How is Trevor Williams Doing It?
I don’t really know. Thanks for reading!
Trevor Williams had another great outing on Monday and now has a 0.66 ERA over his last nine starts, six of which have been scoreless outings. I mentioned Williams four starts ago back on August 14th and basically said to run far, far away. He’s allowed just two runs in 25.7 innings since then, so I totally nailed that one! Look, even an ace-level arm toting a 0.66 ERA for a nine-start run would likely have some unsustainability in his profile even with surges in his base skills. That said, this run from Williams is just wild. Let’s dive in.
First and foremost is the 8% K-BB rate for Williams during this run (and 9% on the season). Earning strikeouts and limiting walks aren’t the only ways to generate success, but the league mark for starters is 14% so anything in the single digits isn’t just below average, it’s often indicative of poor performance. There are 10 qualified starters with a sub-10% K-BB on the season and only Williams has an ERA below 4.11. Four of them sit north of five.
| Name | K% | BB% | K-BB% | ERA | WHIP | FIP | xFIP | SIERA |
|---|---|---|---|---|---|---|---|---|
| Lucas Giolito | 17% | 12% | 5% | 5.66 | 1.41 | 5.47 | 5.47 | 5.31 |
| Andrew Cashner | 15% | 10% | 6% | 4.86 | 1.53 | 5.01 | 5.06 | 5.21 |
| Clayton Richard | 16% | 9% | 7% | 5.33 | 1.38 | 4.67 | 4.23 | 4.49 |
| Reynaldo Lopez | 18% | 10% | 8% | 4.37 | 1.33 | 4.97 | 5.41 | 5.09 |
| Trevor Williams | 17% | 8% | 9% | 3.15 | 1.18 | 4.16 | 4.75 | 4.86 |
| Gio Gonzalez | 20% | 11% | 9% | 4.57 | 1.53 | 4.24 | 4.44 | 4.75 |
| Bartolo Colon | 13% | 4% | 9% | 5.45 | 1.30 | 5.30 | 4.65 | 4.80 |
| James Shields | 18% | 9% | 9% | 4.39 | 1.29 | 4.78 | 5.02 | 4.86 |
| Tyson Ross | 19% | 10% | 9% | 4.11 | 1.29 | 4.39 | 4.25 | 4.56 |
| Alex Cobb | 16% | 6% | 9% | 5.11 | 1.43 | 4.76 | 4.44 | 4.55 |
OK, so neither missing bats nor sharply mitigating walks is driving the success of Williams. I jump from there to the batted ball profile to see if he’s delivering gobs of weak, playable contact for the defense to turn into throngs of outs. That’s not happening, either. At least not as obviously as a Soft contact rate boost would suggest. His profile is almost exactly the same as it was in his first 18 starts and most of the small changes we do see would normally hurt a pitcher outside of the six point drop in HR/FB rate.
| LD% | GB% | FB% | HR/FB% | Soft% | Med% | Hard% | |
|---|---|---|---|---|---|---|---|
| First 18 | 20% | 41% | 39% | 11% | 22% | 48% | 30% |
| Last 9 | 24% | 39% | 37% | 5% | 16% | 53% | 31% |
There is one point of batted ball data where Williams stands tall, both for the season and during this run. His 85.6 mph average exit velocity is 13th among starters with 250 batted ball events and his 85.5 mph mark is 11th since July 11th (spanning his 9-start surge). So while the 16% Soft contact rate doesn’t jump off the page, he is generating a good bit of contact that the Pirates defense is turning into outs with stunning regularity. Too often we only look at the Soft and Hard contact rates for an outlier without giving any attention to the Medium contact. The league average is 46% for starters so Williams’ 53% mark during this run is well above average.
We are seeing major jumps in his BABIP and LOB rates, but without the skills to suggest any improvement, let alone a 29-point drop in BABIP to .233 and an insane 35-point surge in LOB rate to 100%. Yes, he’s stranded every runner he has allowed as the four runs yielded during this run have come from three homers.
I also looked into his pitch-by-pitch approach to both lefties and righties comparing his first 18 to these recent nine and found almost no discernible differences save a worthwhile 7-point boost in high heaters to righties, but that’s it. Williams has also seen a 302-point drop in OPS in four-seamers to righties, but those 45 PA of a .536 OPS can only have so much impact on a run like this.
So I haven’t found any major wholesale changes from Williams in these last nine starts that would generate anything close to this kind of success, especially when compared to his first 18 starts of the season. But this is also not his first big run of the year. Remember, he opened with 2.72 ERA in his first nine starts before a 7.02 in his next nine and then the run we’re currently witnessing. This sort of highlights the margin of error he’s working with as a pitcher who lacks top end velocity, spin, or secondary stuff.
At his best, he’s working the corners, going high-low with a solid four-pitch arsenal, and not giving in to batters, comfortable to issue a walk and take on the next battle instead of dropping a 91 mph meatball down the middle in a 3-ball count that could be sent 450 feet the other way. Let me be clear about one thing: I’m not here to cut down Williams’ success. He’s definitely riding hot and this run no doubt includes a large dose of unsustainable good fortune, but he is contributing to this success, too. The main reason I’ve been so fascinated by this run and wanted to do a deep-dive on it was because I wanted to see if I was missing anything(s).
I think when there isn’t a smoking gun change fostering a run of success, it’s more immediately dismissed as blind luck which is both simplistic and a bit foolish. Just because a run is unsustainable doesn’t mean it’s all happened without input from the player in question. It remains wise to bet against him as a sub-3.00 ERA guy going forward let alone the sub-1.00 guy we’ve seen since just before the break, but this is also part of what’s great about baseball. It doesn’t always go according xwOBA.
Last year, Williams posted 150 innings of a 4.07 ERA and at his core, that’s still who he is and that’s a quality major league arm. Only 35 pitchers met or exceeded both those innings and ERA thresholds last year. You have to keep putting Williams in your fantasy lineup at this point and just make sure you’re expecting a 3.75-4.25 type of arm and take anything better as a bonus.
How has the team with Williams fared in your league during the second half?
This season, I’ve found myself disregarding the Soft/Medium/Hard info on Fangraphs player cards almost entirely, unless there’s an obvious outlier in either Soft or Hard. It’s just not granular enough, now that we have Statcast data available. There’s so much ambiguity in “Medium.”
Just my $.02.
Yes, I agree and was definitely trying to underline that here. I’ve always wondered why Medium was just ignored so often. But as you say, now we have Statcast so we don’t really *need* to focus on the S/M/H rates.
Trying to figure out how you give up 4 runs on 3 HR and have a strand rate of 100%. . . . .
Pulling his game logs here on FG, it shows 100% for his last 9 games but the Aug 5th game shows as 92%. In the 5th inning of that game he gave up a HR to Matt Carpenter and then went single, single, fly out and then another single to score a run. So it’s not 100%.
Doesn’t take away from the main point that it’s unsustainable and it’s still insanely high, but not 100%. Weird that the Game Log feature here is showing 100%. . . .
LOB% isn’t an observational stat, its actually calculated according to the formula: LOB% = (H+BB+HBP-R)/(H+BB+HBP-(1.4*HR))
So since Williams gave up 3 home runs, he “should’ve” given up 4.2 runs, meaning that every run he gave up was attributed to home runs. Now, this does actually mean that a pitcher could have a LOB% above 100% (which is why Williams is at 100% for the 9 game stretch even tho he isn’t at 100% for every game) but its capped at such since it is supposed to based on reality.
I did not know this, learn something new every day!
The next place I’d look if nothing is jumping out would be the data pitch location, and movement. I wouldn’t be surprised to see the location of his pitches turn out to be a bit crisper during this run.
I checked into location — I also looked into his pitch-by-pitch approach to both lefties and righties comparing his first 18 to these recent nine and found almost no discernible differences save a worthwhile 7-point boost in high heaters to righties, but that’s it.
Sometimes luck just plays a role, but he’s been pitching instead of throwing(Giolito) most of the year.
“How has the team with Williams fared in your league during the second half?” Dammit, that explains why I can’t catch him in ERA.
He’s one of few baseball players I follow on Twitter, and while obviously he is not this good, I do think that if anyone is going to beat his peripherals it would be an extremely smart guy such like Trevor.
Also you should of ended it after the first line and we could of all responded #keepnotgraphs
lololol, I don’t think my mgmt team would dig that. maybe in the offseason we can try some shenanigans like that
I don’t know either, but man has it been fin to watch. Maybe it has to do something with a good link up between defensive positioning data and his particular pitch mix? Somehow that is somehow syncing up better than with most pitchers?
He has had 9 starts where he has allowed zero runs this year, crazy.