2023 Fantasy Baseball Preview: Pitching Stats to Absorb and Ignore

If you’re a fantasy baseball player who focuses heavily on offseason research, you’re aware how quickly you can become inundated by numbers. It’s easy to get so bogged down in active spin, a pitcher’s unique arsenal and movement profiles that you forget to see the forest for the trees. Worse, you spend a lot of time researching pitching stats that stabilize a few starts in which — while helpful for in-season roster management — may not be the best route to take for offseason research.

Rather than spending hours rabbit-holing on pitcher stats, I think it’d be better to focus one’s research on the metrics that are “stickiest” year over year. By finding out which metrics are most reliable, we can base our research on certain foundational metrics that will give us the most bang for our buck. I enlisted the help of Kyle Bland — @blandalytics — who pulled dozens of metrics to see what we should be focusing on most when doing off-season pitching research.

Check out the full FTN Fantasy 2023 fantasy baseball draft guide! Here are five bullet points — with data and findings — that I think you should be using to better inform your research.

1. The stickiest metric year over year is GB/FB. Pitchers have shown they will largely maintain their GB/FB rates. Framber Valdez has been a groundball pitcher his entire career just as Marco Gonzalez has largely been a flyball pitcher — as a result, they’re likely to maintain their GB/FB rates moving forward. Where this becomes pertinent is where there are significant year-over-year changes. Case in point: Luis Castillo .

Since 2020, Castillo has had one of the highest GB/FB rates in baseball. Castillo’s changeup has been one of the best at inducing groundballs, and he’s kept flyballs to a minimum. That ground-ball rate plummeted in ’22, however, because Castillo made some arsenal tweaks that impacted his change, resulting in a career high in flyballs. If the stickiness of GB/FB is to be believed, there’s a good chance that Castillo’s new flyball approach to his changeup could be here to stay. Unless of course he finds a way to tweak his changeup back to what it was while maintaining the changes he made to his slider (easier said than done).

While GB/FB may be the stickiest metric, heavy variations on a year-over-year basis can point to a larger change with a pitcher be it with their arsenal or velocity. Uncovering those changes — as we’ve done with Castillo — can help us make better, more informed decisions.

Some other pitchers with distinct changes to their prior GB/FB rates: Adrian Houser , Logan Webb , Alex Cobb .

2. Of all the ERA estimators. the most accurate is SIERA. ERA, xERA, FIP, xFIP and SIERA are very common go-tos when it comes to offseason research, but should they be?

Stat YoY R2 SIERA .342 xFIP .296 xERA .203 FIP .138 ERA .074 The chart above was taken from a six-season sample and looks to see which metrics correlate best year-over-year. The higher the number, the “stickier” the metric. Of the 33 metrics that were tested for year-over-year stickiness, the fifth- least reliable was ERA. It should not be a metric you’re relying on for your research.

Knowing a pitcher is more likely to pitch to his SIERA, it would be helpful to see which pitchers in 2022 had the largest gap between SIERA and ERA. Pitchers will be statistically more likely to regress toward the former than the latter. However, if we simply did a top-10 leaderboard of SIERA - ERA, we’d get something none-too-helpful: Alek Manoah , Tony Gonsolin , Julio Urías , Justin Verlander , etc. We know it will be difficult for those pitchers to sustain their league-leading ERAs because that’s difficult for any pitcher to do year-over-year. To make our leaderboard more informative, let’s do SIERA - ERA where the SIERA was over 4.00.

Starting Pitcher SIERA - ERA Tyler Anderson 1.55 Dean Kremer 1.29 Johnny Cueto 1.23 Michael Kopech 1.19 Martín Pérez 1.19 Zack Greinke 1.16 Cal Quantrill 1.11 José Quintana 1.09 Chris Flexen 1.02 The pitchers listed above were the only SPs (min 100 IP) who had at least a one-run difference between their 2022 SIERA and ERA. While Tyler Anderson ’s changeup tweaks led to legitimate success, it’s tough to believe he’ll maintain his ERA in ‘23, especially with a new team. A lot of guys on this list aren’t strikeout dependent — which explains the higher SIERA — but that also makes them susceptible to higher ERA’s due to being so contact reliant. Add that to the lack of ERA’s stickiness year-over-year and you’re looking at a list of guys whose ERAs will likely be on the rise.

Now let’s take a look at the opposite:

Starting Pitcher SIERA - ERA Alex Wood -1.65 Lucas Giolito -1.11 Hunter Greene -1.11 Sean Manaea -1.11 This is a list of pitchers with 100 IP and a sub-4.00 SIERA who are likeliest to see a drop in ERA. At the time of my writing this article, Alex Wood is going outside of the top 300 in NFBC ADP. Considering his velocity gains and career-high BABIP (when over 100 IP), there’s a significant chance he’ll return a low- to sub-4.00 ERA with a good number of strikeouts.

It’s important to note that while the correlations above aren’t incredibly strong, we’re not looking for perfection. We’re looking for what’s going to provide us the most accurate information. While the .342 correlation from SIERA doesn’t mean it’s the be-all-end-all metric to win you leagues, we do know it should be weighted more than any other ERA estimator.

3. Stick to the whiffs Of the 33 metrics Kyle and I tested, there were a significant amount dedicated to swing-and-miss. We looked at CSW%, overall Swing%, O-Swing% and more. We even got so granular as to look at swinging strike on the edges and middle-middle CSW. We found that three metrics were the stickiest overall: Swinging Strike% (SwSt), In-Zone SwSt and K%.

If we make a leaderboard of who dominated in these respective categories, we’ll see familiar names: Shane McClanahan , Spencer Strider , Dylan Cease , Shohei Ohtani . While there is one name that sticks out as pertinent from the top 10 in Hunter Greene — the Reds rookie made significant strides with his fastball in the second half — let’s dig a little deeper and use this information to find guys who could take the next step.

Starting Pitcher SwSt% In-Zone SwSt% K% Drew Smyly 12.4 13.5 20.4 Tyler Wells 11.3 13.2 18.0 Daniel Lynch 11.3 13.0 20.3 Ian Anderson 12.3 12.9 19.7 Drew Rasmussen 12.1 12.4 21.4 Tyler Anderson 11.7 11.8 19.2 All of these pitchers had above average SwSt% and In-Zone SwSt% but below average K%. Seeing as SwSt% is a little stickier than K% we can posit that there’s a chance these guys could turn more of their whiffs into strikeouts coming into next year.

4. Velocity is king Something happens at the beginning of every year that impacts the fantasy community. We’ve done our research, our spreadsheets have been made, debated and agonized over, we’re ready to start our drafts and what do we hear? Spring training news. “Pitcher X mentioned he’d like to throw his curveball more.” “Pitcher Y has added some velocity.” “Pitcher Z is tinkering with a new pitch.” Just like that, we’re thrown into disarray.

I love talking to members of front offices about baseball because their relationship to metrics is completely different than fantasy players. They don’t care as much about raw spin totals — neither should we, to be honest — or WHIP because they’re looking at things from a different perspective. When I asked a club-employed friend of mine what he thought was stickiest year over year his response was: "Velo.. When I said what else he said, "Velo."

Let that lone word be the piece of advice that allows you to cut through all the other noise. The only thing that you should really give weight to out of the incessant spring training reports is who added velocity.

Some of the pitchers who I could confirm had increased velocity coming into/during ‘22’s spring training were: Mitch Keller , Alex Cobb , Max Fried and Joe Ryan . By ERA, FIP and SIERA standards, literally all four of those pitchers had better seasons in ‘22 than ‘21 (if qualified). This will of course not always be the case. Sometimes velocity wears off as it did for Max Fried ; Lucas Giolito was rumored to have increased velocity and then ended up losing velocity over the course of the year; Tylor Megill gained velocity but had an inflated ERA and an injury shortened season. However, it is a solid piece of news to pay attention to.

All other spring training news should be categorized as noise. It doesn’t matter if a pitcher has a new pitch (remember Tyler Glasnow ’s changeup?) because we don’t know if he will want to throw it or if his pitching coach will want him to throw it or if it will even be good. It doesn’t matter if a pitcher is interested in mixing in a new pitch more because that might not be a good thing and there’s no concrete evidence that it will indeed happen. Velo however: now that’s far more concrete and it’s all we should focus on.

5. You can ignore HR/FB Of the 33 metrics we tested, here are the five “least” sticky.

Stat YoY R2 ERA .074 LOB% .060 HR/9 .041 BABIP .012 HR/FB 0 That’s not a typo. That’s a zero.

There are a lot of reasons that this could be the case. It could be because the sample size is smaller than others (this is taken from six seasons of data, and while that may seem large, there needs to be a pitch, swing, contact, fly ball and home run). It could be because pitchers have more control over launch angle than exit velocity. Or it could be because we’re fans of the only sport in the world that decides to change the ball it’s played with year after year after year. Personally, I’m going with the latter.

Whatever the reason may be, the data show we should not be relying on HR/FB in our offseason starting pitcher research. This is very good news if you’ve rostered Corbin Burnes in a dynasty league, as it means that ghastly 14% HR/FB rate is likely not here to stay. On the other hand, it doesn’t mean that Framber Valdez is likely going to sustain the career low 12.5% HR/FB rate he had in ‘22.

If you’re interested in looking at a contact metric that’s more predictive for pitchers, you’re better of looking at O-Contact% and overall Contact%.

We’re in the midst of a baseball analytics renaissance, where it feels like every day we’re uncovering new metrics for us to research and discuss with our fellow fantasy baseball fans. It’s important to remember the point of this research isn’t to make sure we’re right; that’s impossible. It’s to make sure we’re making the most informed decisions we can in an effort to win our fantasy leagues. Hopefully the above tips can help get you that much closer to hoisting a trophy. Good luck this season.

2023 Fantasy Baseball Preview: Pitching Stats to Absorb and Ignore

This feature is only available to FTN Fantasy Premium Members Only