The Z Files: Pitchers I Rank Lowest

I know we're supposed to keep things upbeat and cheery this time of the year, but there are some pitchers on whom I am the lowest on in the RotoWire Roundtable: Fantasy Baseball Top 300 December Update and I promised an explanation.

Just like last time with those I rank highest, the focus isn't on the raw rank, but the relative rank within the subset of starters and closers. The raw ranking also reflects some draft strategy, and the purpose of this piece is identifying the pitchers for which I am the least confident, regardless of drafting approach. We learned I have a "type" when looking at those pitchers I favor. It will be interesting to see if there are common traits among those I rank lower.

Narrator: There are.

Starting Pitchers

Snell is a tough nut to crack. His BABIP and LOB% scream regression. All of his ERA estimators are well over a run higher than actual. I know we like to use BB%, but his BB/9 was 4.95. That's more than a batter every other inning.

Narrator: And yet, he was a no-brainer for the Cy Young Award.

ARGH!!!!

I talked about Snell with some very smart friends at the recent First Pitch Arizona. One was Nick Pollack from Pitcher List. Nick showed me some compelling data, noting how Snell consistently located certain pitches in areas resulting in weak contact. The notion is he was able to pitch around dangerous hitters, confident he could retire the ensuing batters. The end-result was a higher walk rate, with less luck associated with BABIP and LOB%.

I can't dispute the data; it's quite convincing. However, I have issues with the interpretation and application. No one expects Snell to repeat what he achieved. Nick and others argue he deserved the Cy Young, because he was responsible for the outcomes, at least more than I initially believed. Sure, fine. 

If this is a skill, why doesn't anyone believe Snell can do it again next season? Everyone expressed skepticism. This is my problem. If it were truly a skill, why won't he do it again?

On the other hand, if it were just good fortune, that's an awfully long stretch of "luckily" locating pitches in the perfect spot. Nick suggested Snell just got in one of those grooves, and it lasted longer than usual. I tried to suppress a shrug, but I'm not sure I did. I never pitched competitively. Well, slo-pitch intramural softball, but never hardball.

Narrator: Nick did, quite successfully.

Expectations for 2024 illustrate how much each of us feel Snell will regress. By the rankings, I'm expecting the steepest correction. My projection essentially matched last season's xFIP, but is better than the SIERA.

There is another huge factor, apart from the luck versus skill conundrum. Throughout his career, Snell has pitched exclusively in pitcher-friendly home venues. My formulaic projection assumes a neutral park, so it could get better... or worse, depending where Snell signs. Here is the "on paper" adjustment to show the range:

Even if my colleagues are taking his numbers at face value and not adjusting for venue, I'm still the most bearish.

Narrator: Let's see if Todd changes his tune if Snell signs with Boston, the team he has been following for over 50 years

Glasnow was dealt to the Dodgers after we posted the update. The move didn't alter my rank, but it did slightly change the projection:

Initially, I expected the discrepancy to be innings, and perhaps not accounting for a replacement if Glasnow is hurt.

Narrator: Don't you mean when?

Upon further review, I don't like the projection. Looking at the little black box, I penalized Glasnow too much for his checkered injury history. Look at the strikeout rates, and the delta between the ERA estimators and actual. Glasnow's projected ERA is too high. Please chat with the narrator while I make some tweaks.

Narrator: S'up?

The newly projected ERA is still higher than the estimators, but I continue to make some allowances for the numbers being recorded over less than a full season. Even so, the new rank meshes better with the group (though not the intention).

If you're looking for "my type", it's not evident in the first two starters, but begins to show with Senga.

The narrative is Senga got better as the season progressed. My projections display slight improvements in K%, BB% and HR%, but is it enough? It's clear my LOB% and BABIP regressions helped move Senga's ERA towards the estimators, but that was based on the small uptick in skills.

Generally, I don't put too much stock in first half/second half splits. Maybe I need to soften that stance considering this was Senga's first year in MLB, and it probably took some time to adjust.

How about

This is more in line with Clay and Jeff, leaving Erik on the island.

Let's use Senga as an example of a valuation flaw. Some systems leave counting stats as decimals, while others round them off. I can argue both ways but decided on rounding since that's what the customers preferred. Dropping Senga to 11 wins lowers his rank to 21st. It's not much, but when you're on the clock, and desiring a starting pitcher, three are now ranked higher than where Senga previously landed... because of one projected win. It could even be a fraction of a win (11.4 vs. 11.6). Chances are, the lower ERA resulted in one more win, but the larger point is instead of taking the highest ranked pitcher, the better approach is considering the top four or five to be equal, and then break the tie with other factors like upside, health, etc.

Like with Senga, I am closest to Erik, with Clay and Jeff much more optimistic. Will I change this projection?

Narrator: I bet he doesn't

This is the classic example of being both good and lucky. Perez's 20.6 percent K-BB% was 18th among pitchers with at least 90 innings. If he posted that mark with enough innings to qualify, it would have been the ninth-best mark. That's good.

A .264 BABIP and 80.7 percent left on base are lucky. The ERA estimators are around a run higher than actual.

You'll note a small skills decline in strikeouts and walks, but with fewer homers. The changes are driven by MLE, albeit in a limited sample. I know Perez won't turn 21 until Tax Day next year, but I can't assume he'll match, or even better last season's K-BB%. That's simply not the weighted average of all plausible outcomes. For what it's worth, the projected K-BB% would have landed Perez 14th among qualified starters last season, which is still excellent.

Homers are the key. Even with the projected drop, Perez's projection is 1.36 HR/9 (it's more relatable in this form than 3.6 percent). The caveat is he was an extreme flyball pitcher last season, but his minor-league track record indicates more grounders. If the projection is wrong, it's skewing toward too many homers.

The problem is, homers drive ERA. The usual pattern for a flyball pitcher is a low WHIP, but a high ERA. This is precisely what I have for Perez. Lowering the homers even more is a subjective call, one I'm not willing to make.

Narrator: Told ya

My innings could be low, as Perez compiled 138 frames between the Marlins and Double-A Pensacola. This is not to say 50 more should be expected, but six more stanzas may be light. Even with an adjustment, I'll still be the low man.

This time Jeff is the next lowest. Clay has yet to be second lowest. Hmm.

Cease has been all over the place, but that's what regression and estimators are designed to handle. His ability to miss bats is clearly appealing, and why many continue to take the chance on Cease putting everything together.

It's hard to quibble with any element of the projection, except the walk rate. The projected walk rate doesn't fall in the three-year range of the other metrics. A study by our friends at BaseballHQ demonstrated a pitcher's walk rate correlates with the percentage of balls thrown. Maybe it's intuitive, but I like it when common sense is supported by data. Cease's ball percentage has consistently been well north of league average, so one can surmise he's been fortunate not to have issued more free passes. Guaranteed Rate Park does him no favors since the venue elevates walks by six percent. If Cease is traded, his walks could drop based on park factors.

Narrator: The batter's eye, foul territory and atmospheric conditions all affect walk and strikeout park factors.

In the previous installment, it was determined I am more amenable to a less dominant starter if I trust his ability to suppress ratios over a large volume of innings. If I rostered Logan Webb, I am more likely to override my rankings and pair him with Cease, but I need that buffer. 

OK, this isn't Khris Davis hitting .247 again and again and again, but it's kinda cool how the others all pegged Ragans as SP35.

Hmm, I usually regress LOB percent closer to the league mean, even if the history has been lower. Doing so drops Ragan's ERA to 3.82, but that is still no doubt higher than most expect.

The argument, of course, is Ragans was a new guy after the Royals acquired him from the Rangers and made him a starter.

Clearly, Ragans exhibited better skills, and the results followed. The BABIP and LOB percent may have been a tad lucky, but not much. Ragans starter xFIP was 3.28 as compared to 4.97 as a reliever. His velocity picked up a tad, but what about his pitch mix?

Ragans threw fewer fastballs and more sliders with the Royals. It usually helps to have a fourth pitch as a starter. Ragans is a two-time Tommy John recoveree (the first one didn't take, essentially), so maybe he was asked not to use the high torque slider.

All of this is great, but it's backwards looking. Can he do it again? Who knows? But it's clear from my ranking that I need to see it again before drafting Ragans as my SP3. Sometimes, you have to be willing to be wrong.

The type is beginning to come into focus.

The innings are low. Rodriguez tossed 165.1 combined frames between the Orioles and Triple-A Norfolk. Adding 50 is justified and brings the rank to SP33.

Based on the latest NFBC ADP, Rodriguez is SP18. The narrative is "Look what he did the second time he was called up."

Rodriguez exhibited much better control, and one can argue my projected walk rate is too high as it incorporates Rodriguez's elevated 11.2 percent mark while with Triple-A Norfolk. However, the chance Rodriguez matches the 0.38 HR/9 mark he recorded after his recall is remote.

Adjusting the walk rate, along with the innings...

Rodriguez moves up two spots with the improved walk rate, but I'm still not getting Rodriguez in an NFBC league. Softening my stance brought me in line with my colleagues, but it still lags the market. I just can't anoint him as my SP2. I'll take upside shots later. The ratios Rodriguez needs to be SP18 are not the weighted average of all plausible outcomes.

The next four names on the list are Hunter Greene, Gavin Williams, Brayan Bello, Bryan Woo. You have holiday parties to attend, and shopping to finish, so let's cut to the chase. Especially when it comes to pitching, I need more than one season, or in some cases, part of one season to essentially ignore previous performance, be it in The Show or on the farm. I'll miss out on some breakouts, but I'll avoid just as many, if not more, disappointments.

Narrator: Dude, do you get paid by the comma?

Before calling it a day, and the end to what is hopefully an enlightening series giving you a glimpse into some of the factors that drive rankings (and exhibiting the willingness to make corrections when necessary), I'd like to talk about a couple of...

Closers

Granted, it's not an efficient use of time to micro-analyze closers, but this is a significant difference of opinion.

I don't know, maybe the strikeouts are low, but Duran's swinging and called strikes both dropped last season. That doesn't preclude them from bouncing back, but it makes me uneasy, as does an uptick in walks. A high groundball rate helps keep the ball in the yard, hence a lower ERA, but it also promotes a high BABIP, hence a WHIP higher than other elite closers. Sixty-something innings may not seem like a lot, but it's enough to move WHIP a couple of points if the mark happens to fall within a tightly distributed group.

I suppose others may expect more than 33 saves, even though that would have been tied for the eighth most last season. My research shows save chances correlate best with team wins and team ERA. Currently, the top end of the Twins rotation is fine with Pablo Lopez, Joe Ryan and Bailey Ober. It gets shaky towards the back end with Chris Paddack, Louie Varland and Simeon Woods Richardson all vying for a starting gig. If Minnesota were to upgrade their rotation, I could add a save or two.

Duran is CL6 on the current NFBC ADP, ahead of Raisel Iglesias, David Bednar, Jordan Romano and Paul Sewald. I'd take any of that quartet before Duran.

This is less about Pressly, and more about planting my flag on Bryan Abreu as one of the top speculative closers. Pressly's NFBC ADP is CL12, so we're all in the same neighborhood.

With Dusty Baker calling it a career, Joe Espada is now at the Astros helm. Obviously, Pressly gets first dibs, but he just turned 35 years old, and is beginning to display health and performance concerns. There is a chance Espada isn't as loyal as Baker, with Abreu looming as a replacement.

With that, Happy Holidays. Let's make it a fun, and safe couple of weeks!