. And while LOB% may not be the only reason for their surprising success, it certainly was a driving factor. No team at the top of the rankings in LOB% had a higher WHIP than the Orioles. Essentially, the Os were putting more runners on base than elite pitching staffs like the Nationals but stranding them at the same rate. So, while the Os were putting a ton of runners on, they were able to keep them from scoring.

A lot of that success likely came from an elite defense, which can be extremely helpful in helping pitchers post low batting average on balls in play (BABIP) numbers. The defense fell off a bit in 2015, perhaps correlating with the disaster that was the starting rotation, which went from stranding 77.6% of baserunners (second in baseball) in 2014 to 72.3% (17th best) in 2015. The two biggest culprits in this fall were Chris Tillman and, you guessed it, Miguel Gonzalez. Tillman saw his LOB% drop from 76.7% to 68.2%, with Gonzalez’s fall an even more precipitous 85.5% to 73.1%.

So, what does this mean for 2016? It is certainly reasonable to believe that Tillman can bounce back to being at least the league average mark of about 72%. If Tillman had the exact same stats as last year, but with that LOB mark, his ERA would fall from 4.99 to 4.58. That would certainly be a start, especially if he can get back to around the 200 inning mark. His Opening Day performance, though abbreviated, looked like a great start in that regard. It is likely not realistic for Tillman to maintain an average velocity of 93.1 MPH like he did on Monday, but an increase in his strikeout rate will lessen the overall need for an elite LOB%.

The rest of the current staff did not have the kind of massive change in LOB% that we saw from Tillman and Gonzalez last season, though it is worth noting that Yovani Gallardo had one of his best LOB% seasons in 2015. The rotation turnover that has occurred since the beginning of Spring Training (Gonzalez released, Kevin Gausman on the DL, Mike Wright, Vance Worley, Tyler Wilson all potentially making April starts) makes it difficult to forecast whether 2015 was an anomaly or a sign of things to come this season. That said, with a healthy Jonathan Schoop, JJ Hardy, and Matt Wieters up the middle, along with the emergence of Rule 5 pick Joey Rickard in left, perhaps a return to the defensive greatness (and resulting low BABIP and high strand rate) of 2014 may once again be in order.

7 comments:

Roger said...

Just goes to show that there's more than one way to skin a cat. Does this account for an elite bullpen who strands inherited runners? Do those runners count as LOBs for the starters or stranded runners by the bullpen? Seems like the O's bullpen helps with this statistic too.

Roger said...

I think, in addition to having some injuries on defense that the middle relievers were not as good last year. The rotation is better, if for no other reason than subtracting Gonzalez and Norris. The middle relief is better with the rookies - Bundy/Wilson - and Worley. Defense is better with Rickard, too. There's been a lot of negativity around this team and three games is a very small sample, but they look a lot better than advertised/predicted. Even in spring training they looked bad against the Twins much like last year when it was painful to watch them play the Twins and the Tigers. This initial series says a lot.

Matt Kremnitzer said...

It's too early to say anything definitive. But hey, taking the first three games of the season is great, and there are some positive signs.

Jon Shepherd said...

A reliever is not responsible for runs attributed in LOB.

LOB% = (H+BB+HBP-R) / (H+BB+HBP-1.4*HR)

LOB% is largely considered a constant except for certain pitcher types. Those types tend to be extreme groundball pitchers and guys who rack up strikeouts.

Joe said...

The bullpen is certainly good and helps the starters in a lot of ways, but there wouldn't seem to be an obvious reason for the bullpen to be better for some starters than for others. I think, like Jon is saying, Gonzalez didn't really fit the profile of a guy who should post elite LOB% numbers.

Scott said...

Relievers certainly play into the equation. You would have to stack inherited runners scored and stranded vs. league average to accurately understand the defensive contribution.....Roger is correct.

Jon Shepherd said...

That is not how LOB% is calculated. To see relief contribution one needs to do something differently.