More On Steroids (Part 1)
PROGRAMMING NOTE: Today was supposed to feature the latest Chapter of The A-Rod Chronicles, on the Biogenesis scandal, the 2013 season, and the arbitration hearing that led to his 162-game suspension… but then, last week, ESPN decided to drop a 10-part investigative report on Biogenesis, timing it specifically so we would have to incorporate the new reporting, but not giving us enough time to do so on schedule. Alas, The A-Rod Chronicles should be back next week. In the meantime, the break allows me to revisit, in more detail, a subject from an episode a couple weeks ago: steroids. (Warning: This is a long one…)
When I was planning Chapter 6 of The A-Rod Chronicles, I went back and forth on how much evidence about the effects of steroids to include. On the one hand, I didn’t want to overload the episode with data; on the other, I know that my central claim — that steroids were not the cause of the home run surge in the 1990s — is a tough one for baseball fans to accept. For two decades now, it has been conventional wisdom that steroids were what caused records to fall and balls to fly, so much so that we refer to the whole era as “The Steroid Era.” So it takes a lot of evidence to overcome that dogma — I know I needed convincing, and I still occasionally find myself forgetting what I know, reflexively saying about a slumping player, “Oh, he must be off his steroid regimen…”
If you listened to the episode, you know I ended up relying a lot on the Juiced Ball theory, because that’s what really convinced me. Because we KNOW that Rawlings, MLB’s exclusive supplier of baseballs, moved its production from Haiti to Costa Rica in the early 1990s, studies have shown that balls after the move were springier than balls before. Crucially, a change to the ball is more consistent with the SUDDENNESS of the jump in home runs, which can be dated very precisely to June of 1993. Steroids couldn’t explain that, unless somehow everyone in the league started taking the same drugs at the same time and feeling their impact on the same day.
But whenever I bring up the Juiced Ball theory, I get the same answer. People are not unconvinced by the Juiced Ball theory, but they are unwilling to give up the Steroid Hypothesis as well, so I always get some version of: “Well, ok, maybe the ball played some role, but I think steroids were ALSO a cause.”
I must admit to finding this kind of infuriating. Because, sure, most things in life have many causes. But there’s just no need to invoke steroids. There’s no missing piece of the mystery. If we were trying to figure out why the light came on, and I said, “It must be the whims of Ch’aska, the Incan god of twilight.” And you said, “No, I just flipped this switch right here”; then I bet you’d be pretty annoyed if I said, “Well, I still think both things played a role.”
Changes to the ball have fully explained changes in the home run rate that were far greater than what occurred in the Steroids Era. In 1988, the home run rate fell by about ~30%, or roughly what it went up by in 1993, and everyone acknowledged that was due to changes in the ball. From 2017 to 2019, there were more home runs hit than during ANY three-year stretch of the Steroid Era, and the league seemed to acknowledge it was due to changes in the ball, even explicitly changing the ball to lower the home run rate in 2021. And it worked! Home runs fell by 23% from their height in 2019 to last year. In other words, changes to the ball can fully explain changes to the home run rate like the one baseball saw in the 1990s; we don’t need to go looking for other explanations.
Once you accept that the Juiced Ball theory doesn’t require any support beams to hold it up, you begin to see the Steroid Hypothesis for what it is: an “answer” in search of a question. Steroids become this mysterious thing, an almost magical explanation that gets trotted out for every strange thing that happens on a baseball field. All the vicissitudes of the sports, the ups and downs, the slumps and hot streaks, the ebbs and flows of a season… all are potentially the result of steroids. One of the many reasons I think the comparison to the US Drug War works is that “steroids” become this catchall excuse, the way being “on drugs” used to be used by some squares to explain all deviant behavior.
To be fair, there ARE lots of confusing things that happen, both in life and in baseball, and it’s nice to have a handy explanation for them. But the Steroid Hypothesis doesn’t really explain most of them, and in fact the various explanations it is used for often contradict one another. Let’s go through a couple of them….
The Mark McGwire Question (How did these guys get so big?)
One of the first things everyone points out about the Steroid Era is how big the players seemed to get. Mark McGwire, the most obvious example, was listed at 6-5, and 245 pounds in 1998. Jose Canseco was 6-4, and 240. Manny Ramirez was only 6 ft, but he was listed as 225 pounds. This was a marked contrast to baseball stars of the past, who rarely weighed much over 200 pounds. Even a power hitter like Hank Aaron was listed at only 180.
Steroids provided an obvious explanation, since putting on muscle is the whole point of using steroids. And for many players, that probably DOES explain how they got big. For example, on Roger Clemens’ 1986 baseball card his weight is listed as 205 pounds; in 1998, he was up to 230. So I’m not trying to deny that PEDs helped players put on muscle.
But this is missing two important bits of context. The first is that steroids are a small part of this story. For most of baseball’s history, adding a lot of muscle was discouraged. The conventional wisdom was that being big would make you more prone to injury, or mess up your swing. Throughout baseball’s history, most of the game’s sluggers were NOT big guys: Of baseball’s top ten all-time home run hitters prior to the Steroid Era, only Babe Ruth was listed at over 200 pounds.
That started to change in the 1980s, as weight training became more common. This trend was not unique to baseball: As weight training got more advanced, and more mainstream, athletes in sports from sprinting to basketball found ways to adapt it to their sport.
Steroids were often part of these weight training regimens, but the trend was much bigger than that. In baseball, that meant an influx of sluggers who, in previous eras, might have been overlooked or told to lose weight. Many of these players entered the league before the league-wide surge in home runs, and most of them were never tied to steroids: guys like Frank Thomas, Andrés Galarraga, Jim Thome, and Mo Vaughn were changing the conventional wisdom of how a slugger should look.
Even someone like Mark McGwire, who admitted doing steroids eventually, was big as soon as he entered the league, before he ever started using. All of this is to say that steroids were as much a consequence of players now wanting to get bigger, as they were a cause of those big players. Players didn’t suddenly get bigger because they started using steroids; they started using steroids because they suddenly wanted to get bigger.
And then the second piece of missing context is that it’s not even clear that the original conventional wisdom was wrong. Are bigger hitters necessarily better? If you look at this year’s home run leaderboard, you see names like Mookie Betts and Ronald Acuña, Jr. and Matt Olson, none of whom are built like bodybuilders.* It certainly doesn’t seem like hulked-out dudes make the best power hitters, on balance. Perhaps there was some truth to the opinions that governed baseball for nearly a century?
*This stuff is hard to be precise about because so many of the reported weights of ballplayers are obviously lies. To make comparisons even harder, the incentives for lying have changed direction: guys used to lie about how slim they were, and now they seem to lie about how big they are. For example, Matt Olson is currently listed at 225 lbs, which… I mean, sure, if that’s what he’s telling people. The most ridiculous example of this is Pete Alonso, who is listed as being five pounds heavier than Mark McGwire was in the 1998 season, despite being two inches shorter. Here’s a side-by-side comparison of the two:
If you believe Alonso is the bigger guy, then I have a cow I want to sell you…
Of course, a power hitter can come in many shapes and sizes, and weight training can obviously be beneficial for some players. But on the aggregate, it’s just not clear that steroids are the best way to build a power hitter, or that simply inflating the muscle mass of every player in the league will lead to more home runs. Which leads us to Question #2…
The Sammy Sosa Question (What caused all these historic seasons?)
Of course, people DID hit more home runs in the Steroid Era. The main reason so many people believe the Steroids Hypothesis is all the historic seasons that happened in that era. Sammy Sosa is perhaps the best illustration of this: He hit over 60 home runs in three different seasons. Outside of the Steroid Era, that has only been done three times in all of baseball history. And yet, in none of those seasons did Sosa lead the league in home runs, because there was always someone else hitting EVEN MORE home runs, either Mark McGwire or Barry Bonds. What else but steroids could explain so many 60-homer seasons over a brief, four-year stretch?
Notice, though, that this is a slightly different question than the one we started off by asking. Initially the Steroid Hypothesis was supposed to explain a general trend (why the league hit more home runs as a whole); now it is trying to explain a handful of outlier seasons (why these specific players reached historic home run totals). In the podcast, I alluded to this “soft Steroid Hypothesis,” which blames the ball and other external factors for the league-wide trend, but attributes the historic seasons to specific players who used drugs.
Of course, normally, we wouldn’t look for two different explanations. It’s logical that these things would move together. If the league is hitting more home runs, then the league-leaders are going to hit more home runs. In other words, the reason for all the 60-home run seasons in this stretch is the same reason there were more 50-home run seasons in this stretch, and more 40-home run seasons in this stretch. When the league hits more home runs, you are going to see higher individual home run totals…*
*Making this a little confusing is that, of course, things don’t ONLY follow the immutable logic of prevailing trends. For example, when the home run rate shot up in the 2016-2021 period (reaching higher levels than it ever did during the Steroid Era), nobody ever reached 60 home runs (although Giancarlo Stanton got up to 59 in 2017). And then last year, when the home run rate actually FELL by 12.5%, Aaron Judge hit 62, more than anyone since 2001. This is why it is not really wise to make inferences about broad trends based on seasons that are, ultimately, kind of flukey outliers. Time and chance happen to us all, after all…
People only feel the need to invoke the Steroid Hypothesis because Sosa, McGwire, and Bonds WERE all linked to PEDs, which supposedly explains why they, and not three other home run hitters in this era, reached these historic totals. But this is a classic case of looking where the light is: The reason we have such conclusive evidence against McGwire and Bonds is that, because they set the records, they received an inordinate amount of scrutiny, and so such definitive evidence against them was eventually uncovered.
And even though we know Bonds and McGwire USED steroids, we have little evidence that those steroids CAUSED them to hit home runs. After all, both were already outliers in terms of hitting home runs before they ever used PEDs. McGwire led the league in home runs and slugging percentage as a rookie; Bonds led once in home runs and thrice in slugging in the early ‘90s, before he started his steroid regimen. Both were among the league’s leaders in home runs for basically every season they were healthy.
They did both see increases in Steroid Era, when they each set the single-season record, but the bump in their personal home run totals was largely commensurate with the league-wide increase in home runs. All this is to say that if steroids did not exist, you would expect the home run record to be held by players like McGwire or Bonds – historic home run hitters playing in an era of high home run totals.
Which brings us back to Sammy Sosa, who is different. Sosa DID seem to reinvent himself during the Steroid Era. Prior to ‘98, he was a pretty good but mostly unexceptional player, never hitting more than 40 home runs. Then, from 1998 to 2002, at the height of the Steroid Era, he turned into a monster, hitting 292 home runs over those five seasons. He hit over 60 three times, and led the league in homers the two other years.
Then, almost as quickly as he’d taken over the league, he disappeared. After the 2004 season, the Cubs shipped him to Baltimore for a utility infielder, and by 2006 Sosa was out of the league (he came back in 2007, but was barely league-average as a hitter). The timing of Sosa’s rise and fall matched up perfectly with the idea that record-setting sluggers were a creation of PEDs.
But it’s worth pointing out that the evidence against Sosa is not nearly as solid as we have against Bonds and McGwire. For McGwire, we have the accusations of a former teammate (Jose Canseco), the fact that he used a steroid-like substance in Andro, and, ultimately, a rather detailed confession he made in 2010. For Bonds, we have the extensive reporting on BALCO, plus his own testimony in a criminal investigation.
For Sosa, we have no teammate testimony or legal investigation or confession. All we have is a report, from 2009, that his name was on a list of players that failed a drug test in the survey testing of 2003. Sosa himself has always denied using PEDs, and I’ve written before about how selectively that ‘03 list is wielded.
But even if we grant that he was taking banned substances in 2003, this does not say much about how his use contributed to his prior surge. By ‘03, Sosa’s numbers were already declining, and unlike with Bonds or McGwire or Roger Clemens, we cannot really be precise about when he started using and when he was clean. This makes it very hard to make comparisons about how the steroids may have affected their production.
In fact, outside of examples like Bonds/Clemens/McGwire, we actually know very little about who was on what and when. We certainly do not know enough to meaningfully compare the home run totals of players who were on steroids to players who were not. For most players, we can’t even compare their Before Steroids years to After Steroids years. It feels intuitive that the years Sosa was on steroids should correspond to his home run peak. But this is only true if we already grant the assumption that steroids help you hit home runs, and it doesn’t actually match the limited evidence we have.
The Brady Anderson Question (How did HE do THAT?)
If we want to know why some players experienced home run surges, it’s worth revisiting the story of Brady Anderson. As a rookie in 1988, Anderson was traded to the Baltimore Orioles, where he spent 7 ½ seasons as little more than a decent fielder who could steal some bases. His best year was 1992, when he made the All-Star Game and hit a career-high 21 home runs, along with 53 stolen bases. He was a nice player, but nothing special.
Then, in 1996, at the age of 32, he hit 50 home runs.
Such an out of nowhere surge came to symbolize another confounding mystery of the Steroid Era: how so many home runs were hit by guys who had no business hitting home runs. Steroids seemed like an appropriately nefarious explanation for these surges. People get very worked up talking about how middle infielders or journeymen outfielders somehow turned themselves into sluggers. The 50 homer club used to be reserved for the best of the best, and now it was being crashed by guys like Greg Vaughn and Luis Gonzalez and, most glaring of all, Brady Anderson.
One small problem with this is that Anderson has never been linked to steroids.* He was not named in the Mitchell Report; he was not connected to BALCO; he played before testing, so he has no failed test to point to. None of Anderson’s teammates, or any reporters who covered him, ever accused him, and he never experienced any suspicious weight gain.
*Neither, for that matter, was Greg Vaughn. Gonzalez’s name was reportedly on a disputed list of players who failed a 2003, but he denies ever using and reacted angrily when a reporter sort of idly accused him in 2006.
The other part of this theory — that Anderson’s ‘96 season was caused by PEDs — that I’ve never understood is: Why did the steroids stop working in 1997??? The whole reason for the fixation on Anderson’s 1996 numbers is that the year was such an outlier. In 1997, he fell back down to 18 home runs, and he never again hit more than 24 in a season. But if steroids caused the uptick, then why would his numbers come crashing down? It’s not likely that he would stop using after seeing so much success in 1996 — baseball wasn’t testing at that point, and 1997 was a contract year for Anderson.
The truth is that random outlier years are just a part of baseball. You don’t even have to look much past Anderson to find another one. His manager in 1996 was Davey Johnson. As a player, Johnson had been a defense-first second baseman who only once in his first eight years hit more than ten home runs in a season, and never more than 18. Then in 1973, he hit 43 home runs, before crashing down to 15 the next season and never hitting more than eight again. Bert Campaneris, a speedster who played for the Athletics in the ‘60s and ‘70s, had never hit more than six home runs in a season before whacking 22 in 1970, and then never hitting more than eight in a year. Even Roger Maris, whose 61 in ‘61 stood as the single-season home run record for 37 years, never hit 40 in any other season.
Were these seasons caused by steroids? Of course not. Every year, there are outlier seasons. The extent to which Anderson’s 1996 season is an outlier is unusual, but every player has a career year. Right now, whether we realize it or not, there are dozens of guys around the league having career years, years they will look back on one day and wonder how they ever played so well and got numbers so high.
And in the Steroid Era, which was, thanks to the ball, an era of increased home runs, a lot of those career years involved a crazy amount of home runs. Did Greg Vaugn’s 50 home runs in 1998 make him an all-time great player? Of course not, but he was a pretty good player having a great year, just like Campaneris and Johnson and Maris.
Look at the last home run surge, from 2016-19, for a comparison. Did you remember Mark Trumbo hit 47 homers in 2016? Or that Khris Davis hit 48 in 2018? Do you even remember Khris Davis at all?* He was never worth even 3 Wins Above Replacement, according to Baseball Reference, but somehow hit more home runs in a season than Hank Aaron ever did.
*Actually, you probably do: Davis was the guy in Oakland who managed to hit exactly .247 four years in a row, which is arguably the craziest stat in the history of baseball.
And yet nobody feels the need to accuse Davis or Trumbo of being on drugs. For every other era, fans recognize that circumstance and luck and randomness can contribute to flukey seasons that we look back and go, “How did THAT happen?” For most of baseball history, we recognize that the answer to this question is a mystery. But from 1993 to 2007, the answer is always: steroids.
To be continued…