(Dylan Buell/Getty Images)
There’s a saying that goes: “sometimes it’s better to be lucky than good.” Did you know that Lefty Gomez, a Hall of Fame pitcher who played for the Yankees and Washington Senators between 1930 and 1943, was the first one to say that phrase? Gomez knew it back then and now we have the numbers to prove it: luck is deeply rooted in the game of baseball.
One player this season that has been very good, but extremely unlucky is Giants starting pitcher, Jeff Samardzija. On the surface, it looks as if Samardzija has been struggling on the mound this year. He’s 2-7 with a 4.29 ERA, but basic statistics do not show the full picture.
One of the most widely used advanced stats to evaluate a pitcher’s performance is FIP (Fielder Independent Pitching), which measures a player’s ERA if they “experienced league average results on balls in play and league average timing", and xFIP, which is calculated the same as FIP except with an estimate of how many home runs they should have allowed. These two stats are on an ERA scale, so you can use them the same way you would use a pitcher’s ERA.
I mentioned earlier that Samardzija has an ERA of 4.29 this season. However, he has a FIP of only 2.89, which is good enough for fourth in the MLB, behind only Chris Sale, Stephen Strasburg, and Chris Archer. Samardzija also has an xFIP of 2.72, which is fourth in the MLB behind only Sale, Lance McCullers, and Clayton Kershaw. These numbers put Samardzija in the company of some of the game's best pitchers. This also gives him one of the largest E-F (difference between ERA and FIP) numbers in the MLB at 1.48, which shows that he is pitching significantly better than his ERA suggests.
Jeff Samardzija is also having a career year in terms of strikeouts and walks. He is striking out 10.62 batters per nine innings, which is the highest mark of his career and seventh in the MLB. He is also walking only 1.24 batters per nine innings, which is the lowest mark of his career and the fourth lowest in the MLB. This gives Samardzija the second highest K/BB mark in the MLB at 8.55 and the third largest difference between his K% and BB% in the MLB at 25.5%, behind only Sale and Max Scherzer and ahead of Kershaw and Zack Greinke.
The above statistics show why Samardzija has such a low FIP and xFIP, but what is keeping his ERA up at 4.29? There are two factors that are elevating his ERA. The first is his home run rate. Per nine innings, Samardzija is allowing 1.13 home runs. Now, compared to the rest of the MLB, this is not a very high number— this is the 51st highest rate in the MLB— but it is one of the highest marks of his career.
The thing that is really holding Samardzija’s ERA up at 4.29 is… bad luck. Samardzija has been putting up career numbers and ranks in the top five in many advanced analytics, but is still only 2-7 with an awful ERA. I guess Lefty Gomez was right when he said “I’d rather be lucky than good.”
First, although Samardzija has allowed more home runs this season than average over his career, some of those home runs were the product of bad luck. Below is a chart from Baseball Savant that shows all ten home runs he has allowed this season with distance, exit velocity, launch angle, and the expected home run percentage:
As you can see, some of his home runs would have left the park in almost any setting. However, there are several that may have left the park due to some bad luck as there are four cases in which the expected HR% was below 50%. This shows that with better luck, Samardzija’s ERA would be significantly better and more in-line with his other numbers.
However, Samardzija may have gotten lucky on a few occasions in which an out or a base hit may have yielded expected home run stats. I used Baseball Savant to find non-home runs where the expected ISO (isolated power) was greater than or equal to .750 because that is the point at which half of the batted balls are home runs. Here are those cases:
Rene Rivera hit an RBI double that could have easily been a home run, but for the most part, all of these ‘close-calls’ had a low expected home run percentage. Based off of the above charts, with average luck, Samardzija should have allowed a lower amount of home runs, which would be reflected in his ERA, lowering it to a much more respectable number.
Allowing home runs is not the only category in which Samardzija has faced bad luck. Hitters have a BABIP (batting average on balls in play) of .340 against him. This is insanely high. Average hitters will typically have a BABIP of .300, and Samardzija’s .340 is the eighth highest in baseball. BABIP is dictated by talent level, fielding, and luck. Samardzija clearly has a ton of talent and whether it is due to poor defense or luck, neither one of those factors illustrate how Samardzija has pitched this season. If he had a BABIP closer to the average of .300, his ERA would decline immensely.
The last luck factor that has diminished Samardzija’s numbers is left on base percentage, or LOB%. Good pitchers will usually have a high LOB% because they are able to get out of jams, and the same thing goes for a strikeout pitcher, which Samardzija is. So the expectation is that he would have a LOB% somewhere above the expected average of 72%, but that is not the case. Samardzija has a LOB% of only 66.7%, which is the eighth lowest mark in the MLB. This means that he is allowing a lot of runners to score, a lot more than you would expect from a pitcher with high strikeouts and low walks.
Not only does the low LOB% illustrate the bad luck that Samardzija has faced, but it also shows that he should see a jump in statistics the rest of the season. So do the bad luck on home runs and the high BABIP. If Samardzija had experienced average luck this season, with the way he has pitched, he would definitely find himself in the Cy Young discussion.