Blake Murphy Sports Writing

Title: Sabermetric Mining: FB%, HR/FB, “Lucky” Homers
Date: August 31, 2012
Original Source: Full Spectrum Baseball
Synopsis: The latest in my Sabermetric Mining series took a look at FB% and HR/FB, two home run-related sabermetric stats, and included an exploration of the Hit Tracker tool, with the goal of identifying over- and under-performers in the home run category.

Finding ways to leverage sabermetric statistics for the purposes of finding Home Run value can be a tricky game. Where batting average allows us to delve into BABIP, batted ball type, and more, and there are plenty of peripheral indicators for pitching stats, Home Runs tend to be a stat that most people look at as having been earned, with less luck involved than others. However, that view can be detrimental to our analysis, as we can look to three indicators to aid us in mining for power over- and under-performers: Fly Ball Rate (FB%), Home Runs Per Fly Ball (HR/FB), and the Hit Tracker tool.

My apologies for no DOTF or Sabermetric Mining piece last week. I was driving from Kitchener, ON to Vancouver, BC and then settling in to a new place.

The Stats
FB% – Fly Ball Rate is the percentage of batted balls that a player hits in the air. When we analyzed hitter BABIP, FB% was thought to be a negative as fewer fly balls drop in for hits than ground balls or line drives. For power hitters, however, fly balls are of grave importance. After all, ground balls cannot clear the fence. FB% can help us to determine whether a player has the right batted ball profile to succeed in hitting home runs, but it is the rate at which those fly balls leave the park that is key.

HR/FB – this is the percentage of fly balls that clear the fence. HR/FB is the key item we will examine when trying to determine over- or under-performers, as HR/FB stabilizes at about 300 plate appearances. This means it can help to both identify lucky and unlucky players and players demonstrating a legitimate change in skill. It is important to compare a player’s HR/FB to his career norms, as we must judge if a drop in HR/FB is a blip or a trend, and vice versa. As a reference, an average HR/FB rate is about 10% in recent history. I should note that there is a lot more research done on pitcher HR/FB, if you are interested in further reading, as it is generally thought that a hitter has more control over his rate than a pitcher.

Hit Tracker – Thanks to the great Hit Tracker Online tool, we now have a resource for determining lucky homeruns. That is, a ball that clears Petco would likely clear any park, while a ball leaving Coors may not leave most stadiums. The key areas to view on this site are “No Doubts,” or balls that would cleared the fence by 20 vertical feet and 50 horizontal feet, “Just Enoughs,” or balls that cleared the fence by less than 10 vertical feet or just past the fence horizontally, and “Lucky Homers,” or balls that would not have been home runs on a neutral weather day. Obviously, Lucky and Just Enough homers are less indicative of a power skill than No Doubt homers or other, unclassified home runs somewhere between those end points. It is a lot to take in at once, but I highly recommend exploring the site as it has a ton of interesting information that extends beyond fantasy use.

Park Factors Affecting HR/FB
Park Factors should always be kept in mind, as HR/FB does not control for parks. Again referencing Petco and Coors as our polar examples, a HR/FB of 15% is far more impressive at spacious Petco than it is at the bandbox in Colorado. If you are interested in further and much more specific information on the topic, Jeffrey Gross of The Hardball Times tackled park factors extensively in June of 2011.

How To Use
It is difficult to just provide a link or a chart to help utilize these stats, as they do not all indicate over- or under-performance. The best means of approaching the analysis may be to scan the home run leaders for names that do not intuitively make sense or look out of place, both at the high and low end, and then use these tools to confirm or reject your initial thoughts. Additionally, using the Hit Tracker tool and subtracting “lucky” homers from totals, or simply looking for extreme performances at either end of the HR/FB spectrum, can provide a good starting place.

Edwin Encarnacion – Let’s begin with the league’s leader in FB% and one of the more surprising home run providers of the season, the hitter formerly known as E5. With a 49.7% FB%, you could employ hyperbole to say “he hits everything in the air” and you would hardly be wrong. Because his FB% is so large, even a modest HR/FB rate would lead to a large number of long balls, but Edwin also sports a 17.9% HR/FB rate, a near-elite rate. Edwin has 34 home runs, 8 more than his previous career high, and in less at bats (thus far) to boot. Looking at prior seasons, Edwin displayed an above-average HR/FB every year but 2007 and 2011, with an above-average 12.8% career mark. He also has a 45.2% career FB%. Add it all up, and Edwin has made a modest improvement to his HR/FB, increased his FB% to make the impact exponential, and received consistent playing time, making his home run surge only a moderate, and likely sustainable, surprise.

Billy Butler – People have been waiting for Butler to turn his 240lbs+ into home runs for some time now, and his previous career high of 21 bombs was nearly maddening. Butler had essentially been a monster who hits like a lead-off man. So what’s changed? In terms of batted ball profile, Butler has actually gone in the opposite direction of what you would expect given his homer surge, as his ground ball rate is at a stand-still and he’s traded fly balls for line drives. His home run total of 25 has been fueled entirely by a 22.5% HR/FB rate that is nearly double his previous career high. While Butler’s body type might lead one to expect an elite HR/FB rate, this kind of an extreme jump has to be cause for concern. I would expect Butler to slow down on the long-balls down the stretch, and his 2013 first half rate will be worth watching.

Asdrubal Cabrera – Cabrera was in the Butler/Encarnacion break-out class last year with 25 home runs, but he has fallen back to just 14 this year. When we consult his batted ball data, we see that he has hit slightly fewer fly balls (35.3% compared to 38.7%) and had a fewer percentage clear the fence (10.1% compared to 13.3%), neither of which is surprising given the magnitude of his 2011 breakout compared to his established norms to that point. Even still, we find that he might be over-performing in the category, as he is tied for second in baseball with 4 “lucky” home runs, while just 3 of his 14 have been of the “no doubt” variety. Last year, he came second in the league with 15 “just enough” home runs, indicating that he was getting lucky last year as well, which I’m sure many assumed. It seems likely Cabrera is not even an above-average power hitter, though at shortstop he obviously still holds fantasy value.

Ryan Ludwick – Ludwick technically does not have enough plate appearances to qualify for the leaderboards in FB% and HR/FB, but he sure has enough power to qualify as a leader in the counting stats. In just under 400 plate appearances, Ludwick has smashed 25 homers, a total he hasn’t touched since 2008 when he hit 37. So what happened to Ludwick between then and now, and how did he get back here? Well, Ludwick has always had good HR/FB rates, except last season, but this year he’s setting a career-high mark of 21.2%, a mark that would be top-15 in baseball if it qualified. Ludwick has always hit a lot of fly balls, and though his rate has declined to 43.1%, he’s still in the Edwin mold of ‘hit everything in the air and hope it flies.’ What is even more encouraging is that Ludwick leads the NL in “no doubt” shots with 9, and while he gets some benefit from playing in Great American Ballpark, 20 of his homers (80%) would have left at least half the parks in baseball.

Candidates
I should note here that the ‘candidates’ section this week might be more useful for those in keeper or dynasty leagues, as the month of September may not be a large enough sample to see appreciable correction for any of these players.

Potential Sell Highs and Buy Lows – Instead of identifying both separately like most weeks, this week I will instead show the home run leaders with their relevant statistics heat-mapped, as discussed. Some may have unsustainable HR/FB rates, be getting lucky on home runs, or be legitimate sluggers.

Home Runs are not always spread nice and evenly throughout the year, and power hitters tend to be streakier than contact hitters, it seems. Thus, we must be careful when looking at short-term blips in home run-related statistics, using all three of these tools together to identify the Edwin-like breakouts and the Asdrubal-like over-performers. While September may not be long enough to see full correction to the totals, those lucky enough to be in tight races will want to leverage any potential advantage available to them.

Come get to know me on Twitter @BlakeMurphyODC.
All stats courtesy of FanGraphs and Hit Tracker, for games through August 29.