« Game of Shadows | Main | The Duke Case »



> It certainly isn't fair that people who are sensible about risks to their health are put at a competitive disadvantage.

Isn't the definition of "sensible" subjective? Don't some sports present a risk to the health of the participant? Should we suggest that competitors who over-exercise be limited in the number of hours they may dedicate to their sport?

Sean Roche

You can't have a reasoned conversation about steroid use by Bonds (or others) without acknowledging that the use of steroids was not outside the rules of baseball (and easily could have been) and that steroids themselves don't make a baseball player better, it allows him to work even harder.

To me, there's something at least a little bit mitigating about the fact that Bonds got bigger because he worked out more than he would otherwise have been able.

Justin Poulin

No record is sacred once millions upon millions of dollars enter the picture. To go so far as to create 'a record-review board' would only enter the confounding variable of politics into the equation. I'm not sure which would provide a more invalid result, politics or drugs. Sure, there is a tone of sarcasm in my point, but both politics and drugs exist in their own state of altered reality. Neither of which relates very well to people on the outside.


Some friends and I are working diligently to convince the MLB to suck it up and be a leader in the testing and hopefully eradication of performance-enhancing drugs from major league sports, be it through forensic economics or straight testing. It's cheating, pure and simple.



If I apply for a job cleaning the crappers at the local stadium, the employers will certainly look at my past performance in the crapper cleaning arena (statistics). When they are done, they will certainly require a pre-employment drug test.
If I've managed to land a job in major league crapper cleaning, and establish a work history (stats), the employers will invariably check my performance against my past and present stats. If my job performance tends to vary widely from an already established record, they will certainly ask me to be drug tested again.

Why should our major league idols be held to any less of a standard than the guy who cleans the crappers after the game?


Examining records for their improbability is a pretty circular way to find cheaters. If records were probable, they wouldn't be records. In the ninety years of modern baseball and amongest the thousands of players who suited up, only a tiny number of players reached 500 homers and (aside from Bonds) only Aaron, Ruth, and Mays all managed to hit more than 600 homers without steriods. What this tells us is this (i) hitting more than 600 homers over a career is the apex of baseball ability, (ii) but it can be done without cheating, and (iii) it is likely to be done again.

Maybe Bonds would not have done it without cheating, but someone would have and it would have been perfectly normal. Rather that cast suspicion on a player merely for their unique statistical place in history, its better to cast aspirsions only where there is actual evidence of cheating.

Bonds will be disgraced historically because there is substantial evidence that he knowingly took illegal steriods. If all we had was his home run record, he'd be a hero like Aaron.


Everyone, the steroids-baseball argument should begin and end with this paper by Arthur DeVany.



Just curious, what makes you so sure Bob Beamon would pass this test?


Never mind, I see you addressed that in your previous post.


Here's my idea for proving these guys use steroids:
make them use steroids!

This is purely theoretical, of course, but think about it. If you tell athletes like Barry Bonds to use steroids on an 'experimental basis' for this season, then record his stats, you could observe the variance from expected value.

Athletes who did not previously use steroids would (I hypothesize) see a statistically significant increase in performance.

Good ol' Barry would probably play at the same level which he currently plays... thus indicating he's been using them anyway!

anyways I know this would never fly because the athletes would object and whatnot, but think about it.


There are a number of sub-arguements here that are quite interesting (mostly people's position on a quadrant of acceptable/not acceptable, cheating/not cheating), but I'll try and address what I see as the main arguement.

Is using statistics a good way to regulate sports?

I would argue that leaving things as they are wouldn't necessarily mean a loss of meaning to records. As already mentioned, records are only really relevant with an understanding of that era as it is.

In fact, I would argue the opposite, that making records adhere to a certain level of plausiblity removes the meaning of those records. Sports are exciting, in part, due to their unpredictability. The fact that the unexpected could happen, is central to my interest in it.

Argueably this is a matter of thresholds, but I think on a commercial level sports whose outcome is more predictable will also be less marketable.


"[V]irtually no one bought my idea for loosing the forensic economists on sports records."

This is true. But the two stated points (failure of the system, and success of steroids) made in the follow-up post ("Forensic Analysis Redux" or FAR) don't address the majority of the comments from the last post. Of the commenters on the last post, only around five addressed the efficacy of steroids, and more than fifteen questioned the advisability of applying a statistical model to real life.

Lester Spence articulated the theme best, saying, "[y]ou cannot use statistical inferences to say anything about the behavior of a given individual in a specific circumstance."

There are good arguments for the use of statistical models, despite their inability to completely describe most systems. Unfortunately, nobody here has explicitly made those arguments.

FAR contains the seeds of those arguments with the allusion to jury trials, but I disagree with the assertion that we (here, I'll generalize only to Americans, because I don't have enough other experience) are willing to accept the fact that juries convict innocent people--it does happen, but we seem to fight against this reality when forced to confront it: I thought that was part of the point of appellate courts and the Death Row reviews in Ohio and Virginia.

I expect that there are a host of underlying reasons behind people's apparent rejection of applying models to important situations (like honors in professional sports). I think a dissection of the reasons behind the general unease with applying statistical models would address many more of the questions people have raised (and convert many more people) than another in-depth look at steroids, which is already contained in "Blink". Besides directly addressing the problems people have, a direct look at acceptance or rejection of statistical models would support the arguments from the last two posts, which deploring current drug-testing problems does not do.


Forensic economists can only raise suspicion and doubt, albeit well-founded.

However applying this to future records would enivitably require some revisionist history on past records. The possible endorsement money available to a "home run king" for instance would certainly bring on a fierce and very winnable legal challenge.

So if it goes to past records, Maris' 61 would almost certainly be thrown out. His other high was 39 and he only reached 30 one other time (33). Another reader pointed this out.

But if what we're talking about is suspicion, then there's even an argument about Ruth's records: why should bizarre statistical anomalies only count towards an individual's past history? Ruth hit so many more home runs than his contemporaries that statistically speaking, something must've been amiss. Simply look at him (as we do with Bonds) and he's clearly no Jim Thorpe. Now we obviously don't know what "is up" in Ruth's case, but given the statistical doubt we can throw his records away too.

I refuse to consider the possibility that McGwire or Sosa were "clean".

By my calculations, Foxx and Greenberg would hold the record as they hit their 58 home runs after the explosion of 1930 (against all-white pitchers, but that's for another day...).

Mubic P

The reason your idea sucks is that it eliminates the possibility of the exceptional. People want to think that the unbelievable could happen, and your system tells them that the unbelievable is not allowed.

Forget the fact that your system would be wrong every now and then. Lets say that someone devised an algorithm, and that everyone knew with a Godlike assurance that the algorithm was infallible. That would be even worse than a flawed system, because it would put a limit on what we believe to be possible. That is the point where the utopia becomes the dystopia.

I'm a grown man and I still like to think that someday someones going to do a 360 from the freethrow line. In a game. And it will become their trademark move over the course of their indian summer season.


Because we watch sports for brilliance -- for the anomalies themselves. Marginal greatness is for weekend softball, not for the stage on which we place our heroes.

Sending innocent athletes to the purgatory of allegation and accusation is much worse than sending an innocent man to jail -- it destroys our sense of idealism and hero-worship, which is (after all) why we pay $75 for semi-decent tickets to a mid-season ballgame.


Mr. Joe Average can no longer afford to attend the games anyway, so I expect the anger, apathy, and outing of users will only increase.
I know, let's talk about something of substance, like CEO's pay going up another 20%, while Joe's wage has increased 0%.

Jim Caserta

I think the issue is more do you want to reduce overall steroid use, or do you just want "clean" records? One of the big problems with MLB players juicing is that it trickles down to the college and high-school level, where you have 2 or 3 orders of magnitude more athletes.

I agree with one of Malcolm's two points. First the disagree: It's not hard to catch juicers, it's just hard to get sufficient evidence to prove it. But I agree the drugs do help, at least short-term. As for it helping your work-ethic, it makes workout recovery faster. Take your first weightlifting session after taking a month off - you're wicked sore the next 3+ days. On steroids, you'd be ready to work out the same muscles the next day.

As for McGwire, if you read Canseco's book, big Mac was 'clean' in 1987 when he hit 49 HR to lead the league. Look at pictures from his rookie season and college - he was lean and massive back then.

Another drug to bring up is ephedra (nearly the antihistamines mentioned above). It does help, but is not an anabolic agent, and it has been linked to professional athlete deaths.



I read Ball Four when I was younger and remember Jim Bouton talking about "greenies" that many players took before games and found out later when I was older that they were basically amphetamines. So I guess enhancement has always been around for those that want an edge.

I guess I don't want to know what Bonds did. I want to live in the blissful era of my youth watching the Bash Brothers swat homeruns. It is like wanting to find out if your ex-girlfriend was cheating on you. Better to let your thoughts of a bissful childhood stay intact than to tear down your idols. On the other hand, what's a better way to placate the feeling of anonymity than a celebrity exposed.


So, there is this guy, he used to drive race cars back in the day. He'd make the motors in his cars bigger, faster, and more powerful than any of his competitors in an effort to blow them away. It worked. The only downside was, the faster he went, the more dangerous it became. He continued to win races at a record clip, twice as much as the next best. Pretty soon, the secret got out and everyone started building bigger, faster, and more dangerous motors. Kids could watch this on TV too. What was this guy's name? It was Richard Petty.


I think this is a pretty drastic step when we are just talking about sports records. I am a devout baseball fan and a very statistically inclined one at that. I don't need a panel of experts to tell me which records are "real" and which ones are not.

I'm comfortable with the notion that while Ruth was great his numbers might be inflated because he didn't play against any black players and that while Koufax was great he was somewhat aided by pitching off a higher mound. Just as I'm comfotable knowing that while Barry Bonds was great he was aided by chemistry.

Truth be told, no record is pure. Athelte's do not perform in static, controlled environments. While some may convincingly argue that steroids is different than a pitchers mound that is raised a couple inches this is just a sport. This era should be treated no differently than the others I mentioned. It will be remembered as the "steroid era" and people can see the numbers however they want.

Arnie McKinnis

I'm going to make a very controversial statement - why do we care if PRO athletes take performance enhancing drugs. They get paid to entertain us, the know the better they entertain us, the more money they make - it's really a very simple equation. I know the argument about how "amateur" athletes will use them to get a better chance to make it into the professional ranks - OK - that may be a happen in some cases (but it will not happen in all cases).

I just don't get the fervor about cheating - they are doing what they need to do to continue getting paid for what they do, as long as can do it. Each of us wants to continue in our choosen profession - to maximize our earnings potential - and provide for our families. If I was a pro baseball player, staring at retirement from my chosen profession at age 30, I would do everything in my power (Everything) to get a few more years out of it. Because after that, I'm looking at going to Autograph conventions and participating in fantasy camps. Not much of a choice.

Tanner Boyle

Speaking of statiscal analysis, here is Eric Walker's (featured in The Nu,bers Game and Moneyball), his analysis of the offensive explosion that is frequently dubbed the "Steroid Era".



A batter has less than a quarter second to decide whether to pursue a pitch.

I wonder how steroids aids in that sort of split second decision making?

The answer I suspect is: It

But don't tell the batter that.
He believes he is demonically empowered.

And that belief makes all the difference.

By the way... did you see the work showing coffee improves athletic performance? I suggest we test for that too.


Good point Arnie!
But how far do we run with this philosphy?
Do we meet the devil at the crossroads to make a deal?


One thing we want to know in sports is who won that day. Drug testing has already diminished that (the winner of last years Tour of Spain bike race has not officially been awarded yet), and doing an analysis would make it worse. It would bring a judging aspect to events that don't currently have them. We would have "the Russian statistician gives...". The steps need to assure clean sports are not worth the cost. Walking away from the stadium saying “what a great game, I can’t wait to find out who won after the judges are done, and maybe a lawsuit or two”

Regarding people that think do should be allowed let me use this analogy. The guy in the cube across you works 90 hours a week and uses cocaine to do it. Your boss gives him a big raise and in your review tells you that you need to work more like him. By allowing drugs in sports the governing bodies would be encouraging their employees to commit a felony and risk their health. Seems like a pretty easy lawsuit to me.

The comments to this entry are closed.

My Photo


  • I'm a writer for the New Yorker magazine, and the author of four books, "The Tipping Point: How Little Things Make a Big Difference", "Blink: The Power of Thinking Without Thinking" and "Outliers: The Story of Success." My latest book, "What the Dog Saw" is a compilation of stories published in The New Yorker. I was born in England, and raised in southwestern Ontario in Canada. Now I live in New York City.

    My great claim to fame is that I'm from the town where they invented the BlackBerry. My family also believes (with some justification) that we are distantly related to Colin Powell. I invite you to look closely at the photograph above and draw your own conclusions.

My Website


  • What the Dog Saw

    buy from amazon


    buy from amazon

    buy from amazon UK


    buy from amazon

    buy from amazon UK

    Tipping Point

    buy from amazon

Recent Articles

Blog powered by Typepad