Strikeouts Minus Walks is Better Than Good Enough
We have some fancy tools here at FanGraphs, but sometimes brute force works just as well. In fact, Glenn DuPaul found once that strikeouts minus walks beat the ERA predictors at their own game.
Maybe it’s no surprise. That’s the heart of the game, getting outs and keeping people off the basepaths. And maybe sometimes this other stuff is just pyrite. Just because you get a lot of ground balls doesn’t mean you won’t give up home runs. Even if you have a limited arsenal, if you can get strikeouts and limit the walks, you can have success. So focus in on the Ks and the BBs.
Let’s make a list.
To make this list, let’s look at strikeouts minus walks (in percentages, please). But let’s also strip away all the people that are already doing well (ERA < 3.4). The first number will be the player’s overall rank in K-BB, so you can place them within the 110 qualified pitchers in that stat. Then you’ll see the relevant rates, and then their ERA and BABIP. Let’s find some pitchers that are getting a little unlucky despite their good control and ability to get outs at the plate.
| Rank | Name | K% | BB% | K-BB% | BABIP | LOB% | ERA | FIP |
|---|---|---|---|---|---|---|---|---|
| 5 | Stephen Strasburg | 35.3% | 8.0% | 27.3% | 0.407 | 67.0% | 4.24 | 2.26 |
| 6 | David Price | 27.8% | 2.8% | 25.0% | 0.312 | 76.0% | 4.04 | 3.45 |
| 10 | Drew Hutchison | 29.5% | 8.0% | 21.5% | 0.364 | 82.3% | 3.46 | 3.20 |
| 18 | John Lackey | 25.2% | 5.2% | 20.0% | 0.337 | 74.3% | 4.22 | 3.65 |
| 19 | CC Sabathia | 25.9% | 5.9% | 20.0% | 0.298 | 70.6% | 4.78 | 4.27 |
| 24 | Jordan Zimmermann | 24.6% | 6.1% | 18.5% | 0.351 | 68.1% | 4.05 | 3.46 |
| 27 | Tim Lincecum | 23.5% | 5.2% | 18.3% | 0.395 | 71.4% | 5.96 | 4.72 |
| 29 | Travis Wood | 22.8% | 5.2% | 17.6% | 0.333 | 77.3% | 3.52 | 3.22 |
| 32 | Dallas Keuchel | 23.8% | 6.4% | 17.4% | 0.333 | 78.2% | 3.56 | 3.29 |
| 33 | Dan Straily | 24.1% | 6.9% | 17.2% | 0.264 | 75.4% | 5.14 | 5.30 |
| 34 | Homer Bailey | 23.8% | 6.6% | 17.2% | 0.416 | 78.5% | 6.15 | 5.37 |
| 35 | Corey Kluber | 21.4% | 4.3% | 17.1% | 0.353 | 68.0% | 3.90 | 2.59 |
| 36 | Bartolo Colon | 19.0% | 2.2% | 16.8% | 0.314 | 76.7% | 4.50 | 4.18 |
| 39 | Ian Kennedy | 23.1% | 6.6% | 16.5% | 0.265 | 68.5% | 3.60 | 2.89 |
| 40 | Phil Hughes | 21.5% | 5.0% | 16.5% | 0.353 | 64.3% | 5.14 | 3.37 |
| 41 | C.J. Wilson | 25.4% | 9.2% | 16.2% | 0.309 | 76.3% | 3.69 | 3.78 |
| 42 | Chris Archer | 19.7% | 3.9% | 15.8% | 0.344 | 66.5% | 4.11 | 2.37 |
| 43 | Zack Wheeler | 24.4% | 8.7% | 15.7% | 0.349 | 74.0% | 3.99 | 2.99 |
| 48 | Juan Nicasio | 20.0% | 5.8% | 14.2% | 0.337 | 70.6% | 5.27 | 4.59 |
Obviously not all of these sleepers are built alike. David Price and Stephen Strasburg have track records and will be hard to buy even if they are underperforming their peripherals. And strikeout and walk rates are not super stable in this sort of sample, so there’s still time for guys like Phil Hughes and Juan Nicasio to pumpkin.
But, for the most part, I’d enjoy buying a guy off this list if the price was reduced, especially Strasburg with his league-leading strikeout rate.
From a per-pitch peripheral perspective, Drew Hutchison might have something going on. His four-seamer is getting above-average whiffs for the pitch this year (10.7%, and 7.9% for career is okay too), and his slider (16% career) and change-up (18.8% this year, 12.6% career) looks average too. With his velocity slightly up (91.9 from 91.4), there’s a chance that all this playing up of his secondary pitches will hold. If he regresses to career levels, there’s a chance that he’s league average — after all, what do you get when you add a league average fastball to a league average slider and change-up? Not sure I’d drop an established veteran like Scott Kazmir for him, but Justin Masterson? Definitely Ubaldo Jimenez.
A gaggle of veterans with question marks make the list, so it’s good to give some love to CC Sabathia, John Lackey and Tim Lincecum. Throw Bartolo Colon on the list if you like. The only problem is that they are all older, and declining stuff does seem to lead to more homers. But if you limit the walks, the homers can be solo homers. So I’ll say that I actually like all of these guys in mixers still… except Tim Lincecum. This is where you have to compare their career rates to their present ones and wonder if you really believe that Tim Lincecum can almost halve his career walk rate in one year.
And then there’s a group of guys you’d figure would regress based on other factors, but who are still managing to put up good strikeout and walk rates. We’ve sussed no-name Dallas Keuchel, so you can make up your own mind there. We’ve told you to buy low on Corey Kluber, here’s more evidence. We seem to continually distrust FIP-beater Travis Wood, and he’s up to his old tricks, beating some of his peripherals (like fastball velocity) despite not changing his pitching mix very much at all. He’s another name I distrust.
I’ll end with Dan Straily because I just had a conversation with him about his velocity decline this week. He felt the guns were cold in Oakland, and pointed out that he was 90-92 in Arizona. There is some evidence he’s right, but home or away, he’s been averaging around 90 this year. On the other hand, he felt he was finally making headway with his sinker (12% whiffs) and that his slider has been the best he’s ever felt (23%). Since the change is still average (14%), that’s a good package of pitches. If the work with the sinker holds, he’s still a mixed-leaguer
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
This stat should be available on the leaderboards, imo.
Click “Export Data” and add an extra column. Fairly simple calculation, IMO.
When you export data from fangraphs it’s puts a ‘ mark before anything expressed with a percentage and while I’m no expert I haven’t found a way to get rid of it. You can’t do any calculation with it until it’s a regular number.
Select the column and do a find and remove for ‘ if that is the case. Or you can calculate K% and BB% yourself from TBF and SO/BB.
I could manually calculate ERA or FIP, too. Doesn’t mean they shouldn’t be listed on the leaderboards.
Trapped– not an expert is putting it lightly, then.
Froglegs– space scarcity! Only so much data can go on one page. It is a shame we can’t just add every stat avaiable, I agree. In regards to your remark that it should be available, well it is.
It’s not listed on the leaderboards, which was my original post. If this stat is more accurate than FIP, xFIP, and SIERA and predicting in season ERA, I’m guessing Appelman could find space for it.
Ugh… I get that. The point is that the inputs are RIGHT THERE. It is, in fact, available!
That could apply to *literally* any other stat on the site. I know it’s an easy calculation. I know the input is right there. That’s not the point. If it’s a better in-season predictor of ERA than any stat on the leaderboards, it should be on there.
trapped in a fangraph – when you import that exported data into a spreadsheet, what should pop up is a box asking you how to format the data – along with some boxes below that say something like re ‘detect special numbers’ (ie numbers with a % sign in the field). Make sure that box is checked and the spreadsheet should be able to convert the data to a number that is formatted as a percentage.
We use K-BB as a category in my dynasty league. Makes for a pretty challenging category to win; not too many schlubs can squeeze into the elite ranks of that group unless you find the one-off RP like Cody Allen that really racks up Ks.
i’ve never had this issue with the marks before when exporting. Are you using Excel?
Is there a reason Brandon McCarthy isn’t on this list? His ERA is way over 3.4, his xFIP is 2.89, and his K%-BB% is 15.2. I just picked him up in a dynasty league and I am hoping the xFIP combined with his velocity increase means that he is due for some positive regression in ERA.
This article might have been written before his 12K 1BB outing yesterday. But I’m interested in him as well. He was hitting 94 fairly regularly and even touched 95 a couple times. I’m stashing Stroman and Bauer, but is he worth cutting either of them or is he still wait-and-see? What about for Skaggs?
Those seem like good comps to me. If Bauer, Stroman, or Skaggs had been available in my league, I probably would have taken any of them over McCarthy simply because McCarthy has a significant injury history and doesn’t have a great track record and I just don’t trust any Arizona pitcher right now.
That being said, Bauer and Stroman don’t have rotation spots right now, either. I think all of these guys are #3 or #4 starters and it is really a toss up unless you believe that Bauer’s control issues are gone for good, because he seems to have the skills to be better than a #3.
So you throw out Scott Kazmir’s name like he’s droppable, but tout Dan Straily as a mixed leaguer? Kazmir has been arguably the best pitcher in the AL so far, why would anyone even consider dropping him right now?
What I am confused by is saying Kazmir is more established than Masterson when Kazmir has had basically one good year + This April. I’d still take Kazmir over Masterson, but Masterson would have the lower floor/be more “established”.
Masterson is not more established, it was a group of established guys I was trying to figure out who I would drop for Drew Hutchison. You’re right to say Kaz is not droppable, though, in this scenario. As much as I don’t buy his current walk rate and think he’ll hit the DL once or twice this season, while he’s in, he’s fine.
Eno, it looks like 2014 Travis Wood is better than 2013 Travis Wood.
Wood’s horizontal release point is lower:
http://s27.postimg.org/vm8v1xf4z/Screen_Shot_2014_04_29_at_12_21_00_PM.png
From the release point change, the cutter is moving more:
http://s27.postimg.org/suplbbgmb/Screen_Shot_2014_04_29_at_12_20_49_PM.png
He’s getting more grounders on the cutter as a result: http://s27.postimg.org/fmq7idj37/Screen_Shot_2014_04_29_at_12_20_13_PM.png
And he’s getting more whiffs on the cutters:
http://s27.postimg.org/5b99bz203/Screen_Shot_2014_04_29_at_12_20_22_PM.png
He gets above average whiffs on the four seamer (and has been for a while, for whatever reason). He has an above average cutter right now. His slider is about league average. It looks like Travis Wood improved.
I am confused. I never understood why this statistic was a good judge of a pitcher’s quality.
The ERAs on that list are below average.
These pitchers may not walk a lot of hitters, but their pitches can be hit hard.
I purposely took the guys that have high ERAs but are striking out a ton of guys and not walking guys. The game, at it’s heart, is getting outs and not allowing people on base. 100% of strikeouts don’t get on base and 100% of walks get on base. It’s an easy judge and it’s proven to be more effective than other measures when talking about in-season prediction.