Paul Goldschmidt Is Surging
When the Cardinals traded for Paul Goldschmidt this offseason, they added one of the most consistent and potent hitters in all of baseball to a team sorely in need of a jolt. As players go, Goldschmidt was about as safe a bet as there is. From 2013 to 2018, he had posted a wRC+ between 133 and 163 every season. His wasnāt a story of constant reinvention and tinkering: he was basically the same hitter every year. He walked a lot, struck out a lot, hit for power, and ran a high BABIP through a combination of his surprising speed and consistently above-average line drive rate.
If thatās what the Cardinals thought they were adding to the lineup with Goldschmidt, the early returns were disappointing. Fresh off of signing a five-year extension, Goldschmidt scuffled through the first months of the season. After starting off the season strong with a three-dinger game in his second game in a Cardinals uniform, he put up some alarmingly pedestrian numbers. He ran a 123 wRC+ for March and April, not up to his usual standards, and it went downhill from there. He declined to a 104 wRC+ in May and a shocking 57 wRC+ in June.
Alarmingly, it didnāt look like luck was to blame. Goldschmidtās .302 BABIP was below his career average, but not concerningly so. His strikeouts were up a hair and his walks were down perhaps two hairs from his Arizona form, but nothing about that screamed regression. No, Goldschmidtās problems boiled down primarily to one thing: he stopped hitting for power. In his last six seasons with the Diamondbacks, Goldschmidt had posted an ISO in the top 20 in baseball five times. The one year he didnāt, he propped up his value with a whopping 32 steals and career-best plate discipline.
With half of the 2019 season in the books, Goldschmidtās ISO was below league average, leading to a 97 wRC+. Not just outside the top 20, not just below .200 — it was a puny .159, smack dab between Amed Rosario and Nick Ahmed. As for propping up his value with steals and plate discipline, he had zero steals and the worst K-BB% since his rookie year. Add it all up, and heād been worth 0.7 WAR, less value than heād accrued in his average *month* with the Diamondbacks.
When a player declines like this, theories for what has changed abound. Goldschmidt had a slow start to 2018 as well, and back then, Jay Jaffe surmised that Goldschmidtās difficulties stemmed from trouble dealing with velocity. This year, opinions were more varied. Maybe he had a patience problem, or a groundball problem. Maybe he got too streaky. Regardless of the cause, it seemed like time to worry.
Of course, the season doesnāt end after June 30. Admittedly, the only reason I gave you statistics through that date was to make the transition more dramatic. Transition to what, you ask? Since June 30, Goldschmidt has been, well, heās been Paul Goldschmidt. Heās batting .302/.350/.635, good for a 150 wRC+. His walks havenāt rebounded, but his power has returned with a vengeance (.333 ISO), and his 12 home runs have him batting above .300 with only a .313 BABIP. This version of Goldschmidt isnāt quite the same as his Arizona peak, but heās providing roughly the same value at the plate.
When a change this dramatic occurs, there are four stories we can tell about it. One, maybe he was getting unlucky before, and his true talent level is what heās currently displaying. Two, maybe heās getting lucky now, and the earlier ācoldā streak is the right level. Three, which is the most common story, is that he was getting unlucky before and lucky now. Four, maybe something changed.
Letās cover whether he was getting unlucky first. In a very narrow sense, we could just look at xwOBA, the Statcast metric that approximates wOBA by looking at batted ball data. Using this as our guide, Goldschmidt was indeed getting unlucky — but just barely. He compiled a .323 wOBA and .343 xwOBA through June 30. He was doing worse than he should be, but not by much. That .343 xwOBA works out to around a 112 wRC+ after accounting for park effects.
Nothing looked particularly strange in the more granular metrics, either. Goldschmidt was indeed less patient, swinging at 45.4% of pitches he saw, on pace for a career high. He was making contact with pitches in the strike zone a career-low 79.5% of the time. The walks and strikeouts, in other words, were deserved. He pulled grounders roughly in line with his career average, and turned fly balls into home runs 17% of the time, a disappointing but still above-average rate. Add exit velocity to the home run data, and it looked roughly like he deserved his home run rate — nothing lucky or unlucky to see here, in other words.
Okay, so Goldschmidt wasnāt getting particularly unlucky in the first half of the season. Has he been getting incredibly lucky since? By our standby wOBA/xwOBA comparison, it doesnāt look like it. Goldschmidt has hit to the tune of a .407 wOBA since July 1st, basically what his .401 xwOBA suggests is fair. His plate discipline looks reasonable as well. Heās been swinging at absolutely everything — his O-Swing% is 36.9%, a huge outlier over the course of his career that explains the decreased walks:
Some of the increased production, of course, comes down to a spicy 27.9% HR/FB rate. That rate has increased markedly from earlier in the year, driven largely by hitting the ball harder and more consistently — heās barreled up a quarter of the balls heās put in the air since July 1st, as compared to only 15% before then. Heās pulling the ball more in the air as well, 35% of the time against only 25% before July 1. Pulled fly balls and line drives are the best way to hit for power, and 10 of his 18 extra-base hits over this period have been pulled.
Without real evidence that Goldschmidt was getting tremendously unlucky over the first part of the season or lucky over the second part, weāre left looking for a change. One thing stands out. His batted ball distribution has changed massively:
| Statistic | Through 6/30 | Since 6/30 |
|---|---|---|
| GB% | 43.4% | 33.7% |
| LD% | 19.6% | 21.1% |
| FB% | 37.0% | 45.3% |
| GB/FB | 1.17 | 0.74 |
| Pull% | 33.3% | 43.2% |
| Cent% | 42.0% | 37.9% |
| Oppo% | 24.7% | 19.0% |
| HR/FB% | 17.3% | 27.9% |
Goldschmidt has never been a fly ball revolution guy; his GB/FB ratio has never been below one for a full season. His swing is anachronistically level, designed to work with what the pitcher gives him, which means that heās generated most of his power on high pitches:
Thatās changed. His power during his recent surge has come almost exclusively from the middle of the strike zone:
Want a demonstration of how heās tapped into this power? Take a look at two pictures of Goldschmidt facing fastballs middle-in:
One of these balls became a weak grounder. The other was hit for a home run. Iāll save you the guesswork. Houserās pitch, which occurred in April, became an out, while Darvishās in late July became a souvenir. Notice anything different? To me, one thing jumps out. Goldschmidt is more crouched and coiled against Darvish, while he appears almost folded over against Houser. Watch them in real time, and the difference becomes more stark. Against Houser, Goldschmidt starts in a crouch before seemingly popping halfway out of it mid-delivery:
Against Darvish, he stays in his stance, smoothly transferring forward into his swing:
This is a tiny change, something Iām only seeing because I want to look for a reason Goldschmidt has put more balls in the air. But small changes add up. Goldschmidt has put nearly 65% of balls thrown in the lower two-thirds of the strike zone into the air since July 1, significantly higher than his career rate of 56%. Heās putting 80% of balls on the inner third of the plate in the air over that same span, and heās started swinging at those inside pitches far more. Early in the season, he leaned inside and high but had a balanced swing profile:
Now, rather than attack everything equally, heās looking further and further inside:
Of course, swinging at inside pitches comes with downsides — look no further than all the swings off the inner edge of the plate. For now, Goldschmidt is managing the tradeoff, and itās not hard to understand why. He has a career .630 wOBA on balls hit in the air, miles higher than his .258 wOBA on grounders. If the cost of turning groundballs into line drives and fly balls is marginally worse plate discipline (his K-BB is 2% worse during his recent hot streak as compared to earlier in the year), thatās a swap anyone would make.
Goldschmidt hasnāt fixed everything that ails him. His walk rate during his recent hot streak is worryingly low, and his performance on high-velocity fastballs, which Jay identified as a problem last year, is the lowest of his career. Heāll need to adjust again if pitchers counter his recent aggression on inside pitches. For a batter who was metronomically consistent for so many years, his future is startlingly unsettled.
Ask Goldschmidt himself, and he’s not convinced there was ever a reason for concern. When he spoke to Rian Watt on July 3, near the nadir of his season, he had this to say: “I wouldn’t use the word ‘concern.’ There’s always times that you struggle. This time has been a little longer than other times in my career, but that’s part of the game, and that’s going to happen. I’ve just got to find a way to make it happen out there.” In the month since then, he’s absolutely made it happen, though with a slightly different approach than we’re used to seeing from him. Goldschmidt may not have been concerned, but he seems to be adjusting.
At the moment, the chaos is working for him. Goldschmidt has been one of the best hitters in baseball for the past month, powering a Cardinals offense desperately in need of consistency. Heās shown, with adjustments small and large, why the Cardinals were so eager to acquire him this offseason. After all the early-season struggles and worries about age-related decline, it seems surprising, but this season has gone about how many would have predicted this spring. The Cardinals are locked in a three-way dogfight for the NL Central, and Goldschmidtās hot bat is keeping them in the race. Who knows what will happen tomorrow, but for now, a season of highs, lows, and dramatic adjustments has ended in a predictable place.
Ben is a writer at FanGraphs. He can be found on Bluesky @benclemens.
The opinions about the source of short-term volatility don’t seem to matter for anything. The underlying metrics vary just as wildly as the real outcomes. You can drop all of the batted ball data and get to exactly the same places. There is no inherent truth in the batted ball data – its an overly narrow view of batted balls for my taste. The one thing that is way down are his doubles – that and a depressed babip is what has Goldy having his worst year. The FB explain the babip and the lack of doubles. It is entirely possible that he hasn’t made any changes or fixed anything at any point as things will vary either way. Things never go the way players want them to go – if they did, then everyone would have a career year every year. The whole thing looks like he is hitting more balls in the air to me, which may or may not be intentional and/or sustainable. Many hitters have gone that route as the physical skills begin to decline. When you can’t make as much quality contact there is a point where it makes sense to maximize the limited quality contact you do make. Its not something the best hitters in the game do, but it can keep a player relevant for a while longer. It doesn’t really seem to be working for him, but you don’t really know that it is intentional. For Goldy it could just be what decline looks like, it could be an injury or it could be intentional. I think the FB/GB rates tell you more than the EV derivatives. it seems like the worst analysis is often rooted in EV derivatives – the Taylor Motters and Keon Broxtons of the world come to relevance through that school of thought.
I watch real baseball and I don’t see players adjusting but I do see a lot of inconsistency from game to game and at bat to at bat – just as it has always been. I see a lot of pitchers with poor command of limited repertoires – I don’t think your average arm is exploiting or adapting anything. I have a hard time believing that it is a game of chess as much as it is a game of poker which is much more dependent on luck in the short term, but lesser so over a long run. At the end of the year things seem pretty close to predictable with wild variations in between. I think if you skipped all of the modern predictive models and just used past counting stats with some consideration of lineup changes, park changes, advancing age and longer-term track record you would end up way ahead of what you would get from statcast-driven contemporary analysis. There are no profits to be made from that line of thinking though. Snake oil has always been a seller.
Real baseball? I’m not even sure what that means. What happens on the field is reflected in the numbers — both descriptive and predictive. It basically comes down to what numbers you trust.
Using “past counting stats” (with the carpton of “modern” adjustments you stipulate) is a perfectly logical approach.
Bu to end up “way ahead,” you’d kinda have to sort of … do it.
I encourage you to make the attempt.
I should however add (and back to the OP, jam-packed as it is with all this new-fangled snake-oil) that the most interesting part about all this isn’t necessarily the — descriptive – numbers. It’s Paul G’s approach. This is obviously a guy at or about the top of his game who is entirely at ease with making adjustments on instinct.
Baseball is an entrancing sport. Numbers or not.
lmao sure.
no
imagine pulling the ball in air.
Goldschidt has a 91 wRC+ in August so far, including an 0-for-14 with 1 walk in a very recent 4-game stretch. I think it remains to be seen whether he’s really surging or not.
Yeah, Iām not buying it. His timing continues to be awkward and seems to be both watching really good pitches go by and swinging through center cut fastballs
This is likely to be age
Goldy is a very big man and that is the type of physical profile that tends to lose the quickness required to catch up to heat earlier than the slimmer body. Last year he looked awful for half the season then turned it on. He has rebounded to some extent this year but the Cardinals should be worried as his contract plays out.
Likewise, I don’t see a huge difference between the two highlighted swings. In the first he simply just almost misses the ball, and tops it straight into the ground. In the second he doesn’t.
His weekend against the Pirates pretty much squashes that theory.
Goldy has an OPS vs the Pirates over >1.200. He’s under .750 vs non-Pirates teams. Take the Pirates away, and he’s garden variety average.
Players don’t magically perform better against certain teams for any reason other than randomness. You can’t just throw out his numbers against Pittsburgh.
Is facing a crappy pitching staff 18x a year “randomness”?
Try to step away from the dogma.
Other teams have bad pitchers on their staff. Sometimes you come into Washington and face Scherzer, Corbin, and Strasburg, and sometimes you come into Washington and face Hellickson, Fedde, and Sanchez. These things mostly balance out. Most players beat up on bad pitching and suck against the good ones…that’s why the good pitchers are good. There’s literally no reason to single out Goldschmidt for his success against one team.
You basically just supported my point. The Pirates have no Scherzer, Corbin or Strasburg. Facing them is like facing Hellickson, Fedde and Sanchez every single game. Maybe Keller changes that going forward, but w/o Taillon and with the crap bullpen (minus Vazquez), they’ve been the worst non-Coors Field staff in the NL, and by a wide margin.
I tried to edit the above, but wasn’t in time to do so. What I wanted to add was this:
With all the data available about how the Pirates pitching philosophy lags behind much of MLB and their continued reliance on fastballs, it surprises me to see that folks are so willing to abandon the obvious and defer to simplistic dogmatic rationale on this topic.
Obviously looking at only 2019 vs one team is small sample, but it merits discussion that the only Pirates starting pitcher with any success vs Goldschmidt this season is the junkballing Agrazal. Otherwise, the Pirates throw Goldschmidt 4 seamers at a 47% rate and FBs in general at 65% when MLB throws him 4 seamers on 33% of the time and FB 57%. Could that not be the actual problem?
Generally I’m in full support of analytics explaining a whole lot of what goes on in MLB, but sometimes one needs to step back and really look at a specific tree in the forest.
I’m just not sure “dogmatic” means what you think it means.
The Cardinals also have to face the Cubs’ starters 18 times a year as well. Meanwhile, teams in the NL East get to feast on the Marlins’ pitching 18 times a year, while teams in the NL West get to take 3 trips to Coors Field. Overall, these things mostly even out.
So Goldschmidt was just pressing?
Is this what is going on with Manny Machado?
I think that now the game is cleaner, we are seeing the real aging curve of a baseball players. They decline quickly over the age of 30!
Just to make what I was saying below crystal clear, today there’s an article here on how bad Pirates pitching has been lately.
The Goldschmidt article claims he’s on a resurgence since June 30. Turns out that since June 30, he’s faced that incredibly bad Bucs pitching in 10 games, accounting for 31.1% of all his plate appearances during the sample time cited to suggest his resurgence.
In those 10 games vs Bucs, Goldy had a 1.406 OPS. In the non-Bucs games, he had a .801 OPS, better than the .750 or whatever he started July with, but if he’d only compiled an .801 OPS over the sample time, there wouldn’t have been an article on his resurgence.
Yes, everyone faces bad pitching and good pitching. But c’mon. Between the short sample size used to make a case for Goldy’s resurgence and almost a third of his PAs during that sample being against a woeful Bucs staff, I’m going with “no actual resurgence.”
If he doesn’t keep up this tear (at least until next time he faces Pirates pitching), I believe I’ll be proven correct on this.
What you’re saying is that his “resurgence” is the result of facing a lot of bad pitching, and when he goes back to facing an average amount of bad pitching, it will disappear. This could be tested in a number of ways. One fairly straightforward way: Was he beating up on bad pitchers early in the year too? If not, then he may well have made an adjustment that has allowed him to beat up on them now.