Chris Archer’s Last, Best Hope

Fantasy owners have been chasing, to no avail, Chris Archer’s 2015 season, during which he recorded a 29% strikeout rate with a 3.23 ERA. After finishing just outside the top-50 overall by National Fantasy Baseball Championship (NFBC) average draft position (ADP), Archer averaged the No. 50 pick from 2016 through 2018. Unfortunately, the outcomes annually and in aggregate have been awful…

Chris Archer’s Career Halves
2012-15 564.2 3.33 1.19 24.1% 8.1% 46.3% 3.36 3.47 3.51
2016-18 550.2 4.12 1.28 27.5% 7.5% 44.8% 3.64 3.44 3.54

… even though his peripherals before and after 2015 have been nearly identical. That’s the persistent problem with Archer: he has given us perpetual reason to chase results he may never again achieve.

I’m here to argue Archer’s woes started not in 2016, when his ERA ballooned to 4.02, but in 2015 — yes, his career year. That’s because he stopped throwing his sinker in 2015, opting instead to rely on a pitifully bad four-seam fastball as his primary offering. His 2015 success can be attributed primarily to his slider, which he began to feature much more prominently, but the remaining success was thanks to good luck.

In the absence of his sinker, Archer’s four-seamer allowed a 121 wRC+ in 2015 — in other words, it performed 21% worse than league-average. Yet the pitch recorded a positive pitch value — just barely positive, but still positive — despite being resoundingly below-average in aggregate to date. This speaks volumes to the good fortune he incurred on a mediocre pitch he threw more than 50% of the time.

(Had his four-seamer been unlucky, or at least true to form, we might have been able to diagnose Archer’s problems much earlier on. We might’ve taken more seriously his abandonment of his sinker in favor of an exclusively four-seam approach. Instead, we accepted Archer’s 2015 season as vintage production, and it blindsided us.)

Alas, we’ve reached the important wrinkle: after Archer was traded from Tampa Bay to Pittsburgh, he started throwing his sinker again. (Jason Rollison of Bucs Dugout discussed the potential implications here.)

The fantasy community has recently become more adept at diagnosing pitchers’ struggles. The primary fielding-independent pitching statistics (FIP, xFIP, SIERA, etc.) all assume batting average on balls in play (BABIP) and home run-to-fly ball rate (HR/FB) regress toward league-average (or, in other words, are outside the pitcher’s control). This is, or maybe was, an adequate assumption in the grand scheme of things; home runs and batting average are terribly noisy within single seasons for pitchers.

But there’s plenty of evidence pointing to varying levels of season-long success based on pitch-level skills and optimizing a pitcher’s repertoire. I would like to think I’ve contributed meaningfully to this paradigm shift by emphasizing and embracing the importance of pitch-type analysis. Considering our best pitching estimators hold constant batting average and (sometimes) home runs, I know we as an analytic community still have room to grow in regard to incorporating the nuanced effects of repertoire optimization.

I’m saying all this because: Archer’s fastball is very bad. Fastballs typically allow below-average BABIPs and can be homer-prone, which pretty soundly explains nearly all his struggles the last three years. His sinker, on the other hand, was actually pretty spectacular before he abandoned it.

I admit I’m staunchly anti-sinker; it’s usually a terrible pitch, and almost everyone should not throw one. But a select few pitchers can, with success. Blake Treinen can. James Paxton can. Archer, well, I don’t know if he can, still, but he once could and maybe should.

Because Archer threw — and, evidently, might still be able to throw — a power sinker. Among 77 pitchers who threw at least 500 sinkers in 2014, Archer’s ranked 17th in True Average (TAv). It allowed a mere .333 slugging (SLG), good for 7th among this same list, and allowed an even-better .297 SLG in 2013 — 2nd-best among 106 pitchers (min. 350 sinkers thrown). The pitch was far from a wipeout offering, but it doesn’t need to be. But it did very, very clearly limited contact in a way Archer fails to do now. Considering most other sinkers are flaming turds, Archer’s power sinker was one of a select few to give the pitch a good rap.

Now, consider the annual slugging percentages allowed by Archer’s four-seam fastball, which displaced his sinker as his primary offering, in the years since eradicating his sinker usage: .405*, .527, .506, .512. Aside from 2015 (as marked by an asterisk), Archer’s fastball has been utter garbage. The pitch is incredibly hittable, as mediocre fastballs tend to be. Velocity does not a fastball make.

It’s evident (to me, at least) the Pirates saw something — ideally, for my pride, the same something I’m seeing — in Archer’s power sinker. It allowed a .152 isolated power (ISO) and .359 BABIP in August and September last year, both far off what we saw from it half a decade ago. I would call it a small sample, though (just 226 thrown), and I might give him the benefit of the doubt having not thrown the pitch for almost five years. Looking past the subpar (but not terrible) outcomes, arguably the most important and bankable trait for his sinker would be velocity. At 94.9 mph per FanGraphs (95.24 per Brooks Baseball), the pitch mostly has the same zip it once did. Archer can’t quite reach back the way he used to; his sinker’s maximum velocity reading last year fell about a mile-per-hour short of its previous peak (according to both FanGraphs and Brooks), which might stifle the “ceiling” for the pitch. But, at first glance, it resembles the pitch it used to be.

A pitcher can only succeed on two pitches (effectively operating as a fastball-slider guy) for so long, especially when the fastball is crap. The sinker, while not a knock-’em-dead offering, keeps hitters not only honest but also off-balance, probably. At the very least, the pitch, when featured heavily, should perform no worse than his fastball; at the very most, it helps everything click for Archer, and he approaches peak form. By implementing this change, the Pirates have nothing to lose (if it fails catastrophically, simply ditch the sinker again) and everything to gain.

Let’s build in a few assumptions. If Archer throws more sinkers (displacing his four-seamer in the process), he likely accrues marginally fewer strikeouts in exchange for much-improved contact management. Given he features his slider much more prominently than he did in 2014, the strikeout rate should not dip significantly below its current rate (or, if it does, it shouldn’t fall back to 2014’s relatively mediocre levels). When synthesized, what does it all mean, ultimately? A 29% strikeout rate (like the one he achieved in 2015) is a lot to ask for, but if he can repeat last year’s strikeout and walk rates with league-average BABIP and home run rates — well, that’s exactly what Steamer is projecting, right? A 3.64 ERA with a 1.23 WHIP?

In a full season, that’s excellent. It’s not elite, like it was in 2015, again, 2015 never should’ve happened in the first place. Drafting Archer at 133rd overall (his current NFBC ADP) is, at worst, a break-even proposition given clean health. Despite how it sounds, breaking even on any draft pick ain’t bad, especially when the return on investment (ROI) of the average draft pick is well below 100%. A sinker-heavy Archer at least builds in a new ceiling, although it’s hard to envision exactly what that ceiling might look like.

But after years of tolerating lackluster results with no end in sight, what more could we reasonably ask for? For the first time in years, Archer might actually be a positive expected value (+EV) play — and, if he wins more than, like, 11 games, may actually possess top-50 upside.

Currently investigating the relationship between pitcher effectiveness and beard density. Two-time FSWA award winner, including 2018 Baseball Writer of the Year, and 5-time award finalist. Featured in Lindy's Sports' Fantasy Baseball magazine (2018, 2019). Tout Wars competitor. Biased toward a nicely rolled baseball pant.

newest oldest most voted

For 3 years now the only plus value pitch has been the slider and it appears to be in steep regression (22 to 19.5 to 6.1) as well. It takes the rosiest of glasses to see how adding another pitch that was negative value (yes small sample but negative all the same) makes him anymore valuable a pitcher. Top 50 is an extreme stretch as now he heads to the wrong side of 30 already regressing. Worth noting that in 2014 the last year of the sinker he had career low K/9 and outside of his 29 inning debut in 2012 a career high BB/9

Travis L
Travis L

I admit that I generally stay away from pitch values, as I haven’t seen much written about using them predictively. My (admittedly biased) perspective is that they seem subject to the same variances as surface level stats such as ERA.

Further being biased, I tend to believe much more in component stats than results. I am led to think if we rewound the 2016-18 seasons, and played them several million times each, we’d see Archer’s numbers much closer to his xFIP/FIP than his real world results. This is only because I don’t think 3 seasons is that big a sample to declare someone breaks the trend, although that’s more from a general skeptical approach instead of one I’ve validated thoroughly.

On that note — what pitchers have consistently beat their peripherals? Are there legitimate studies that indicate this isn’t due to normal variance? Matt Cain comes to mind as the poster boy, but he only did it until he didn’t. (And SF home park certainly helped turn some dingers into outs — I think this was in the day when we didn’t appreciate the batted ball profiles like we do now.)