2023 Magic Formula Qualifiers - Pitchers
I go through each of my "magic formulas", explain what they are and why they matter, and then examine the players who qualified for each.
Check out the hitters version of this post here.
We have two magic formulas to cover here on pitchers. My first one is pretty elementary, because it mostly uses K% and BB% - which is nothing new and nothing you won’t see every other half-decent baseball writer citing. The lesson common thing that I have thrown into the mix is GB%. Not as many people talk about liking pitchers with higher ground ball rates, but I sure do.
Magic Formula #1: K% vs. BB% vs. GB%
Two quick and ugly scatter plots to get started, just to drive home the point once again about K-BB%. When you’re talking about ERA indicators like FIP or SIERA, you’re mostly talking about K-BB%. The main inputs into those regressions are strikeouts and walks, so we can really simplify things just by showing how K-BB% does a nice job all by itself predicting ERA:
This is a one-year sample, and ERA is very noisy, so there will be outliers, but you can see the downward trend. From 2023, here are the average ERAs of groups of pitchers based on their K-BB%
K-BB% less than 10%: 5.19
K-BB% between 10%-15%: 4.53
K-BB% between 15-20%: 3.97
K-BB% greater than 20%: 3.69
GB% all by itself tells you basically nothing about ERA, as evidenced here:
But let’s zoom in for a moment. If we look at just the pitchers with at least a 15% K-BB% (decent to very good pitchers), we do see the high GB% doing better in ERA than the low GB% pitchers. Among these pitchers:
Sample with a GB% above 47%: 3.59 ERA
Sample with GB% below 43%: 4.10 ERA
So GB% by itself - no good. The real goal of pitching is to not allow a ball to be put in play at all. But if you’re going to allow a ball in play, it’s very much preferable for that ball to be hit on the ground. To make my point further, here are hitter stat results on each batted ball type:
GB: .248 average, .273 slugging
LD: .644 average, .910 slugging
FB: .273 average, .845 slugging
It’s pretty clear there. All home runs and almost all extra base hits come from line drives and fly balls, so keeping the ball on the ground will certainly keep runs off the board much better than allowing balls to be hit in the air.
All of that said, I would much rather have a high K-BB% pitcher with a low GB% than a high GB% pitcher with a low K-BB% - so there is some weighting needed here.
Let’s get to the formula results and talk about some names. Later on, we’ll get to our newer magic formula of SwStr% vs. Ball%, which is one I’m really liking and leaning on more and more. I’ll talk about that in full and give you some really interesting names below, but you’ll have to be paid subscriber to get the rest of this piece. Sign up today, check out the about page here for more details.
We are below the paywall now, shout out to you for paying me! And here’s your interactive scatter plot for this part of the post, it data from 2023 only.
When we have the time to get deeper into the numbers, we should do that. And I do that below, but first, for old time’s sake, let’s just give the list of all the SPs that met the given threshold of
K% > 27%
BB% < 7%
GB% > 45%
Well whoopsie-freaking-daisy, Pablo Lopez is the only qualifier (29% K%, 6% BB%, 45% GB%).
Let’s loosen everything
K% > 25%
BB% < 8%
GB% > 42%
Zach Eflin, Pablo Lopez, Tyler Glasnow, Kevin Gausman, Griffin Canning, Zac Gallen, Aaron Nola, Freddy Peralta, Kyle Bradish, Mitch Keller, Clayton Kershaw
So there you go. If you want to do more threshold stuff on your own, check out that dashboard or download the Excel file I link to below.
One thing mathematical trick I have have been using over the years is to take a few different percentiles and then average them together to get one number to represent multiple categories. We can do that here with K%, BB% and GB%. Find the percentile for each category for each pitcher and then average them together and see who comes out on top.
If we do that, Zach Eflin is actually your top dog here.
26.5% K% (84th percentile)
3.3% BB% (99th percentile)
50.4% GB% (90th percentile)
And that’s a 91st percentile average to put him at the top of the list. The rest of the top 10 when doing it this way:
Zach Eflin (24.2%, 3.3%, 50.4%)
Cristopher Sanchez (24.2%, 4.0%, 57.7%
Pablo Lopez (29.2%, 5.9%, 45.4%)
Logan Webb (22.85, 3.6%, 62.7%)
Justin Steele (24.5%, 5.0%, 50.3%)
Braxton Garrett (23.7%, 4.4%, 49.0%)
Tyler Glasnow (33.4%, 7.6%, 50.9%)
Kyle Bradish (25.0%, 6.6%, 49.1%)
Joe Musgrove (24.4%, 5.0%, 46.0%
Framber Valdez (24.8%, 7.1%, 55.2%)
But any top ten list that does not include Spencer Strider or Gerrit Cole you should either majorly adjust or outright throw away. K% is much more important than GB% here. Take an extreme example. If a pitcher struck out 90% of the hitters he faced, it would not matter at all what happened on those few balls in play, you wouldn’t care one bit what his GB% was.
So let’s do some weighting here. We’ll weigh
K% x 5
BB% x 3
GB% x 1
This is probably still not weighing K% enough, but here’s your top 20 when we do that:
Eflin
Lopez
Glasnow
Wheeler
Ryan
Cole
Sale
Gausman
Sanchez
Canning
Gallen
Steele
Maeda
Castillo
Strider
Garrett
Nola
Peralta
Gilbert
Musgrove
Crappy screenshot here if you want the details:
If you want to check out that Google Sheet, here you go.
Automatic ranking is a really tough thing to do with pitchers. I’m certainly not going to draft Zach Eflin ahead of Cole, Strider, Glasnow, Wheeler, etc. this year just because he pops up here. This stuff takes manual input and thought.
Strider’s K-BB% was 29.4%, that’s lapping the field and it makes me really not worried at all about the 11th-percentile GB% he put up (although it’s not like it doesn’t matter at all - his 3.86 ERA was much higher than you’d expect for a 29% K-BB%, and that’s almost certainly because hitters were able to get balls over the fence against him when he was allowing contact).
The pitchers that most stand out to me here:
Tyler Glasnow. He’s pretty close to Strider in K% and BB%, and he gets an above-average amount of ground balls while he’s at it. If he stays healthy, I think he’ll be right there with Strider in the Cy Young race, and he’s usually the 8th-10th SP off the board - a much better price.
Zach Eflin. He’s at the top of the list, so we have to highlight him. His career 21% K% does throw the 26.5% spike year into question, but he was with Tampa Bay and his SwStr% made a jump upward as well. He’s also only 29 years old, so there’s some value in this price (SP #26 in the average draft right now).
Cristopher Sanchez. He is the Phillies #5 SP right now on Roster Resource, so there are question marks about his job. But if he’s in the rotation and being used as a traditional starter, I’m very interested given his output here. He put up a 62nd percentile K%, a 96th percentile BB%, and a 99th percentile GB% in 2023 - very exciting stuff.
Nick Pivetta. I really want to hesitate on Pivetta because he’s been around for so long and has never really turned into a strong and consistent SP, but there’s no denying how great he was last year. Only eight pitchers managed a 30%+ K% in 15+ starts, and while there was some help from some bullpen stints there from Pivetta, he did rip off a 33% K% and a 4.3% BB% in his final five starts against the Rays, Red Sox (twice), Blue Jays, and Orioles. Very impressive stuff there. I expect some serious regression from him in 2024, but at the very least he seems to be a solid SP.
Justin Steele. I’m always been hesitant on the guy because he hasn’t ever gotten a ton of whiffs and none of his pitches actually grade out well, but now he’s been very good for two full seasons, and that’s about when I start accepting that a guy might be an outlier. And to his credit, he does grade out well in this trio of stats with a 24.5% K%, a 5.0% BB%, and a 50.3% GB%. None of those (besides the BB% maybe) pops out individually, but the combination of them turns him into a guy that is tough to score on.
Griffin Canning. I mean, of course. 26.0% K%, 6.5% BB%, and a 43% GB%. And that isn’t even to mention the SwStr%, which we’ll see below. Canning is going to be the poster boy of this blog this year, so feel free to judge everything I say from here forth on how he performs in 2024. My flag has been planted.
Let’s wrap this up, I never want these posts to be overly long. But before we move on, I’ll just give you the top 10 in each category.
K%
Strider 36.9%
Glasnow 33.4
Ohtani 31.5%
Snell 31.5%
Pivetta 31.2%
Gausman 31.1%
Peralta 30.9%
Greene 30.6%
Sale 29.5%
Ryan 29.5%
BB%
Kirby 2.4%
Eflin 3.3%
Webb 3.6%
Greinke 3.9%
Sanchez 4.0%
Garrett 4.4%
Mikolas 4.5%
Gilbert 4.6%
Hendricks 4.7%
Bryce Miller 4.8%
GB%
Webb 62.7%
Cobb 57.8%
Sanchez 57.7%
Stroman 57.4%
Bello 56.3%
Cabrera 55.7%
Peterson 55.5%
Valdez 55.2%
Houck 53.3%
Brown 52.2%
Magic Formula #2: SwStr% vs. Ball%
Defining these stats:
SwStr% = Whiffs Generated / Pitches Thrown
Ball% = Balls Called / Pitches Thrown
Note that this is balls, not pitches out of the zone. Foul balls count as strikes, so these are only pitches that result in another ball being added to the count.
Of course, we want high SwStr% and low Ball%. Some distribution explanations first so you know what we’re looking for:
SwStr%
25th percentile: 10.3%
50th percentile: 11.8%
75th percentile: 13.4%
90th percentile: 14.6%Ball%
25th percentile: 34.2%
50th percentile: 35.8%
75th percentile: 37.1%
90th percentile: 38.7%
If you want really, really simple rules we want to see SwStr% above 13% and Ball% below 35%. The Ball% distribution is really tight, meaning even a half of a point difference is significant.
The Ball% king was George Kirby at 29.1%, and the worst in the league was Michael Kopech at 41.8%. So that’s a range of 12.7 points. The ranger of SwStr% goes from Spencer Strider at 20.5% down to Adam Wainwright at 5.4%.
Here’s your interactive plot.
And here’s the screenshot:
Let’s check a few combinations:
SwStr% >= 14% & Ball% <= 34%
Spencer Strider
Shane McClanahan
Tarik Skubal
Luis Castillo
Pablo Lopez
Domingo German
Bailey Ober
Joe Ryan
Max Scherzer
Bryan Woo
Kevin Gausman
Kutter Crawford
Brandon Woodruff
The quite interesting thing here is that there are a few cheap SPs on the list, and they aren’t guys that proven to be mediocre or poor pitchers, they are just guys we haven’t seen a lot of.
Bailey Ober. He made 26 starts and put up a 15.2% SwStr% with a 33.2% Ball%. That turned into a middling 25% K%, but a strong 3.43 ERA and a 1.07 WHIP. He was good, and right now he seems locked into an MLB rotation job for 2024. He doesn’t have dominant stuff, but the guy is practically seven feet tall, and that kind of extension and arm slot can go a long way to keeping hitters confused.
Bryan Woo. There are at least four SPs ahead of him on the depth chart for 2024, the M’s are pretty loaded in that regard. But he’s got a shot to claim the fifth spot. He threw 88 Major League innings last year with a mediocre 4.21 ERA and 1.21 WHIP, giving up a high HR/9 of 1.3. But the impressive part was the 14.3% SwStr5 and 33.9% Ball%, and that turned into a solid enough 17% K-BB%. I think his job security is probably fine, I don’t see anybody on the roster that would make any more sense than Woo, and it wouldn’t seem likely for the M’s to go attack SP in the market given their current strength.
Kutter Crawford. This is another guy that gets it done with his delivery. He has like a hitch thing going on, which most hitters have never seen before, and that probably helps with deception. Regardless of how he did it, he put up a strong 14.2% SwStr% and 34.0% Ball% to qualify here. That turned into a 19% K-BB%, really good as well. The end result was a 4.04 ERA and a 1.10 WHIP, so all of these numbers are pretty darn good. The issue is that he doesn’t have a locked in job in the rotation. He’s currently the #5 on roster resource, but guys like Garrett Whitlock and other youngsters might have something to say about that.
As for the prices on these guys, we can hit up the ADP Dashboard and run a comparison:
So we can grab all three after pick 150 most of the time, and that’s something I’m going to be doing a lot in drafts this year.
Let’s back off the criteria a bit and find more names of interest. If we go to a SwStr% above 13% and a Ball% below 35.5%, I pick out these names:
Griffin Canning (of course)
Carlos Rodon
James Paxton
Jon Gray
Michael Grove
There are a bunch more names I left out because they’re already names you know about and are being drafted where they should be.
Rodon and Paxton are known commodities and it’s really just a question about health with them, so it’s fair to not want to take on that risk at all. Michael Grove also put up a solid 13.4% SwStr% and 34.1% Ball% last year on his way to a 24.2% K% and 6.3% BB% (18% K-BB%). For your splits-truthers, he went for an elite 28% K-BB% in his four starts in the second half.
And I’ve already said enough about Griffin Canning.
The more I research and write, it seems pretty clear to me that I want to build most of my pitching staff after pick 100. My general strategy right now is to take at least two hitters first, but to get one of these top SPs in the first four or five rounds. That puts me at the end of the SP1 range, I don’t think I’ll end up with any Strider, Cole, Burnes, Wheeler, Castillo, or Gausman. I will probably pull the trigger on one of these guys as my ace:
Glasnow
Lopez
Nola
Skubal
And then maybe I go hitter heavy again to fill out the offense before really going hard after SPs in the 100-250 range, here are some names I’m excited to draft right now along with their ADPs.
Hunter Greene (134)
Jordan Montgomery (142)
Gavin Williams (150)
Bailey Ober (166)
Cristian Javier (169)
Shane Bieber (178)
Nick Pivetta (183)
Bryce Miller (188)
Braxton Garrett (189)
Bryan Woo (190)
Brandon Pfaadt (214)
Triston McKenzie (239)
Cristopher Sanchez (244)
Nick Lodolo (248)
Emmet Sheehan (251)
Kenta Maeda (258)
Kutter Crawford (273)
Nestor Cortes (279)
Lance Lynn (299)
Jameson Taillon (301)
MacKenzie Gore (311)
I didn’t really mean for this to turn into an SP strategy piece, and there will certainly be another one of those coming soon - but there you go we’re ahead of schedule.
I’m going to really focus on the getting the team-by-team previews and my ranks done before doing too many more extra pieces like this, but I intend to go heavy on this kind of content in February and March. Thanks for coming along for the ride!