The only sure thing about luck is that it will change

Famed playwright and raconteur, Wilson Mizner

That quote sums up what I am aiming to apply to this fantasy baseball analysis. I’m looking for starting pitchers that have been either lucky or unlucky in 2017, and digging deeper to see if I can take advantage in 2018.

There are a lot of metrics I like to look at to get a feeling if a pitcher was lucky or unlucky in the previous season.

Earned Run Average minus Expected Fielding Independent Pitching (ERA-xFIP)
I like to think of xFIP as the pitcher’s “true” ERA, or his ERA based on the things he can control (strikeouts and walks), and removes the things he has little to no control over (defense and BABIP). Subtracting xFIP from the pitcher’s ERA gives us an idea of whether he is lucky or unlucky. If the pitcher’s ERA is higher than xFIP it tells me that the pitcher may be unlucky and is actually pitching better than his ERA indicates. The opposite can be said if his ERA is lower than his xFIP.

Outside of ERA-xFIP, I like to look at batted ball metrics. We know that pitchers do not have complete control over what happens to a baseball once it’s put into play, but they do have some control over the type of batted ball they allow (ground ball, line drive or fly ball).

Batting Average on Balls in Play (BABIP)
A pitcher with a high BABIP is seeing more balls fall in for hits thus raising his ERA and WHIP. Things should even out over the course of the season and lower his ERA and WHIP.

Left On Base Percentage (LOB%)
A LOB% lower than 72.1% indicates that he is allowing more runners to score than the league average thus increasing his ERA. Like BABIP, this should even out over the course of the season and lower his ERA.

Ground Ball Percentage (GB%)
Ground balls are the best for a pitcher. That is because they tend to go for hits more often than fly balls (although they don’t result in extra base hits as often). But the higher a pitcher’s ground ball rate, the easier it is for their defense to turn those ground balls into outs. I like to compare the pitcher’s GB% rate to the league average and his career average to get any idea if luck played a role.

Home Run to Flyball Rate (HR/FB)
Pitchers generally do not have control over their HR/FB from year to year. The ballpark that a pitcher pitches in can have an effect on their HR/FB rate. A high HR/FB indicates that the pitcher is serving up more home runs than normal and thus their ERA is probably high. Over the course of the season their HR/FB should lower towards the league average and lower their ERA.

Here are the MLB totals for these metrics over the last three seasons, plus the 10-year average. These results are from starting pitchers only, no relievers.

Here are a few things that interest me from this data:

• Total innings pitched by starters is going down each season.
• All of these metrics are pretty consistent over the past three seasons, except HR/FB which appears to be steadily climbing.

For this exercise I went to FanGraphs and exported 2017 stats for starting pitchers only (134 pitchers). I then simply determined if each pitcher was lucky/normal/unlucky for each metric. To determine lucky vs. unlucky I came up with these guidelines. Anything in-between was considered “normal”.

For the players identified below, sabermetric stats are from FanGraphs.com, 2018 projections are from Steamer, and ADP is from FantasyPros on March 7, 2018.

LUCK COULD TURN AROUND

Lance McCullers Jr., HOU
ADP SP #33, Overall #119

In 2017, McCullers was one of nine pitchers that had his xFIP a whole run lower than his ERA. Combine that with his high BABIP and low LOB%, and you could say that McCullers was quite unlucky in 2017.

Why should you think he can bounce back in 2018? Well, he’s always been a groundball pitcher with a career 54.5% GB% (league average is 44.3%) and a 2.21 GB/FB (league average is 1.26). When he does let a ball get hit in the air it tends to stay in the park with a career HR/FB of 10.8% (league average is 13.0%). Those facts combined with the thought that his luck should turn indicates that he is someone you should not shy away from during your draft.

Chris Archer, TB
ADP SP #16, Overall #55

In 2017, Archer brought you a lot of K’s and D’oh!’s as he struggled through an up and down season. A high BABIP (27th highest out of the 134 pitchers in this study) is the metric that sticks out the most. But, if you look at the others they are all leaning towards “unlucky.”

Why should you think he can bounce back in 2018? Well, his xFIP in the first half of 2017 was 3.42 with an ERA of 3.95, while in the second half his ERA was higher at 4.27 but his xFIP was lower at 3.23, with the telling stat being HR/FB. In the second half his HR/FB was 18.0% compared to 10.4% in the first half. Prior to 2016 his HR/FB ratio was consistently below 11.7% even in the minors. My gut tells me that Archer just took a little longer to “right the ship” in 2017 but eventually got there in the second half. I’m expecting his 2018 numbers to end up more like his 2015 numbers.

BUYER BEWARE

Gio Gonzalez, WAS
ADP SP #38, Overall #140

In 2017, Gio had the 7th lowest* ERA-xFIP of the 134 starting pitchers in this analysis. I’m pretty sure that you don’t need to be a card carrying member of SABR to see that Gio was very lucky last season. Along with his ERA-xFIP he had the 12th lowest BABIP and the 6th highest LOB%.

Why should you avoid drafting him at his current ADP in 2018? While enjoying metrics on the very lucky spectrum his 45.8% GB% was not far from his 47.5% career GB% and his 11.1% HR/FB was higher than his 9.3% career HR/FB. This all adds up to a big red flag telling me to avoid drafting him this year.

* The other six are, in order of lowest ERA-xFIP, Andrew Cashner, Chase Anderson, Ervin Santana, Jose Urena, Parker Bridwell and Lance Lynn. I selected Gonzalez because he has the highest ADP of the seven. Safe to say you should avoid these other six just the same. As I was putting this article together I nearly went entirely in the Buyer Beware direction.

Stephen Strasburg, WAS
ADP SP #6, Overall #27

In 2017, Strasburg turned in arguably his best season with career lows in ERA and WHIP in 28 starts (most since 2014). Anytime someone has a “career year” you should take pause and evaluate whether or not luck played a role.

Why should you not immediately pencil him in as the #5 SP on the board? Although, Strasburg had a career best ERA (2.52) he had his career worst xFIP (3.27). Think about that for a second. His 2017 .274 BABIP was a good bit lower than his career .295 BABIP. His 8.7% HR/FB in 2017 was the 6th lowest of the 134 starting pitchers and also below his 10.9% career HR/FB.

Put it all together and you see why I’m asking you to temper your expectations for him in 2018. He’s going to be one of the better starters in any format but I’m comfortable passing on him at the top of the 3rd round and selecting someone like Noah Syndergaard, Jacob deGrom or Luis Severino later in 3rd or even taking Archer in 4th.