On Matt Harvey and Steroids

Steroids! They're bad, right? They're "performance-enhancing drugs," and the people who use PEDs are cheaters who should be suspended for lengthy periods of time, possibly given lifetime bans, and certainly not voted into the Hall of Fame even if they did flat-out goofy things at the plate like getting on base more than half the time over a four-year span while drawing more intentional walks than strikeouts. (Yeah.)

But... why?

Because, y'know, "performance-enhancing drug" doesn't really narrow it down. Here's a performance-enhancing drug: water. Athletes who consume H2O on a regular basis will perform considerably better than those who don't, on account of those other guys being, y'know, dead. Maybe you say water isn't a drug, but "drug" actually has a remarkably broad definition: it's basically any chemical substance that has a physiological effect on you. Water definitely has one of those! And if that seems a rather trivial example, consider something like the Gatorade found in clubhouses, which is pretty deliberately chemically engineered to enhance performance as much as possible. I'm pretty sure they advertise about that, electrolytes and such. We want our athletes consuming chemical substances that are designed to enhance their performance. They're supposed to want to be good, after all. So what demarcates a "PED" as such?

There is, of course, an answer: a PED of the sort we like to ban is bad for you. There aren't a lot of downsides to drinking water. As far as I know there aren't major downsides to the electrolytes in Gatorade. There are, however, serious downsides to anabolic steroids, and other PEDs. These sorts of drugs essentially give players the opportunity to trade their own health for a performance enhancement. Perhaps in the abstract if each individual player could make that decision in a vacuum we might, in a nice, non-paternalistic way, let them, but of course the world doesn't work like that. There's a ton of pressure on players to be the best that they can, both because they want their teams to succeed blah blah blah and because, y'know, it gets them paid. And so in a world with a whole bunch of drugs offering health/performance trade-offs, we worry, sensibly, that everyone is going to be pressured into making those trade-offs, and we don't like that. It's sort of like how each individual worker has an incentive to offer to work for less than the other guy: it's individually rational but collectively disastrous. So we have minimum wage laws and PED bans designed to prevent these pressures from forcing everyone to give away their own welfare in an effort to out-compete the other guy.

So that's it, right? Competition is all very well and good and everyone should try to be the best, but you're not allowed to trade your own physical well-being for an edge on the field?

Rubbish. Of course you are. Hell, you're supposed to. We valorize people who do this, if they do it the right way.

Why Does No One Suspect Randy Johnson of Steroid Use?

I'm not saying he did steroids. I just think it's weird no one else is. One of the main markers of steroid use, people think, one of the only things that actually shows up in the empirical record is that they allow players to remain effective later into their careers. Traditionally baseball players would start declining around the age of 30 to 33, and then drop off sharply after something like 35. During the Steroid Era, players were staying successful into their forties. So here's a career trajectory comparison between two of the best pitchers of the 90s and 2000s: Randy Johnson and Pedro Martinez. Let's look at each of them through the age of 28 and then after the age of 29:

Young Pedro: 125 W, 56 L, 1576.1 IP, 2.68 ERA, 168 ERA+, 10.4 K/9, 2.5 BB/9, 6.7 H/9, 38.5 WAA, 52.0 WAR

Pedro through his age 28 season was historically dominant. Actually, his age 28 season itself was historically dominant, as his ERA was less than half that of the next lowest qualifier in the AL.

Old Pedro: 94 W, 44 L, 1251.0 IP, 3.23 ERA, 139 ERA+, 9.6 K/9, 2.3 BB/9, 7.5 H/9, 22.9 WAA, 34.0

That's... not much of a decline. He was a bit worse, but still pitched at a Hall of Fame level. What he didn't do, though, was pitch for very long. Well less than half his career was after his age 29 season.

Young Randy: 49 W, 48 L, 818.0 IP, 3.95 ERA, 101 ERA+, 9.0 K/9, 5.7 BB/9, 7.1 H/9, 0.1 WAA, 7.6 WAR

During the same part of his life when Pedro was rocketing to stardom, Randy Johnson was... okay? Average-ish? His strikeout and hit rates showed promise, but he was walking batters like his name was Oliver Perez.

Old Randy: 254 W, 118 L, 3317.1 IP, 3.13 ERA, 146 ERA+, 11.0 K/9, 2.7 BB/9, 7.3 H/9, 68.1 WAA, 96.7 WAR

Randy Johnson had a Hall of Fame career entirely after his age-28 season. And not even a borderline one, a greatest-of-the-great one. It's a good thing, too, 'cause he might as well not have been in the league until age 29 at all. Unsurprisingly, he's second all time in WAR after the age of 29 to Cy Young; Phil Niekro, knuckleballer extraordinaire, is third. He was 880th in WAR up to age 28. Yeah.

Now, there's a perfectly good explanation for why 28-year-old Randy Johnson had never been anything but a mediocre pitcher but 29+-year-old Randy Johnson was one of the most dominant pitchers of all time: he figured out his control. He always had the unhittable, strikeout stuff, but as Sandy Koufax will tell you, that only matters once you figure out how to throw it in the zone. Still, he did the same thing Bonds and Clemens and Palmeiro did. Actually, he did it even moreso: he had the most WAR of any player in baseball over the age of 35 from 1994 on. Yeah, including Bonds. Together they have a big lead over Clemens, and he's a ways in front of #4 on the list, Jamie Moyer. (Who, uh, wasn't doing steroids, I don't think.) So why, of the three people who put up more than 40 WAR after the age of 35 in the steroid era, is Randy Johnson exempt from suspicion? No one elsee seems to be. Well, that's not quite right: various people seem to be, like Frank Thomas, for no articulable reasons. But it cannot be denied that Randy Johnson had exactly the lack of age-related decline thought to be distinctive of steroid use. That's more evidence than we have on Piazza or Bagwell.

If You Want to Subtract the Steroid Guys, You Should Add Someone in Their Place

Well, that was anticlimactic. No one got voted into the Hall of Fame this afternoon. No one. From a ballot with quite a few of the greatest names in baseball history on it. Not Bonds, not Clemens, not Sosa or McGwire or Palmeiro. Not Biggio or Bagwell, or Mike Piazza. Not Jack Morris or Tim Raines or Lee Smith. Not Curt Schilling or Edgar Martinez or Alan Trammell, or Larry Walker or Fred McGriff or Dale Murphy, who was in his last year of elligibility (though he by no means deserved to get in). No one. Since I would have voted for quite a few of those players, obviously I disagree with that outcome, most emphatically when it comes to my favorite player, Mike Piazza, the best-hitting catcher of all time. At least there's always next year.

But I do have one point I want to make, which is that any voter who wants to weed out the steroid users and vote against them on the grounds that they cheated really ought to find replacement people to put in in their stead. Now, some of the voters said that they just plain refuse to vote for anyone who played in the Steroid Era at all, which strikes me as being deliberately unfair to any player who played clean and was therefore the victim of all the cheating. I am not addressing those voters; there would be no point, other to say that they're being pompous, obnoxious, and mean. No, this point concerns only those voters who believe that all and only those players who we have decent reason to believe were themselves cheating should be excluded from the Hall. Those voters should, in my opinion, couple their withholding of votes for the likes of Bonds, Clemens, McGwire, etc. (and hopefully they'll leave Piazza off that list in the future) with the awarding of votes to some guys who would have looked a whole lot better had Bonds, Clemens, McGwire, etc. not been in the league playing the way they did.

Take John Olerud, for example. Over a 17-year career spanning 2234 games, in which he had 9063 plate appearances and 7592 at-bats, Olerud had 2239 hits, of which 13 were triples, 255 were home runs, and 500 were doubles. He scored 1139 runs and drove in 1230 runs. He drew 1275 walks but only struck out 1016 times. All that is good for a .295/.398/.465 batting line, which is good all-around but in particular is an elite on-base percentage, the 31st best in the integration era (1947 onwards). Olerud was an integral part of not one but two championship teams with the Blue Jays and the nearly-championship 1999 Mets. He won a batting title, hitting .363 in 1993, and in a sense deserved to win another, hitting .354 in 1998 and finishing second behind Larry Walker's Coors Field-aided .363. In each of those years he was clearly an MVP-level player, and in quite a few other years he was an All Star-level player. And while he was a first baseman, he was a really really good defensive first baseman, winning three Gold Gloves, probably deserving quite a few more, and (according to Baseball Reference) contributing almost enough high-quality defense to cancel out the penalty for being a first baseman in the first place.

All that adds up, according to Baseball Reference, to 53.7 career wins above replacement and 27.4 wins above average. Those are borderline Hall of Fame numbers, and probably slightly on the wrong side of the border. But keep in mind, WAR is an adjusted-for-league-average stat. And Olerud was playing in a league full of cheaters, like McGwire and Palmeiro and Sosa and Bonds. They make his adjusted numbers, which are already on the borderline for Hall consideration, look worse than they should be. And he was playing in a league full of cheaters like Roger Clemens, who he faced 107 times, more than any other pitcher, and against whom he hit just .205/.335/.373. So, first, Olerud's numbers are worsened because a good chunk of the pitchers he faced were cheating, and second his numbers get worsened again in the adjustment process because he gets compared to other hitters who were cheating. What that means, I think, is that if you want to try to pretend the steroid thing never happened, i.e. to imagine that all the cheating players just hadn't been there, John Olerud has a pretty strong Hall of Fame case.

Now, you don't necessarily have to agree with me about Olerud. But I think the basic principle is sound: if you're the kind of voter who wants to go through the list of great players from the previous two decades who might look like they deserve induction and strike the names of anyone we have sufficient reason to believe used steroids, you ought also to take a second look at players like Olerud, and see whether they deserve a few bonus points for having been really really good while playing clean in a league of cheaters. Merely subtracting is not enough; you should add something back as well.