clock menu more-arrow no yes mobile

Filed under:

What can we learn from Spring Training stats?

Do the stats actually mean anything?

If you buy something from an SB Nation link, Vox Media may earn a commission. See our ethics statement.

Cincinnati Reds v Milwaukee Brewers Photo by Ralph Freso/Getty Images

Now that we’ve completed Spring Training and are already four games deep into the regular season, I think it’s time to try and answer the age-old question:

Do Spring Training stats mean anything in the regular season?

The easy answer: No.

The hard answer: Well, yeah, sort of, kind of.

So of course, I enter the season trying to answer a loaded question. But baseball is analyzed in every way possible. There is tons and tons of data out there to study. And Spring Training is no exception. In fact, plenty of research on this has already been done.

For instance, this article, “Spring Training Stats That Matter” by Mike Podhorzer on FanGraphs, analyzed the K% and BB% of pitchers (and in a subsequent article, hitters), and found that there was a decently strong correlation between spring K% and BB% and regular season K% and BB%. To quote his conclusion:

-Spring K% and BB% actually do mean something and may help identify breakout and bust performers for the upcoming season

-Good and bad springs carry the same level of significance and they should therefore be treated equally

-Spring ERA is completely useless

So, that’s a start. K% and BB% matter, somewhat, for both hitters and pitchers. What else is there?

This study, published by Dan Rosenheck on The Economist, dove deeper into Spring Training stats and drew a similar conclusion regarding K% and BB% and noted that stats like ERA and batting average didn’t have much value. However, he did find an interesting correlation regarding younger players:

In every peripheral category, forecasts that included a finely calibrated dose of spring-training numbers outperformed ZiPS by itself. The impact was particularly strong for first-year players (“rookies”), for whom spring training is their first taste of proper big-league competition. After adding the peripherals back together to get an all-in-one value measure, incorporating spring training improved the correlation between preseason projections and final results from .578 to .593 for hitters (using OPS) and from .354 to .387 for pitchers (using ERA).

Now, keep in mind, this isn’t going from spring OPS/ERA to regular season OPS/ERA. Rather, it’s simply making this claim:

(Preseason Projections + Spring Training Adjusments) > Preseason Projections alone

So where am I going with this? I’m going to look at the Spring Training K%’s and BB%’s for the 25 Dbacks that made the MLB roster and compare them to their ZiPS projections. From there, I’m going to see if we can spot any trends and maybe make a prediction for a boom or a bust in 2018. And we will, of course, revisit these over the course of the season to see how they turned out.

Dbacks Hitters ZiPS vs Spring Training 2018

Batter ZiPS K% ST K% Trend? ZiPS BB% ST BB% Trend?
Batter ZiPS K% ST K% Trend? ZiPS BB% ST BB% Trend?
Nick Ahmed 19.4% 22.6% Worse 5.7% 13.2% Better
Alex Avila 33.8% 38.1% Worse 16.0% 8.8% Worse
Daniel Descalso 22.0% 15.4% Better 11.1% 7.7% Worse
Jarrod Dyson 13.9% 15.5% Worse 6.9% 3.1% Worse
Paul Goldschmidt 22.4% 28.3% Worse 14.7% 11.3% Worse
Jake Lamb 26.3% 22.6% Better 11.5% 15.1% Better
Deven Marrero 28.8% 38.5% Worse 5.7% 5.5% Even
Ketel Marte 13.7% 10.2% Better 7.2% 14.3% Better
Jeff Mathis 31.1% 22.7% Better 6.0% 22.7% Better
John Ryan Murphy 18.8% 10.4% Better 6.8% 13.8% Better
Chris Owings 21.8% 10.9% Better 4.9% 9.1% Better
David Peralta 17.5% 15.3% Better 7.4% 10.9% Better
AJ Pollock 14.7% 10.8% Better 7.3% 3.6% Worse

Hitters look pretty interesting. Lots of improvements in plate discpline, especially in our younger players: Ahmed, Lamb, Marte, Owings, and Peralta. Even the non-Avila catchers (Mathis and Murphy) are joining in on the fun. I wonder if this has anything to do with us leading in BB% so far to start 2018 (MASSIVE SSS ALERT!!)?

For the batters that are doing worse, it seems to be primarily the older and/or experienced guys: Avila, Dyson, and Goldy. And Pollock just seemed to put the ball in play as much as possible.

It’ll be interesting to see if any of those improvers follows through with a legit breakout in 2018.

Dbacks Pitchers ZiPS vs Spring Training 2018

Pitcher ZiPS K% ST K% Trend? ZiPS BB% ST BB% Trend?
Pitcher ZiPS K% ST K% Trend? ZiPS BB% ST BB% Trend?
Brad Boxberger 31.7% 30.4% Worse 12.9% 4.3% Better
Archie Bradley 27.8% 24.3% Worse 9.1% 13.5% Worse
Andrew Chafin 24.3% 25.0% Better 9.8% 15.6% Worse
Patrick Corbin 20.1% 16.0% Worse 7.5% 10.0% Worse
Jorge De La Rosa 20.5% 36.7% Better 10.6% 6.7% Better
Zack Godley 24.5% 25.0% Better 9.5% 9.2% Even
Zack Greinke 23.4% 19.4% Worse 5.5% 0.0% Better
Yoshihisa Hirano 21.3% 27.8% Better 8.3% 0.0% Better
TJ McFarland 12.7% 17.1% Better 7.6% 7.3% Even
Robbie Ray 31.5% 25.0% Worse 9.9% 16.7% Worse
Fernando Salas 22.1% 17.4% Worse 7.0% 4.3% Better
Taijuan Walker 21.3% 22.1% Better 7.7% 4.4% Better

Pitchers don’t look as optimistic as our hitters. Seems to be a mixed bag. For the older guys (Greinke, De La Rosa, Salas), I don’t see much value from this. Hirano seems to be difficult as his ZiPS projection is based entirely off Nippon stats. And the other relievers (Bradley, Chafin, McFarland) just didn’t face many batters. That does leave Boxberger, who I’m only noting that his K% looked impressive in his return from injury in the past two seasons - and we’ve seen it so far twice at the MLB level with that filthy change. But again, I’m not looking much into these numbers.

So, that leaves us with the non-Greinke portion of our rotation. Godley pretty much matched his projections to a T. Walker beat them ever-so-slightly and Ray/Corbin performed worse. Not much to see though I guess I would consider myself mildly concerned about Ray’s walk rate.


So, there you have it - a projections versus Spring Training comparison of our active roster for K% and BB%. Again, I can’t reiterate this enough, this is mostly SSS noise rather than meaningful data. But it will serve as a useful tool to look back at as the season goes on. I imagine we will find some elements of truth from these differences. Maybe we will actually learn something!