Some Evidence of a Greenie Effect (by JonathanF)

So some colleagues were discussing PEDs and the HOF, a topic which really doesn’t interest me much at all. (I say let ‘em in.) The inevitable amphetamine argument was broached, you know, Mantle used greenies, what’s the difference? This got one of my more quantitative colleagues to try and figure out if one could estimate whether or not amphetamines had an effect on batting stats. He formed a hypothesis which, for all its obvious infirmities, at least should point in the right direction.

He opined that (1) the effect of amphetamines should be most pronounced on day games after night games; (2) batting should be more affected than pitching, on the grounds that even Whitey Ford probably drank a little less the day before he was supposed to pitch; and (3) the effect should be bigger in the 1970-74 period than in the 2006-2010 period, since in the latter period there was testing; indeed, in the prior period I don’t even think it was illegal.

He then said: “Hey, JonathanF: you’ve put together a database of every baseball game ever played. Can you try it?” So I did. I really didn’t expect to see anything, so I was a little surprised.

Taking every game from those two periods, I compiled a simple TeamOPS number for every game (H+W)/(AB+W) + (H+D+2*T+3*H)/AB. I then compared the average teamOPS (simple averages here – you don’t do anything fancy when you don’t think you’re going to get anything) separately for day games after night games (which I called greendays) and all other games.

I did this year by year and got the following results. The “difference” column measures the team OPS difference between greendays and non-greendays. The bold results are statistically significant.


About 15 percent of the early games are greendays, and about 22 percent of the later games.

In all but one of the early years, TeamOPS was higher on greendays than non-greendays, and very nearly the same in 1973. Now that doesn’t indicate anything, since it is probably easier to hit in day games, and greendays are all day games while only some of the non-greendays are. But here’s the interesting part: in the later period, greendays were worse in three years out of five, and very, very close in the two other years, much closer than they ever were in the earlier period. Over both periods, there is almost no difference. Standard significance tests confirm that this result is unlikely to be due to chance.

Now that doesn’t mean it’s due to amphetamines, either. Off the top of my head I can think of quite a few possible reasons for this. But when someone makes a hypothesis, and you don’t expect to find anything, and you do find something the first place you look, it bears some thought. Mantle and Ford! Kick ‘em out! (Not really.)

94 thoughts on “Some Evidence of a Greenie Effect (by JonathanF)”

  1. That is a tremendous and elegant study. One “beneficial” effect of the steroid era was to paint in bold relief the notion that off-field habits can have an immense effect on on-field performance, and that this effect is not limited to those at the margins (say, Mantle’s historically heavy drinking, Bonds’ state-of-the-art roiding, or Josh Hamilton’s drug abuse). Matt Kemp’s off year in 2010 was because he was clubbing with Rihanna, and nobody will ever convince me otherwise. I believe the disco Pirates were buffeted by cocaine in the late ’70s, and were brought down by it in the early ’80s.

    You referenced other factors which might contribute to this effect. Given that pitching is a much more labor-intensive activity than hitting, I wonder how well a heavy drinker (of which there were many more in those days, I would think) would pitch in day games, and how this might be shown. My first thought was that you might see a more pronounced inning-by-inning decline in effectiveness during daytime starts. Then again, maybe sweating out the booze over the course of the game would result in the opposite effect. Hmmm….

  2. Ha, ‘Greendays.’ Here in the service industry we call them ‘Clopens,’ and they suck. I’m on the back end of one now. Anyone know a good greenie dealer?

    Really digging this series, by the way.

  3. This is a fantastic post, fellas. I really like the uptick in “original research” contributions from the BravesJo team. Well done, Jonathan. Clearly not definitive, but extremely interesting results.

  4. Hard to believe Earl Weaver was only 82. He looked like an old man 35 years ago.

    Jim Leyland is only 68. Smoking is deadly, even if you’re not dead yet.

  5. It should be pretty easy to check if this effect is due to the relative improvement of offensive players at night over time (whether it be due to better lighting, playing more night games, or whatever). You can figure out the same difference for all day games vs all night games each year, and you know the percentage of day games that are Greenday games each year.

    Also, could there be an over/under counting of certain teams? If the Cubs, for example, didn’t have night games in the 70s, they’d have fewer day games after night games, so the Greenday set from that era has proportionally fewer Cubs games in it, contributing to the bump.

  6. I really want to see it. I’m actually a fan of Bradley Cooper and Jennifer Lawrence is my current obsession.

  7. Why is hypothesis (1) sound? I would imagine that players using amphetamines would use them for any game regardless of first pitch time. It seems to me the day-night confounders would overwhelm any effect due to amphetamines.

  8. Also, in my view, it seems to me the use of amphetamines simply could have allowed a player to play in a day-after-night game, as opposed to play better in that game. Not sure how that effect could be measured, since it wouldn’t necessarily improve statistical performance over the earlier game.

  9. Jeff K, they very well may have used them for every game. I think the premise is they would be most likely to show a positive effect in the games where the players were most tired, which would be the games with the quickest turn around.

  10. @19, I think I see this a bit differently – the hypothesis (1) is the assertion he’s trying to examine data to prove or disprove. This by no means proves anything – it’s just a first red-face test on the idea. Does this data support or refute the assertion is the issue at hand, I think.

  11. Stan Musial is dead. The Man and Ted Williams were probably the two best hitters of my youth.


  12. @manvaldez
    #Braves José Costanza was awarded a Gold Glove in Dominican Winter League. @ajcbraves

    That seals it. You know he’s going to be our opening day LF.

  13. @27 Addition by substraction. We also lost JJ and TH. Pitching staff will be better. Good chance that Uggla and McCann should have better years. Bourne had poor second half. Simmonds for full year. FF and JH should improve. If too many players regress having a year older CJ would make little difference.

  14. @28 No way. Just no way this year’s roster is better than last year’s. We lost our best offensive player and have done jack to replace him.

  15. Yeah, I have to agree with Bethany. We can say that Upton/Walden are capable of Replacing Bourn/Durbin. But there is no chance that Laird/Constanza/Teheran can match Chippèr/Ross/and Hanson.

  16. Full year of Simmons will certainly be an upgrade, but hes got to stay healthy. Either way, this is just still a boarderline playoff team if everything goes right.

  17. Full year of Simmons could be much better than 120 games of aging CJ. How many games did JJ and TH cost last year? If there are injuries we are doomed as all teams would be.

  18. Was getting rid of Hanson that big an improvement? Was he really that bad? I’d rather have him pitch for the Braves in 2013 than Delgado, to be frank.

  19. 112 games for CJ in 2012 at 40. 100 at 41? 2.6 WAR an be replaced. What was total SS WAR last year? COmbined SS & 3B WAR this year will be close.

  20. 2012 SS WAR Hitting 0.0
    Defense 1.8
    AS in 49 games had 0.7 Hitting WAR and 2.4 Defensive WAR. We cannot expect 9 or 10 WAR, but 4 or 5 would be great.

  21. Greenies: Very interesting the statistical differences between the time periods, but we certainly cannot conclusively blame it on Greenies. There is too much conjecture in the use being significantly limited to day games after night games.

  22. I love this quote about Stan Musial:

    Asked once how best to handle Mr. Musial, Brooklyn Dodger pitcher Carl Erskine said, “I just throw him my best stuff, then run over to back up third base.”


  23. I couldn’t get over this quote from Musial:

    “I consciously memorized the speed at which every pitcher in the league threw his fastball, curve, and slider,” he said. “Then, I’d pick up the speed of the ball in the first 30 feet of its flight and knew how it would move once it has crossed the plate.”

    Stan the Man.

  24. @37

    Says it right there in the title – “some evidence”. This is the nature of any study of this type – you try to apply valid logic, and isolate a variable as best you can. Certitude is impossible, and not really the goal.

  25. PaulV and Parish: I don’t think anybody disagrees with you, certainly not me. Definitive? Definitely not. Correlation, not causation? Yep. But studies like this can’t be definitive and can’t prove causation. Really, no study can. Causation is nothing more than correlation plus rhetoric in any study of historic data. If you don’t believe the causation story, nothing about the correlation will do anything to convince you. I agree that the danger really goes the other way — when you believe in the causation story, you’re likely to fall for any ol’ correlation. The trick is to follow what sansho1 says: come at it from multiple angles, a correlation here and a correlation there, pretty soon the sets of plausible causation stories you can tell narrow down to just a few. Does this story narrow the unvierse of possible causation stories very much? Nope? At all? Sure. (Imagine the results had come out strongly the other way, for example.) Does that bother me? Nope.

  26. In a dream last night, this happened:

    D’Backs get: J.J. Hardy, Wilson Betemit
    Orioles get: Chris Johnson, Juan Francisco, Randall Delgado, Luis Avilan
    Braves get: Justin Upton

  27. @23 – Sure the hypothesis is what the study is trying to test (i.e., the dependent variable), but if the study doesn’t control for the independent variable (i.e., amphetamine use) and you can’t verify whether another independent variable that you can control is an unconfounded surrogate for amphetamine use (i.e., day-after-night games), then I think it is a stretch to say that statitically significant increase in TeamOPS is some evidence of an amphetatmine effect.

  28. @44: It’s definitely a stretch. But indirect tests of this sort (through a hypothesized mediating effect) happen all the time in social science. If anyone is thoroughly convinced by this study they have a very low credibility bar. Please write me privately for investment advice.

  29. Sorry to harp on this… (OK, we’re all waiting for the Falcons game so there isn’t much else to do.) But this study has a result: OPS differentials between day games after night games and all other games shrunk between these two periods. The cause of that result is unknown, and will always be unknown. If you think that a change in amphetamine use would have had this effect then your strength of belief in this cause should be at least slightly higher, although not necessarily against any particular alternative. But, Jeff K, there could be confounding. PaulK had one in the very first response to this post, and Joey T suggests a test to try and help separate this potential confounder @14. In the absence of quantitative evidence of the underlying independent variable, amphetamine use, we will never answer this question beyond a shadow of doubt. Even if we had that variable, we might not be able to answer it — depends on the size of the signal relative to the noise and the existence of unmeasured confounders. But this, in a nutshell, is how nonexperimental science works. Some experimental friends of my acquaintance (that is, friends in experimental sciences, not friends I’m doing trials of friendship on) sometimesdoubt that such investigations deserve the label science. Whatever.
    Go Falcons.

  30. My son asked me a question I couldn’t answer off the top of my head:

    What year did MLB officially ban the use of steroids?

    Thanks. (And, yes, I know I could probably Google it but am hoping for some discussion of this topic. Hence my post.)

  31. Francisco, Delgado and Avilan for the other Upton?

    D-Backs would be crazy to say no. Braves would be crazy to offer. He’s not Matt Kemp.

    I know some of you would take that deal in a heartbeat, but we’re starting to do in our minds what Wren did in his that resulted in the contract for Derek Lowe.

  32. A plague upon the House of DeBartolo. (Yeah, I know, but I still hate those guys.)

    Go Falcons! Shut Harbaugh’s mouth – at least for a day.

  33. @48 You mean, trade a bench 3B, a 5th starter and a great left reliever (of which we have 2 others), for an above average LF, for 3 more years? And you don’t do that? Wow.

  34. Just a brief defense of causal inference here.

    While proving causation through an observational experiment (like the one JonathanF did above) is problematic, you can prove causation (at least as far as you can “prove” anything, but we’re quickly devolving into philosophy of science naval-gazing at that point) through carefully designed experiments where factors of interest are systematically varied. This would basically be the equivalent of getting a bunch of baseball players and, over the course of a season or 3, having them use greenies on specific, predetermined days of the schedule. You’d have them use them on some day games, some night games, some days back-to-back, some games against certain teams, some days not against those same teams, etc.

    The results from a study like that would go a long way to showing whether there was any causal link between greenies and performance at baseball.

    So yeah, there are ways to show causal relationships. It’s just really tough to do through observational studies.

  35. Thanks, JonathonF. So there’s really no gray area for the injectors, is there?

    Bethany, you’re correct in your characterization of what those three are right now to the Braves. I just think all three are going to be instrumental to our future ability to stay competitive. And they’re cheap. Remember, we’re broke.

    Let’s say we get the other Upton and we find out why the D-Backs are so ready to divorce. We’re doooooomed.

    Or, you could be right and the Upton brothers lead us to three straight World Series.

  36. Francisco is a low average, low bop, platoon 3B with average defense, Avilan the untouchable is a lefty reliever, Delgado is a good trade piece and could be a middle of the rotation starter. You easily trade those 3 for Upton

  37. @45-46 – Right, and I’ve learned what I know about social science statistical studies from my wife, who is a political science professor at American U. But, it isn’t just whether this study has low predictive value. It seems to me that the data from 2005-2009 tend to disprove the hypothesis that there is a drop in performance in day-after-night games. The analysis of TeamOPS shown above shows no such effect, so I think we need to revisit the hypothesis.

  38. The grey area for injectors is that the Players Union never officially signed off on baseball’s steroid ban — it was never collectively bargained, it was presented as a fait accompli by the owners (who then, obviously, opened their pocketbooks for players who were obviously juicing). So the owners can cry foul, but they didn’t do anything until the mid-2000’s, other than promulgate a toothless policy.

  39. @56

    In defense of the study, I would say that the current climate shows that whatever day-after-night performance effects exist are evenly distributed between pitchers and hitters, and thus no overall effect is shown.

    In the earlier period, a significant rise in day-after-night OPS performance may point (in part) to the particular effect that greenies could have on the reflexive action of hitting, as opposed to the the more deliberate initiating action of pitching.

  40. @61: “Things don’t work at 40 the way they used to.” My advice to Chipper…. hold that thought until you’re 56.

  41. It seems like every reference to a ‘greenie effect’ I’ve heard of before this was in regards to second half slumps. I wonder if the data would show a pre- and post- all-star game difference. Athough, that probably effect hitters and pithers equally, so I doubt the numbers would be that different.

  42. It’s a good game so far. I’m glad SF decided to take some shots down field. It really feels like they can run the ball all day, but they’ve also got down-field throws there, too.

    At the same time, it also seems like Atlanta’s getting all of the “one inch” plays. Ryan’s gotten off great passes immediately before (and at times while) being sacked, and White and Jones have made tremendous plays down field. If they keep executing at that level, it’ll be hard for SF to ever stop them.

  43. To borrow from Kevin Stallings, it was a great professional football game if you didn’t care who won. Going to be tough to get a better one in the Super Bowl.

  44. Same story line…got the lead, committed mistakes, gave up the lead, and failed to comeback. The Falcons have learned a lot from the Braves.

  45. @73

    That’s narrative applied over random variance. If the 49ers flip sequence on one of their first three drives the whole “had a lead, lost it” narrative is shown to be crap.

  46. As I mentioned last week, if you let the bad guys make easy throws you are going to lose. The Falcons had nothing for two quarters, and were kinda lucky on the funble/missed FG not to have yakked it up earlier.

  47. The fumbled snap was probably the biggest play of the game. Silly mistake and they were in the red zone. They at least get 3 there and that would have let them kick the FG in the 4th quarter for the lead. This one is going to bug me for a few days.

  48. Yup, as much as anything, the Matty Butterfingers play was deadly.

    Woulda been nice to get a defensive holding call on that 4th & 4, but you never wanna be in a position to need that flag.

  49. Anyone blaming Matt Ryan for that loss is just completely irrational. He was the only reason the Falcons were in the game. He was awesome.

  50. The irony was that the Falcons finally played pretty well overall – better than last week, easily – but the luck (Roddy White slips, Ryan fumbles, Matthews’s hand grazes the QB’s helmet, etc.), at the same time, finally went against them. I mean, they averaged more per play than the opponent, for a change; they actually outplayed the 49ers in most meaningful statistical categories. And yet they lost. Such is the cruelty of a one-and-done playoff system.

  51. A Falcons fan was stabbed outside the stadium after the game. No idea how Ray Lewis managed to get from Atlanta to Foxboro in time for kickoff.

  52. Ryan had an all-time first half & 2 second-half turnovers that were totally on him.

    The Falcons did not score in the 2nd half. Everyone, especially the defense, can shoulder the “blame,” but that dropped snap was huge. Even 3 points there & it’s a different game.

  53. The INT was clearly on White, not Ryan. Clearly.

    Anyway, my response was to 77, which looks like blame and is coming from someone with a history of blaming Ryan. And the rest of Atlanta’s great athletes.

  54. Yep, the D looked outclassed most of the game. Way too many gaps in coverage and little pressure on the passer. If they hadn’t gotten that lucky TO at the goal line it wouldn’t even have been close.

    That said, Matty Ice turning into Keystone Ice, again, lost that game.

  55. Every football team suffers injuries and I know it is meaningless to play the what if game regarding them, but I’m going to do it anyway. If the Falcons had Brent Grimes this year they would have won the Superbowl.

  56. From MLB Rumors. Not sure how true it is but I find it incredibly funny…

    “Chipper Jones was honored at the New York BBWAA dinner last night and confessed to the audience he was starting to get the itch to go back to Spring Training, Rubin reports. “I was on the Braves’ web site, just kind of messing around on my computer,” said Jones. “I was thinking to myself, ‘You know what? I think I’m going to go down and get me a workout in, see how everything feels.’ I was down there for about five minutes and I figured I’d go to Hawaii instead.”

Leave a Reply

Your email address will not be published. Required fields are marked *