A lot has changed this week. The excitement of a big trade Friday came in the wake of a column on Thursday in which Jeff Passan wrote about how each team is trending, and he declared the Padres one of the seven early disappointments in a section discussion which of these teams might be the most “doomed” (i.e. not good enough to make the playoffs, nor bad enough to “earn” a high draft pick).
Passan suggests the Padres won’t make the playoffs because they can’t prevent runs. And the Padres are definitely having trouble preventing runs; coming into this weekend, the Padres were allowing 4.79 runs/game. That’s not the worst mark in the league, but it would have put them in the bottom ten last year, and that’s usually bad enough to tank a team’s playoff chances. The big question is: Will they be this bad at run prevention all year? If yes, the playoffs are probably out of reach, and the Padres might very well land in Passan’s “doomed” zone. But when you look into why the Padres haven’t been able to prevent runs it gets very interesting. It always is with the Padres. Let’s see what the numbers say.
The 2023 Padres were the second-best team in baseball at preventing runs. They allowed 4.00 runs per game, a fraction behind the MLB-leading Brewers, who allowed 3.99 runs per game. A big part of the Padres’ 2024 winning thesis was keeping that elite run prevention going. There’s reason to believe that they should be able to do this; they brought back most of their outstanding defense, and they might have upgraded a few spots: The hideous fielding of Juan Soto was replaced with the not-as-bad Jurickson Profar, and Austin Nola/Gary Sanchez was replaced with the unspectacular-but-competent Luis Campusano. The question mark was center field, where Jackson Merrill was replacing gold glover Trent Grisham, who was 93rd percentile in outs above average. Merrill hasn’t graded out as well as Trent Grisham, but the gap is small: Merrill’s outs above average has been in the 91st percentile. So, the defense should still be good. Some widely regarded defensive metrics do in fact look good. But the defensive outcomes have not been good. And you have to look into the dark recesses of baseball statistics to understand why.
As of Sunday the Padres stood third in MLB in outs above average with an estimated nine runs prevented as a result. This is evidence that they are capable of making plays a lot of defenders would not. This is evidence that there are skilled defenders in the field that have good range. But for a defense this skilled they have been failing to make the plays they should make at a higher clip than expected. The Good Play/Misplay Runs Above Average (Rgood) is not a predictive analytic so much as a descriptive one. This isn’t a metric that shows a player’s range or skill per se, it just describes the balance in runs prevented by good plays and the runs allowed through misplays including some effort based plays. A lot of the plays on the list could be considered fluky, not representative of expected future outcomes. Here’s a partial list of the 82 different play types considered for Rgood:
These are the misfit outcomes that don’t have a place in the traditional box score. We referenced Rgood last month because the Padres were the worst in MLB at that time with -3 runs above average due to Good Plays/Misplays. We speculated that this would likely get better. It has not. They’ve moved up to -4 runs on net, a mark that would have been eighth worst in all of MLB last season:
Four runs through 37 games is not a lot, but it’s part of the story. Specifically allowing runs through these non-box score plays is one way in which leaks can spring from an otherwise elite run prevention defense. And keeping track of these events can be complimentary to our interpretation of box score errors. Failing to cover a base in a situation in which it is needed is one of the misplays. And you can see this in Sunday’s loss to the Diamondbacks. Watch this play which was scored a throwing error on Jackson Merrill:
The ‘error’ here is Matt Waldron not covering home plate, he’s off to the bottom right of the screen, almost, in position. But not. He undoubtedly knows this is his responsibility. It was a fluky lapse in concentration. It probably doesn’t have predictive value about how likely it is Waldron will make that misplay again. If anything, he’s less likely to misplay it next time because of having the memory of his failure in this instance. And his lapse led to a base advancement for the opposing team. These misplays are rare and only account for fractions of a run most times but they add up.
These stats do not represent strong evidence that the defense is bad, just that through the first month of the season a lot of plays that should have been made, have not. This is to say the first month has been pretty fluky, defensively. Fernando Tatis Jr and Ha-Seong Kim, the team’s best fielders, each made four errors in the early part of the season. In contrast, last year they committed six and seven errors, respectively. They’ve since steadied the ship. Manny Machado and his 97th percentile defense spent almost all of April playing DH, and during that stretch, his fill-ins committed four errors, putting them on pace for 21 this year. Machado has never made 21 errors in a season, and the closest he’s come was his rookie year, when he made 19. This team should get better defensively just by having guys get back to playing how they usually play.
It's also not just that the Padres are making errors: They continue to be punished harshly for the errors they make. They’ve given up 20 unearned runs, third worst in baseball. The first month of the Padres’ season has been defined by an unusual amount of unusually costly errors.
There are even more hidden sources of extra opponent runs on the margins. The Padres lead major league baseball in combined wild pitches and passed balls at 21 (16 WP + 5 PB). What passed balls, wild pitches, and errors have in common is they always lead to at least one opponent base advancement, they don’t have much to do with the skill of the opponent, and they are somewhat rare events that contain a lot of noise; for example the league leader in errors or wild pitches is not necessarily the worst defender or pitcher in the league, and some passed balls don’t even have much to do with the skill of the catcher:
Courtesy: @PitchingNinja
We’ll call the combination of wild pitches, passed balls, and defensive errors the Butterfingers Quotient:
Butterfingers Quotient (BQ) = WP + PB + E
And the Padres again find themselves grouped among teams they shouldn’t be:
While the Phillies are merely a bad defensive team, the Red Sox, Marlins, and White Sox are incredibly bad defensive teams. The Padres don’t belong along side them in terms of the caliber of the average defender.
While the BQ doesn’t have much predictive power, it can have explanatory power. The Padres have gifted more than 41 free bases to opponents (errors and WP/PB with multiple runners on base lead to multiple free bases at a time). A quick way to estimate the run value of all those free bases in the Butterfingers Quotient is to presume they carry a similar run value to a stolen base, or about 0.3 runs. So by this ballpark estimate the Padres BQ has lead to an additional 12.3 runs scored (BQ * 0.3). We said above that Outs Above Average, the current standard bearer for defensive competency, estimated that the Padres saved nine runs through the plays they made that an average team wouldn’t. They’ve given away more through these fluky events that keep defining the season. This is why it’s so hard to make sense of the team. Because it is the case that they are very good at defense in the traditional sense, but fluky events have found them with alarming probability. It’s not clear there’s anything they can do about it. And it’s not likely (we hope) this this will continue much longer. Despite the lousy first month, this team’s defense is probably still good. Regression to the mean to the rescue.
Of course, defense isn’t the only part of run prevention: There’s also pitching. And, at the time of Passan’s article, the pitching had been severely underperforming. Most of that performance is down to two players: Joe Musgrove and Michael King. Through their first seven starts, Musgrove and King had surrendered 9 and 10 home runs respectively, the two worst marks in major league baseball. For some perspective, the most home runs ever given up in a season by a starting pitcher is 50 by Bert Blyleven in 1986.
Going into his start Saturday Michael King was on pace for 51.4 home runs this season, so if he stayed at this pace, he wouldn’t just be bad: He’d be historically bad. So, what are the chances of that happening? The answer that question, let’s look at the source of Michael King’s struggles this season: command issues. We’ll start with his outing against the Phillies last week.
Bryson Stott is not a formidable hitter. But he performs like one on certain pitches:
He hits like a slugger on pitches up and in. Oh, and he almost never whiffs at pitches in this part of the zone:
One way to approach a player who is not very threatening overall but is very dangerous in one part of the zone is to not throw pitches to that part of the zone. In the clip below, Campusano sets up low and away – one of the frigid parts of the Stott heat map:
But King’s command is not there:
Courtesy: @Phillies
That’s about as perfectly in the no-fly zone as a pitch could be. In Stott’s second AB, Campusano sets up high and outside on 0-2; Stott did not barrel a single pitch in that zone in either 2022 or 2023:
But again command is just not there:
Courtesy @DevineGospel
The bad process here speaks for itself: King is missing his spot, and he’s missing his spot in the wrong spot. Of course, there’s still some bad luck here: Stott’s best zone is up-and-in, but it’s not like he hits a home run every time he sees that pitch. King’s process was bad, but the outcome was worse than that bad process usually yields. King’s lack of command is causing problems elsewhere: Through his first seven starts, he was walking 5.25 batters per 9 innings pitched. Last year Blake Snell, Blake Snell led the league in BB/9 with 5.0. King’s career BB/9 is 3.2. Something was very off about King’s command the first seven starts.
Michael King looked better in his start Saturday, pitching six innings and not allowing a run. He also didn’t walk a batter, which might be a key performance indicator for his command going forward. Watching the game you could see he seemed upset when his missed his spots, suggesting that he might have been focusing on control. When he did miss his spot, the outcomes were far better than in his last outing:
Courtesy: @CFahrentold11
Both of those pitches led to good outcomes. But the first pitch was perfectly executed, while the second pitch was a pretty bad miss. King’s stuff is good enough that he can get favorable results even when he misses his spot, but he’s playing with fire every time he misses (see: The Bryson Stott outcomes above). If his control can just be not-extremely-bad, he should be much better. It still looks likely that there’s a good starting pitcher here.
Musgrove had also struggled with his command, though the real cause for concern in his first seven starts was his stuff:
He got shelled in his start against Philadelphia, giving up four home runs on pitches that weren’t necessarily bad misses with location.
But in his start on Wednesday, it was notable how his stuff was playing, velocities and spin rates up across almost every pitch:
Musgrove had reportedly worked out much more vigorously between his starts than usual. It seemed to have worked, because in addition to his spin rates and velocity being up on Wednesday, he only yielded two earned runs and struck out a season high nine over six innings. The combination of a great outcome with improved underlying pitch metrics suggested there was real reason to believe the performance was not a fluke, and that Musgrove was rounding into his usual form. This would have been the strongest evidence to date that the Passan “doomed” prediction was unlikely to come to pass. Tragically it was announced Sunday that Musgrove was placed on the 15 day injured list with Triceps tendonitis. It’s being downplayed as a precautionary move. A serious injury would be a real blow to the Padres projected ability to prevent runs across the rest of the season. Let’s hope, as Musgrove suggested to Kevin Acee, the IL stint is indeed precautionary.
Dylan Cease has been an Ace, and Yu Darvish has been his typical efficient self. But the other point of concern in the starting rotation comes from Matt Waldron. Going into Sunday he had pitched to a 4.35 ERA, better than many fifth starters. But he was completely ineffective Sunday giving up eight runs in three innings while getting almost no swing and miss and lots of loud contact. He likely would have been removed from the game much sooner had it not been for the important Chicago Cubs series starting Monday that nudged Shildt to spare the bullpen the extra work. The challenge with Waldron is he’s both a known and an unknown entity. His four seam fastball and sinker are not major league quality. We know this. The current theory of him as a pitcher relies on the effectiveness of his knuckleball, and the ability to sequence his pitches so that the fastballs play more effectively due to the drastic difference in speed an movement between the knuckleball and the hitter having to account for both. It’s a similar to the pitching thesis for Tom Cosgrove. And similar to Cosgrove, it hasn’t been working. Far and away his most effective pitch is his knuckleball. What we don’t know is the answer to a question raised on Padres Hot Tub: why is Waldron only throwing the knuckle ball 35% of the time. It may have something to do with this, though:
That’s all 196 knuckleballs he’s thrown in 2024. About 52% of them have been strikes. A couple of them look like they might kill a man. It’s not good to throw around the zone too much, but this chart looks like Waldron has a very hard time reining in the pitch. The bottom line is that currently Waldron is throwing below major league quality pitches 65% of the time. It’s hard to imagine that the results going forward are going to be a lot better with that same process. But we’ll hedge here. Pitch sequencing is one of the most poorly understood aspects of the modern game. It’s extremely difficult to forecast whether Waldron might continue the same process but expect better results as he figures out how to better sequence his offerings. And while it’s easy to believe that he could be a more effective pitcher with better command of his pitches, he’s presumable already putting maximum effort into that and hasn’t seen the results yet. Along with Musgrove’s health, Waldron’s outlook carries a lot of weight in the Padres run prevention forecast.
The bullpen, on the whole, has been league average, but 14 of the 60 total earned runs have been given up by Tom Cosgrove, who was sent down to AAA and his innings replaced by the combination of Adrian Morejon and Jeremiah Estrada. Morejon has been electric with 19 strike outs in 11.1 IP, and his stuff metrics and whiff rates are some of the best in baseball. Estrada was laying waste to hitters in the extremely offense friendly confines of El Paso, with an almost unbelievable 20 strikeouts in 8.2 innings. He’s pitched to a zero ERA through his first 6.2 IP with the Padres, racking up six strikeouts. He also has superb Statcast metrics for velocity and pitch shape, and his splitter has a 66% whiff rate. It is not insane to think that a good bullpen could take shape before the year is over.
So how should we feel about the run prevention question raised in Passan’s article? There are some really strange April outcomes weighing down the run prevention calculus. From a skill level the defense looks very good. In future months simple mean regression should allow the butterfingers quotient to normalize, and the Padres should be punished less for what errors they do commit. One of their very-bad April starters, Michael King, should be better, and the reliever who gave up 25 percent of the bullpen’s earned runs is already in AAA replaced by more promising arms. But May brought news of the Musgrove injury, and yet another outing where Waldron was relying on sub-major league quality pitches. And that’s the real reason for concern. There’s also a way forward. If Musgrove returns to form and any of Waldron, Jhony Brito, or Randy Vasquez can become a credible fifth starter, it’s very likely that the Padres will achieve the elite run prevention they appear capable of.
Maybe the best defense is a good offense. AJ Preller added a unique talent on Friday which changed the offense’s outlook significantly. We’ll analyze the move later this week. Until then: Welcome to San Diego, Luis Arraez!
Is Campusano "unspectacular-but-competent"? Fangraphs/Statcast seems to think he's literally the worst starting catcher in the league, defensively.
https://baseballsavant.mlb.com/savant-player/luis-campusano-669134?stats=statcast-r-hitting-mlb
https://www.fangraphs.com/leaders/major-league?stats=bat&lg=all&type=8&season=2024&month=0&season1=2024&ind=0&rost=&age=&filter=&players=0&sortcol=20&sortdir=asc&pos=c&team=0&qual=50&pagenum=1