Five Sabermetric Overperformers: 2014 Studs Due for Regression

March 21, 2015

Jake Bogardus uses sabermetrics to analyze 5 potential fantasy baseball busts in 2015, or MLB players who were fantasy overperformers including Cueto, Altuve, Cruz, Upton, Fister.

Hitters and pitchers will disagree on the ratio - but there is a general idea in baseball that for every hard-hit out you make, a blooper will fall in for a hit. This idea that chance plays a part in the game is the concept underlying the definition of regression. A hitter can sting the ball ten straight times, but if the ball happens to be hit right at a defensive player the quality of this hitters at-bats will not show up in our traditional stats such as batting average. This means that we cannot always take our traditional stats, which happen to be the numbers that we use to measure value in fantasy baseball, at face value.

Regression is defined as a measure of the relation between the mean value of one variable and corresponding values of other variables. A negative connotation is normally associated with regression, but it's important to note that regression works both ways. This means that regressing to the mean could indicate an increase in performance going forward.

Oftentimes, when fantasy players search for regression candidates, the first thing they do is look at players who had unexpected big seasons in the year prior. While this thought process can sometimes be successful, it is certainly not foolproof as it is not based in the underlying numbers that tell the true story.

If one used this simplified regression thought process on a player like Jose Bautista after his 2010 breakout, you would have missed out on Bautista's fabulous 2011 follow-up season. Bautista is a great example of why simply looking for the surprise breakout season and immediately considering that player a regression candidate is a dangerous line of thinking. The fact is, Bautista completely recreated his swing going into that 2010 season. That recreation led to more fly-balls which Bautista was trying to pull and as a result his HR/FB (Home Run per Fly Ball) ratio skyrocketed. The power surge was totally explainable.

Fantasy Producers That Will Decline in 2015

For the purpose of this post, we will be looking at players who we would expect to see a negative statistical shift moving from last season to the upcoming season. Potentially more important, we will also be identifying the types of underlying statistics that can tip you off to these regression candidates. These underlying statistics are great to focus on in the offseason during your draft preparation, as well as in season when looking for buy-low trade candidates. It's important to note, that while the underlying numbers suggest that these players will see a negative trend in their traditional statistics, there is no guarantee as luck will always be involved.

Jose Altuve, 2B, Houston Astros

Fantasy Regression Statistic: AVG (Batting Average)

Reason for Regression: BABIP (Batting Average on Balls In Play)

Let's start with an example that is similar to the discussion in the opening paragraph. Batting average is a statistic that can be greatly affected by luck. While players may say that for every hard-hit out there is a blooper that falls in, over the course of a season that ratio may not necessarily hold true. BABIP, or Batting Average on Balls In Play, is a statistic that attempts to identify players who are getting lucky or unlucky. BABIP is essentially a measure of your batting average that excludes K and HR to only account only for balls that were within the field of play. Typically, a BABIP  of .300 is the mean for the Major Leagues.

Before last season, in over 1400 Major League at-bats, Altuve sported a BABIP of .317. BABIP is dependent upon several factors, including which types of balls you put in play (ground balls, line drives, fly balls) . Being a line drive hitter and a speedster, Altuve's above-average BABIP is no big surprise, but his .360 BABIP in 2014 is something that warrants attention. Even with his slight increase in line drive percentage over the years, a .360 BABIP is likely unsustainable. Altuve should see that number drop closer to his career average, meaning his AVG is likely to drop from a league-leading .341 last season down into the .300-.310 range for the upcoming year. Steamer, a baseball projection system, is a bit more pessimistic, projecting a .299 AVG. Furthermore, this dip in AVG means less time on-base which could ultimately result in fewer R and SB this year. Altuve still can be considered somewhat of a top-of-the-draft player, but do not expect him to duplicate last season's numbers.

Nelson Cruz, OF/DH, Seattle Mariners

Fantasy Regression Statistic: HR (Home Runs)

Reason for Regression: HR/FB% (Home Run to Fly Ball Percentage)

Nelson Cruz was a fantasy stud last year. He was a low risk high reward signing for the Orioles and turned out to be the same for anyone who took a chance on him in last season's fantasy draft. Cruz hit a career-high and Major League leading 40 HR last season, after averaging 27 HR in the previous three seasons with Texas. In the process of hitting those 40 HR, Cruz sported a 20.4 HR/FB% , which was well above his 17.3% career average.

In addition to the HR/FB numbers, Cruz will turn 35 years old during the season and has had a history of injuries. Lastly, Cruz is moving from a HR friendly ball park in Baltimore to a HR stingy ball park in Seattle. All in all, Cruz is an aging, injury prone player moving to a pitcher's park with an expected decrease in his HR/FB%. This is the type of player that screams regression. Cruz may struggle to get to 25-30 HR this year (Steamer projects 26) and should be drafted with more caution.

Doug Fister, SP, Washington Nationals

Fantasy Regression Statistic: ERA (Earned Run Average)

Reason for Regression: FIP (Fielding Independent Pitching)

Fister has had some issues staying on the mound in his career, only topping 200 innings twice, but last year's 164 innings were truly spectacular from a fantasy standpoint. Fister racked up 16 wins on the back of a career-best 2.41 ERA which was sixth in all of baseball. However, if we look at the underlying numbers, once again a different story is told.

FIP, or Fielding Independent Pitching, is a statistic that attempts to measure what a pitcher's ERA would look like if they experienced league average results on balls put in play as well as league average timing (how often and when runner's are stranded on base). Fister's 2014 FIP came in at 3.93. His difference of 1.52 between ERA and FIP was the biggest in the Major Leagues last season. To put it in further perspective, Clay Buchholtz had a FIP just above Fister (4.01), but ended up with a 5+ ERA. Chance truly does play a role in these numbers, and Fister was on the lucky side last season.

Johnny Cueto, SP, Cincinnati Reds

Fantasy Regression Statistic: ERA (Earned Run Average)

Reason for Regression: LOB % (Left-on-base Percentage)

Cueto was one of the biggest breakouts and fantasy studs of the 2014 season, and if it wasn't for an other-worldly year from Clayton Kershaw Cueto would likely have some Cy Young hardware in his trophy case. Cueto finished the year with 20 wins while sporting a nifty 2.25 ERA in 243+ innings. However, once again there is more to that ERA than meets the eye.

The second part of our definition of FIP from the Doug Fister section mentions "league average timing." A large part of "timing" is how often a pitcher is able to strand runners on base. Generally, the league average LOB% falls between 70-72%. Cueto has always performed above this average, falling in the mid-high 70's most seasons, but last year his astronomical 82.5 LOB% hints at a rate that is unsustainable. Expect Cueto's LOB% to fall back to his career norms this year, ultimately resulting in a higher ERA (Steamer predicts a 3.48). Of course, there is a domino effect here, where a higher ERA will generally mean less wins. And fewer innings pitched meaning less strikeouts as well.

Justin Upton, OF, San Diego Padres

Fantasy Regression Statistic: Across the board (AVG, HR, RBI, R)

Reason for Regression: Park Factors

No surprise here, San Diego is a really tough place to hit and Upton will feel the impact when he starts his first season with the Padres. While Padres pitchers have always enjoyed the confines of Petco Park, hitters quickly learn to dread the deep power alleys. Petco has ranked as a below average hitters park since it's inception, and has been in the bottom three parks for offensive production in recent years. This does not bode well for Upton, who is moving from a league average hitters park to the canyon that is Petco Park. Upton put up a monster fantasy season last year, but the days of near 30 HR, 100 RBI and .270+ AVG seasons may be behind him, at least as long as he stays in San Diego.