Fernando Abad and Eliminating The Cutter

"If you're lefthanded and breathing, you'll find a job." The adage is so important that I've already ingrained it on my wife, such that when we have a son in the future, we will ensure a baseball ends up in the kid's left hand. It took Neal Cotts a while to get signed, but he's also coming off a season where he allowed 12 home runs in just over 60 innings pitched. The adage is certainly true.

Someone that didn't take long to find a new home was Fernando Abad as he quickly signed a minor league contract with the Twins after being cut by the Oakland Athletics. I'm sure many of us first thought that Abad was being brought in as some competition from the south side to battle with Ryan O'Rourke as a third lefty in the 'pen. That was until the Twins signed Byung Ho Park and gave Brian Duensing's number to him before Duensing was officially released, though. After that happened, it started to look as though Abad may have a decent shot of starting the season on the Opening Day roster.

Abad has had an up-and-down career thus far. Though he started his rookie campaign with the Astros in 2010 on a high note, his sparkling 2.84 ERA was marred with a lack of strikeouts, an impossibly low .196 BABIP, and a slight issue with the home run. The strikeouts picked up over his next two seasons but his ERA skyrocketed as well, though it should be noted the Astros turned to him for six very bad starts in 2012. (His work that season as a reliever was still mediocre overall.)

Abad joined the Nationals in 2013 and enjoyed the best season of his career. He finally got his home run problem under control, he had a fair number of strikeouts and his control improved, combining to give him a 3.35 ERA and 3.26 FIP. However, he still gave up too many hits (.271 batting average allowed) and found himself out of a job the next offseason. That was when the A's picked him up and in typical Billy Beane fashion, it appeared to be a savvy pickup. A 1.57 ERA, more strikeouts, fewer hits, it looked as though Abad was starting to cement himself as a solid lefthanded reliever.

Then last season happened. His issues with walks and home runs returned. He did strike out a career-best 8.5 batters per 9 innings, but he also allowed 11 dingers in just under 50 innings pitched. In a lot of ways, his season was just like Cotts' last year.

After signing with the Twins, Abad made it known that he was going to simplify his arsenal. He attributed his struggles last season to his experimentation with a cutter, a pitch he had never thrown before in his career. Instead, he was satisfied with sticking with his four main pitches; a 4-seam and 2-seam fastball, changeup, and knuckle-curve.

At first glance, it appears as though Abad's memory was failing him. FanGraphs has pitch values (called "linear weights") that demonstrate the effectiveness of a pitch, measured in runs. Though the values for individual seasons can fluctuate wildly, the linear weight over a career almost always settles in between +2 and -2 runs. Even Clayton Kershaw with his otherworldly curve and slider sees them both check in at 1.8, though his curve has rated above +2 runs for four consecutive seasons now. Going back to Abad, FanGraphs first says that Abad threw his cutter 10.9% of the time last season with a linear weight of +1.02. This is actually quite good and right next to it we can see that his knuckle-curve rated as a -3.82 last year. Therefore, why would he get rid of an effective pitch while keeping one that was markedly inferior?

Well, if you play around with pitchF/X data enough, you learn that the technology isn't perfect. Sometimes pitches get misclassified and something caught my eye. In addition to the 10.9% cutter usage, I saw that Abad supposedly through a slider 1.1% last season and those two pitches are often mistaken by pitchF/X (a slider typically drops a few more inches but has a similar horizontal break). Furthering this evidence is that the "slider" was clocked at 86.2 MPH while his cutter was at 85.6 and sliders are supposed to be slower than cutters, so they're most likely the same pitch. But, that rare "slider" seems like it wouldn't be that big of a deal. How much negative damage could have been done 1.1% of the time last year that could outweigh a +1.02 linear weight?

Oh no. Nope. Nope nope nope. Now we're on to something. Last season, this "slider" of Abad's had a linear weight of -36.36. I italicized and bolded that to show that it was not a typo.

I did say that there are wild fluctuations in small samples here, but I meant that you'll see some positive or negative threes, not a double-digit number. If we lump in that terrible "slider" with Abad's seemingly positive cutter, now we see that he threw a cutter 12% with a linear weight of -2.4. This is likely what Abad remembers, that his cutter was getting whacked around the ballpark. With a bad season and a bad pitch, who can blame him for associating the two together?

Additionally, we can see the results from the two lumped together thanks to FanGraphs. The 99 pitches resulted in a .353/.389/.889 triple-slash with two doubles and three home runs allowed in 19 plate appearances. That does leave 80 pitches that resulted in a ball or strike, but I don't blame a pitcher for remembering those 19 more vividly than the other 80, especially when they were getting whacked all over the field.

Admittedly Abad's knuckle-curve fared poorly as well last season (.407/.429/.704) but that was a pitch he's thrown his entire career versus the cutter that was introduced for one year. I'd think that it wouldn't hurt Abad to give his cutter another shot in 2016, but I suppose I can also see why he'd want to eliminate it. Nevertheless, he's pitched well and poorly without it, so ultimately it may not even make a difference.