Athletics Nation: An SB Nation Community

Navigation: Jump to content areas:


Sports blogs for fans, by fans.
New Blog: Sounder At Heart for Seattle Sounders Fans!

Moneyball and PEDs, a Successful Combination

Well first I'd like to thank Manny for putting steroids and performance enhancing drugs back in the news today.  I had originally planned to do more analysis on the franchise's ability to develop young players, but I was really too lazy to try and find the right mix of players, collect the data, and then get it in the right format.  Plus, I don't have access to forecasting metrics like PECOTA, and didn't feel like spending hours on end doing all this work only to be refuted for the same reasons as in the first post. 

Star-divide

 

Like everyone around here, I read Moneyball back in 2003, and was thoroughly impressed by the results Beane and DePo had in regards to competing against competition with payrolls doubling Oakland's.  The only reason I mention the book is because it seems to have given Beane a free pass in regards to fans objectively evaluating trade results, draft picks, and team performance.  As a disclaimer I don't remember a lot of specific details in the book anymore, but I do know it touched on much more than the aforementioned stuff.  Anyhow, as time has passed Oakland's success has started to significantly dwindle.  The decreasing performance doesn't just apply to the A's in general, but it has also followed the stars of those early decade A's teams to the point where any valid contributors from Oakland and the early 2000's have become essentially unknowns.  Aside from losing a few stars to free agency there really wasn't much I could come up with that caused the downturn in performance within the organization.

One thing my friends always like to give me hard a time about is teasing me that Oakland was the organization that introduced steroids into baseball.  From Canseco and McGwire, to Giambi and Tejada, the biggest stars in Oakland always seem to somehow be linked to PED use.  It also doesn't help that BA in BALCO refers to Bay Area.  Citing mlb.com's Drug Policy Timeline I saw that the MLB began random drug testing in 2001, and then in 2003 every player on the 40 man roster was required to be tested at least once.  Coincidentally enough, the A's offensive struggles began right around that time.  Therefore, I decided to plot the A's overall team performance in runs scored, home runs, and slugging vs. that of the rest of the league to see if there were any noticeable results or trends since the start of drug testing.  Originally I performed the tests against all of the MLB, but I realized it would be much more fair to do it only against the AL so the data wouldn't be skewed by the poor hitting NL teams.  The results were pretty astonishing.

Runs_mediumHrs_mediumSlg_medium

As you can see before the steroid testing, through 2000, the A's were outperforming everyone by a pretty significant margin in every statistical category shown.  You can see the steady increase within the league and in Oakland up until 2000, and then the sudden drop off after the start of testing in 2001.  From there, the league begins a slight downward trend that could be expected from such a change.  The A's on the other hand drop absolutely off the map.  The results speak for themselves, but I wanted to be sure that the mean values before and after PED testing were significantly different.  Since the runs scored data was normally distributed I was able to perform a Two Sample T-Test at an alpha level of 95% on the runs scored data to verify that the hypothesis of runs scored from 1995-2001, and 2002-2008 were statistically equal.  That provided these results:

Two-sample T for 02-08 vs 95-01

       N   Mean  StDev  Mean

02-08  7  755.9   52.0    20

95-01  7  840.4   77.3    29

95% CI for difference:  (-161.2810, -7.8618)

T-Test of difference = 0 (vs not =): T-Value = -2.40  P-Value = 0.033  DF = 12

Both use Pooled StDev = 65.8664

Because zero is not within the range of the "95% CI for difference" and the P-Value is less than .05, statistically speaking we can reject the hypothesis that the two data sets are equal.  This means that from the data there is in fact a difference between the runs scored prior to testing and after. 

So what do we take from this?  Well getting back to Moneyball I'm not so sure that Billy Beane was ever the expert we made him out to be.  I think we unfairly built him up to be an infallible genius, remember "In Billy We Trust?"  I'm of the opinion that Beane's success in finding undervalued players, and having them instantly turning into All-Stars upon landing in Oakland was because these guys were on PEDs.  Now that steroid testing is in full force, successful teams have become more reliant on efficiently developing their young talent.  PED testing has also weeded out the younger players that are instantly able to come into the league and mash 20 home runs, and then five years later be completely out of baseball.  I'm staring at you Ben Grieve, and Terrance Long.  I won't even touch Matt Stair's and John Jaha's 35 home runs years.  Heck, even Eric Chavez was healthy enough to have back to back 30HR years and even made the cover a video game at one point in his career!

It should also be noted, of players on those teams, Adam Piatt, Jeremy Giambi, Jason Giambi, Miguel Tejada, and Randy Velarde have all been mentioned in the Mitchell Report as being known PED users.  I won't do it, but you could even go as far as making the case that Beane intentionally sought out known steroid users with the idea they were more likely to have huge years. David Justice, Mike Piazza, and Jack Cust all come to mind and fit the bill.  Alternatively, you could also make the argument the A's reduced offensive performance is based on Beane seeking to eliminate any potential PED users within the organization leaving the franchise with the complete lack of power we see today. 

Don't get me wrong with this, I'm not trying to question Beane's ethics, but I do think it's time to seriously question his aura of infallibility.  The A's old station-to-station approach of taking a walk and just waiting for your big power hitters to knock you in is obviously a thing of the past.  As I alluded to in the player development post I don't know for sure if the organization is still preaching that philosophy, but if they are, I feel like the A's are now the organization that is lagging the trend.  I do realize the Big Three is no longer around, but I'm not so sure even they could win with the offense of recent years.  That goes without saying Hudson is out after TJ surgery, Zito has been throwing BP for two years up until the past week, and Mulder... well I'm not even sure where he is these days.

I hope this post opens up a few eyes.  I've got a feeling questioning Beane will bring me some trouble, but as long as I don't have to hear any correlation doesn't equal causation nonsense in regards to steroids not affecting power numbers I'll be alright.

 

0 recs  |  Comment 50 comments

Story-email Email Printer Print

Comments

Display:

I won't make you hear any "correlation doesn't equal causation" nonsense

I’ll just say that you seem to have proven, rather conclusively, that Jason Giambi and Eric Chavez in their prime, along with Miguel Tejada, hit, slug, and score runs better than Eric Chavez playing through back and shoulder misery, Daric Barton, Jack Hannahan, and Bobby Crosby. Who knew?

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 7, 2009 6:38 PM PDT reply actions   0 recs

Yeah, it definitely seems like this could be chocked down to...

Earlier this decade, the A’s had good hitters. Now they have bad hitters.

KMoAsFan, you say you’re not trying to question Billy’s ethics right after you say you think he sought PED-users because he knew they would have huge years. Uhh, what? Do you think if Billy was seeking out an advantage by finding PED-users, that other GMs weren’t doing the same? As a GM, you have to take whatever advantage you can find—and this is not to say that Beane did this; it’s a completely unsubstantiated accustion.

I do like, graphs, though, and I think it’d be cool if you did something about A’s hitters between years with batting coach A and years with batting coach B, or same with pitchers or fielders.

by NateHST on May 7, 2009 6:57 PM PDT up reply actions   0 recs

With the A's, that would have to be

batting coaches A-H.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 7, 2009 7:09 PM PDT up reply actions   0 recs

I would then say.

Who’s fault is it that for the last 5 years there haven’t been any good hitters on the team?

I also didn’t say he sought PED users, I specifically said “I won’t” say he did that, but you “could” “go as far as” saying he did. Nothing in that says I think he sought PED-users. You mentioned that if Beane had in fact done that, how would I know other GM’s wouldn’t be doing the same. Thats exactly why I didn’t criticize or claim Beane did that. With as much money that is at stake, no GM would’ve sat back and tried to avoid PED users, because they obviously brought wins, records, attention, and crowds.

What about Barry?
"Barry who?" Forst said, and I felt like I was in the middle of a knock-knock joke.

by KMoAsFan on May 7, 2009 7:09 PM PDT up reply actions   0 recs

So you're not saying he did,

but you’re saying he could have. OK. You’re saying maybe he didn’t seek out PED users but maybe he didn’t actively seek to avoid them. OK.

So at best you’re saying that your whole “possible theory” is baloney, and at worst you’re saying maybe the A’s were guilty of a “don’t ask don’t tell” ignorance that ruled the game at large. OK.

Your point, then, is that the A’s didn’t uniquely go out of their way to find out who was using PEDs and who wasn’t and then to avoid those who were, when the other 29 teams weren’t doing that and weren’t required to? OK.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 7, 2009 7:15 PM PDT up reply actions   0 recs

I don't understand what's so difficult about this.

Are we not allowed to say anything negative about Beane for fear he won’t come back to AN for interviews?

I’m saying the A’s offensive struggles are due to Beane’s ineptitude. His success in the early 90’s was either luck that can be attributed to players playing above their expected level due to PED use, or due to Beane being the more shrewd GM at the time, and being one of the first to seek out players that were on steroids. Because I can’t prove the second assertion I’ll stick to him getting lucky with the slew of guys that were on the team. You know, the guys that were on steroids and won MVP awards.

The team has been terrible offensively since 2003. Why is that? Could be poor coaching, could be poor talent evaluation, could be stupid trades. All falling back on the GM. This isn’t complicated. Since none of us have the kind of resources to find that type of inside information on the franchise I had to look elsewhere. The start of steroid testing provides a very acceptable drop in performance in regards to the entire league. The drop it causes the A’s is ridiculous.

What about Barry?
"Barry who?" Forst said, and I felt like I was in the middle of a knock-knock joke.

by KMoAsFan on May 7, 2009 7:28 PM PDT up reply actions   0 recs

Ah, the old "We're not allowed to say anything negative about Beane" try.

Have you been reading the site the past two weeks? You can say all the negative things you want about Beane; it’s just that if you make arguments that are easily refuted and then have no answer for the refutations, you might get called on it.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 7, 2009 7:36 PM PDT up reply actions   0 recs

The A's offensive struggles are caused by

the hitters’ ineptitude. And in this, you’re leaving no room for Beane being a good GM, saying either A) He got lucky because everyone he chose was doing them or B) He sought them out because they were doing them. What about C) He found good hitters that took the easy advantage when they saw one.

And you’re walking a damn fine line with saying you think he did this and not accusing him of doing it.

Also, do you think the A’s offensive decline could have been caused by injuries and forcing the A’s to get replacement level players for them because they don’t have the Yankees’ wallet?

by NateHST on May 7, 2009 7:39 PM PDT up reply actions   0 recs

Yeah. I blame Beane plenty for the A's offensive ineptitudes,

but not because I think his way of getting good talent cheap before was to scan the “overperforming due to PEDs” market or anything related to PEDs.

Beane’s failures, IMO, have revolved around banking far too heavily on Chavez in the face of all evidence, taking far too long to recognize that Crosby’s shortcomings had little to do with health, and focusing so much on pitching (which is generally a great idea) that the team and system both became bereft of hitting talent, actual or potential.

He also, even recently, got Cust for pennies, acquired Cunningham, Carter, and Cardenas, drafted Doolittle and Suzuki. So it’s not even all bad.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 7, 2009 7:46 PM PDT up reply actions   0 recs

I agree for the most part, but I have a couple little quips with Cros/Chavy

With Crosby, the potential was there. After his rookie year, there were injuries that appeared to hinder his performance up until last year. When it became clear that maybe Cros really did suck, there was really nothing Billy could do. Nobody even wanted him off waivers. Why he’s still on the team, I will assess in a moment.

With Chavvy, Beane was unfortunate enough to give him an albatross of a contract instead of retaining Miguel Tejada. He was injured and expensive.

Unlike other teams with, you know, money, the A’s can’t just cut their losses and buy another good player. Not replacing Crosby, I understand why this happened. Not replacing Chavvy, that’s another issue, (Warning: Now entering Cliche City) but through the fog I can still see the rock and I can see the hard place and I can see Beane stuck in between.

by NateHST on May 7, 2009 8:19 PM PDT up reply actions   0 recs

I guess I feel it was pretty clear Crosby wasn't that good

as long as 3 years ago.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 7, 2009 8:25 PM PDT up reply actions   0 recs

Beane didn't become GM of the A's until 1998

so I’m not sure what you mean by “his success in the early 1990’s”. He did work as Assistant GM from 1994-98 under Sandy Alderson, but Alderson was very much the man in charge/

Beane has pointed out in the past that he was not responsible for drafting some of the A’s biggest stars of the late 1990’s and early 2000’s, such as Giambi, Chavez, Tejada, and Hudson.

I don’t doubt that Beane did not seek to avoid suspected PEDs users, but then which GMs did?

One could make an argument that the criteria he developed as GM for selecting players in the draft, via trade, or free agency have not resulted in success as far as getting the best offensive players. (I’m not arguing for or against that, but it would be an interesting subject for research.)

by OaklandSi on May 7, 2009 8:07 PM PDT up reply actions   0 recs

My bad.

I didn’t mean to say 90’s, I meant early 00’s.

Ok Nico, for the third time, it doesn’t have anything to do with him scanning the landscape for players that could have potentially been on PEDs. It has to do with guys he signs blowing up and having career years between 1999 and 2001. Coincidentally enough (or not), those were the years everyone in the league was blowing up and having career years. It probably wasn’t even his fault he got so much attention from Lewis’s book, but Beane ended up being the one who came off as a genius over the whole thing. That so called genius then created an infallibility around here which irritates the crap out of me.

It was less than a year ago we were all sitting around here talking about how great the team would be in a couple years being anchored by Carlos Gonzalez, Travis Buck, and Daric Barton. Heck, I bought into the moves myself and was totally pumped, after all, Billy was in charge, he would never steer the team in the wrong direction. It was unfathomable to think CarGon would’ve been dealt at this time last year. We were all itching so bad to see him in the green in gold. People used to talk as if success was guaranteed in the coming years, but why? Then Harden and Gaudin get traded for Gallagher, Patterson, Donaldson, and Murton. No one really questions the move of giving up on the season. Everyone just says it will make the team better for the future. Never mind the performance and location of any of those players right now.

I hate looking back at trades and saying would’ve could’ve should’ve, but the only thing I will say is that he hasn’t really been any more successful than any other GM at coming out on top lately. Especially not enough to get the publicity, confidence, and ownership stake in the team that he has. Lets not talk about the complete depletion of the farm system that required all the trades to be made in the first place. How you can let that happen is beyond me. It’s understandable that the A’s can’t easily fill spots on the roster after guys move on from free agency, get hurt, retire etc, due to their lower budget. The thing is, Beane got the reputation for being able to weather those occurrences by finding players no one else was picking up on. He has failed to do that in a long time. Well actually, just since the league adopted a more formal PED policy.

I would be arguing against me too if I was somehow claiming Beane intentionally sought out PED users, but thats not what I’m doing. I just wanted to point out the startling effect that steroid testing had on the A’s during his tenure, and how much worse an affect it had on the A’s than the rest of the AL. The other main thing is that I think he gets way too much of a free pass, and I had my own opinion it was because he got lucky during the steroid era. I’m fine with people saying the players aren’t as good now, but not so much the injuries. Every franchise in every sport deals with that, not just the A’s. Chavez and Crosby being injured does not result in 100 fewer home runs being hit in 2008 than in 2000. When you say the players aren’t as good you have to look at who put them on the team or gave them their contract extensions in the first place. That responsibility falls on the GM. The team is in it’s third straight year of offensive ineptitude and has been GETTING WORSE during the REBUILDING years when they should be getting better. The A’s are walking a fine line between becoming the Pirates, Royals, and Rays of the last decade, and the sad thing is, no one seems to look at it that way. The scary part is that everyone is so quick to point out the weak AL West, and all the flaws that apparently befall the Angels, Mariners, and Rangers, but it’s really our own favorite team we should be more concerned about.

What about Barry?
"Barry who?" Forst said, and I felt like I was in the middle of a knock-knock joke.

by KMoAsFan on May 7, 2009 11:57 PM PDT up reply actions   0 recs

i see your point

and yeah, i agree timing was a bitch in terms of our players, their heydays and drug testing. we had talented players who were enjoying their best years plus a lil extra right before baseball as a sport cracked down on PED accross the board. that doesnt mean that Beane’s team building durring those years wasn’t genius. it was. nor does it mean that those players weren’t great players without PED. they were. but let’s call a spaid a spaid. we were an offense heavy team because some of our sluggers were juicing. maybe giambi and miggy would have had MVP seasons without PED… we’ll never know. all teams saw a drop in production once testign was implemented, the A’s were no different. Beane did his job, assembling talented teams. in my mind: free agency + drug testing+ general competitive ebb and flow = the last 5 seasons.

i was wrong to do that stuff

by jaylikewise on May 8, 2009 8:34 AM PDT up reply actions   0 recs

This is Beane's weakness

Building an offense of ineptitude is right IMO.

And Nico is right, he banked too much on Chavez. You would think he would have drafted several players for this position over the last three or four years.

by Trainman on May 8, 2009 3:24 PM PDT up reply actions   0 recs

This post

"Until Holiday comes around, I will refer to him as Coors Light." Ak_A

by doctorK on May 7, 2009 7:44 PM PDT reply actions   0 recs

I think this is MUCH too harsh

The post makes a point we all think has some merit. Even though KMOAS is missing a few explanatory variables (like talent), I don’t have too much trouble believing the A’s would be more impacted by testing than other teams.

Good post…

by ohmangoAs on May 8, 2009 12:32 AM PDT up reply actions   0 recs

oh, and really,

no post deserves a one word dismissal. Especially not one with genuine effort by a regular AN member.

by ohmangoAs on May 8, 2009 12:34 AM PDT up reply actions   0 recs

I quite agree

There is definitely some merit. Leaving aside the utter comtempt I feel every time some uses the word ‘fail’, there’s something to be said.

I mean, can anyone really say that Jack Hannahan + PEDs = Eric Chavez?

This is not to say that Eric was doping (actually, I don’t think he was, for what it’s worth), but rather, we have no way of knowing if the current A’s players lack talent or chemical help.

It’s certainly something worth discussing, rather than pissing all over

It's never too soon to jump to conclusions

by alea iacta est on May 8, 2009 1:41 AM PDT up reply actions   0 recs

Mea culpa

Obviously, I was overreacting. In my defense, I had just been reading these other posts:
“I think Ziggy got the Swine Flu (even though I have no evidence)”
“Troy Neel, Father of the year” – since been deleted because the post was simply pasted text from an AP story.

I still disagree with this gist of this post. I seriously doubt that Beane was actively seeking out known ’roiders as being undervalued players. More likely he turned a blind eye to the PED issue just like the rest of baseball, and instead concentrated on getting players who, for whatever reason, were not considered worthy by the rest of baseball, like Hatteberg, Bradford, etc.

"Until Holiday comes around, I will refer to him as Coors Light." Ak_A

by doctorK on May 8, 2009 8:35 AM PDT up reply actions   0 recs

It's a wonder so many people have been leaving the site lately.

This wasn’t some stupid unsubstantiated post. It takes time and effort to collect all this data, compile it, figure out the right statistical tests to use, and then in fact interpret whatever the heck the tests show. If I didn’t feel there was merit to it I wouldn’t waste hours out of my life making a fanpost about it. Heck, I really have no motivation to ever do it again. There is a clear correlation between steroid testing and the A’s offensive performance. Much more so than the rest of the league. It’s practically irrefutable, but yet somehow the majority of the people on here keep ignoring the statistical proof and are taking it as an attack on Beane’s morals.

I would really like people to observe the difference in performance because of steroids and not. There is much more of a dip than one or two players on the roster could affect. The reduction in performance reflects a team wide drop in power, not just the loss of Giambi or Tejada, or an injured Eric Chavez. I could plot batting average and OBP to see how the non power based numbers show up. I’m completely guessing here, but it wouldn’t surprise me one bit if there wasn’t near the drop, or maybe an improvement up until 2008. What would that say? Would those charts even matter to you guys?

What about Barry?
"Barry who?" Forst said, and I felt like I was in the middle of a knock-knock joke.

by KMoAsFan on May 8, 2009 9:30 AM PDT up reply actions   0 recs

Hmm...

Except:

a. On AN, whether it’s right or wrong to do so, folks have always been quick to be dismissive of analysis they found to be flawed in premise or data analysis. You say “don’t give me that correlation is not causation nonsense,” almost as if you know it applies.

AN’s traffic is stronger than ever. People have always left and people have always come.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 8, 2009 10:32 AM PDT up reply actions   0 recs

what the fuck?

honestly, I’ll take my first CGV here

I absolutely, completely, cannot believe you just said that, Nico. Correlation vs. causation is a reasonable argument. Disagreeing about it is even MORE reasonable. If he thinks his results are strong enough to indicate causation, that’s a reasonable stance.

For you to fucking say that an AN member is expendable because you think they fucked up in their ORIGINAL research is unfuckingbelievable. I’m really pissed.

AN’s traffic is stronger than ever. People have always left and people have always come.

I’ll memorialize this as the worst thing I’ve seen I’ve seen on AN in a while. I doubt anyone affiliated with SBN, least of all Blez, thinks this way, or would say it.

Users who have over 800 comments and 5 posts in 1 year and have (or, it appears, sadly, had) the motivation to do real research are a fucking far cry from expendable. That is the group of users who are disappearing faster than they are appearing. Believe me when I say that having users who have the time and want to try to do intelligent analysis is the ONLY thing that makes AN different from other sites, and, frankly that uniqueness is dying ever without our telling people they can take a hike.

by ohmangoAs on May 8, 2009 10:51 AM PDT up reply actions   2 recs

Nico would have to speak for himself, but I think when he wrote
AN’s traffic is stronger than ever. People have always left and people have always come.

it was in direct response to this comment above

It’s a wonder so many people have been leaving the site lately.

I am pretty sure he was not making a commentary on whether or not the OP was expendable, but was just replying to that particular comment.

by AsFanInLA on May 8, 2009 11:46 AM PDT up reply actions   0 recs

well

I have no ax to grind…if that was all it was, Nico can just say so (or not, if it was obvious, and the consensus is that I’m crazy).

In context, Nico argues the point, then says people leaving is irrelevant/not an issue. I found that really callous in the context of the fact that KMOAS had just said:
 It’s a wonder so many people have been leaving the site lately.

This wasn’t some stupid unsubstantiated post. It takes time and effort to collect all this data, compile it, figure out the right statistical tests to use, and then in fact interpret whatever the heck the tests show. If I didn’t feel there was merit to it I wouldn’t waste hours out of my life making a fanpost about it. Heck, I really have no motivation to ever do it again.

emphasis mine. The point is, Nico’s “oh well, people come and go” is not paying attention to what KMOAS said. Frankly, this post really was of value. If we had 10 a month of these, AN would be more awesome (if for no other reason that people can learn from the subsequent debate if not the post itself).

KMOAS said he was discouraged by the response, and not planning to do it again. Nico said “oh well, no worries, people come and go”.

at best, negligent. Still, it’s no biggie- Users will continue to be dismissive, as Nico noted above. Hopefully people (and really, as we all know Nico is far from the only one) will try to curtail it.

by ohmangoAs on May 8, 2009 12:37 PM PDT up reply actions   0 recs

FYI

I view this as more damaging that your typical “if you think Adrian Beltre is a bad player, I despise you”…
That’s just people on the internet being superficially insulting, as is typical.

What riled me was the frustration clearly in KMOAS comment, and then the response that just piled on (unintentionally, I’m sure)

by ohmangoAs on May 8, 2009 12:40 PM PDT up reply actions   0 recs

LOL- yeah I liked your private email better. :-)

For clarification, I don’t mean for a minute that the author of this fanpost, or anyone else, is expendable. I just meant to say that when someone refutes the post it’s not because “no one’s allowed to criticize Beane” and that when someone says “that’s why everyone’s leaving” I want to correct them.

But as I said privately, better for me to let others make those points than for me to make it myself. Mea culpa for that.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 8, 2009 1:55 PM PDT up reply actions   0 recs

emailed in response

doing the responsible thing I should have done before :)

by ohmangoAs on May 8, 2009 2:18 PM PDT up reply actions   0 recs

Oh, and I just noted a formatting weirdness

I didn’t write the last sentence to appear in bold. It should read with the heading “b.” and then it should not be in bold. If the boldness helped cause confusion over what I was trying to say, then I believe Apple and its evil HTML gremlins should get a “strike”.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 8, 2009 4:30 PM PDT up reply actions   0 recs

there is a clear correlation between PED abuse

and offensive performance throughout all of baseball — and actually, through many sports.

As far as the A’s, Moneyball, and any unique effect from the “end” of the steroids era? One of the basic tenets of statistics is that “correlation does not imply causation.” As much work as you evidently put into your post (and that is appreciated), much more work would have to be done — and probably for many other teams for comparison — before you could make a truly substantiated arugment.

The book Moneyball, by the way, was a business book about looking for competitive advantages through exploiting undervalued assets. The A’s under Beane proved to be a terrific case study in part because of Beanes’ use of statistical analysis to find players who had good numbers in undervalued assets such as OBP. Those players would come cheaper and therefore aid Beane in building a competitive team and system that could compete against the big budget clubs. This is no longer such an advantage as the use of statistical analysis and attributes such as OBP have become commonplace throughout baseball.

In an environment where players are cleaning up from the PEDs (or perhaps moving on to still-undetectable PEDs), and where other organizations with lower payrolls (the A’s are certainly not the only ones!), perhaps the next issue to research might be the scouting staff at all levels.

by OaklandSi on May 8, 2009 10:42 AM PDT up reply actions   0 recs

oops, left out part of a sentence

“…and where other organizations with lower payrolls are having success in drafting, developing, trading for, and using good offensive players…”

by OaklandSi on May 8, 2009 10:44 AM PDT up reply actions   0 recs

First of all

I was impressed by the work you did- It’s a lot harder to compile original data and do your own statistical analysis than it is to grab a boutique statistic from fangraphs and interpret the results. You designed a study- it’d be really cool if AN users regularly defined their own study and executed it- that’s real statistics.

The bummer, and what I think most people are reacting to, is that I think you have a lurking variable. Two major things changed in 2002 (actually many more): 1. Drug testing, and 2. The A’s personnel.

They both changed radically. It’s actually very difficult to identify which one caused the change. You feel strongly that the personnel changes weren’t THAT important, but baseball only has 9 hitters. Replacing good Chavez with bad Chavez, Miggy with Crosby, and Giambi with our other 1B over the last few years represents HUGE downgrades on 1/3 of the team. Anyways…without another study (which actually would be interesting) we can agree to let it go.

Finally, ignoring the issue of the lurking variable, your results are not as unambiguous as you suggest.

95% CI for difference: (-161.2810, -7.8618)

That’s a huge interval, and the upper end is damn close to a league average decrease (unless I’m misinterpreting your results). A small difference (say -40) could easily be explained by the personnel shift.

Still, and I’m dead serious, I loved the work. It’s fun seeing original work, and just going through it and thinking about it, even if I think you overlooked or ignored something. Thanks.

by ohmangoAs on May 8, 2009 10:43 AM PDT up reply actions   0 recs

Ohmango, I really appreciate the comments.

Nico, I’m starting to get the feeling you have an agenda to try and discredit anyone you disagree with. Heck, it might be fun to try and search through the AN archives and look at the people you’ve singled out, and then track their posting trends afterwards…

I specifically point out correlation vs. causation because it’s the trendy thing to say on AN to throw out stuff you don’t agree with it. People use that comment, but really have no clue what they are talking about. I used this same example before, but when summer comes around you’d probably notice an increase in ice cream sales, and also an increase in shark attacks. The increase in ice cream sales does not have any correlation with the increased number of shark attacks. On the other hand, the increased number of beach goers just might. Thats why there are so many types of run, time series, and regression plots. MINITAB is a super awesome statistical software package that provides dozens of tests designed to help quality control people determine the root cause of inefficient aspects of various processes. When you are looking at a string of observations and you see a run of defects or points above or below the mean, or any kind of trending that is outside the spec limits the first thing you do is try and investigate the cause for this. If you were manufacturing a gear and every day there was a run of rejected out of spec parts starting around 3 in the afternoon you’d then go and investigate whats going on, on the manufacturing floor at that time. Maybe it corresponds to a new shift of employees coming on. Maybe it corresponds to the drill bit on the end mill becoming worn out through the day.

It’s not rocket science to apply the same principles commonly used in process improvement techniques like Six Sigma to a major league baseball franchise. Thats all I’ve done here. Ohmango, the hypothesis test isn’t as accurate as I’d like it to be, because the sample size is ridiculously small. Because I noticed the data for runs tended to be normally distributed I felt it might be fun just to throw that in there, and since I’ve always had to work out those type tests by hand, I wanted to try out the software. It didn’t do much, but confirm the means were different, but it is good to know that. There are plenty of other tests to do to show that the change is more significant before and after the testing, like a regression fit, or plotting the area under the points, but I didn’t feel like anyone wanted to see more time series plots. I’m glad someone at least appreciated the work that went into it, and at least understood the results for what they are.

What about Barry?
"Barry who?" Forst said, and I felt like I was in the middle of a knock-knock joke.

by KMoAsFan on May 8, 2009 11:44 AM PDT up reply actions   0 recs

i liked it too

and feel you are well within your rights to hypothesize, especially when backed by data. people can agree or disagree. debate is clearly one of the more prevalant aspects of AN, and your post contributes to that. also appreciate the effort involved. thanks.

i was wrong to do that stuff

by jaylikewise on May 8, 2009 12:14 PM PDT up reply actions   0 recs

I'm not sure about this
Nico, I’m starting to get the feeling you have an agenda to try and discredit anyone you disagree with. Heck, it might be fun to try and search through the AN archives and look at the people you’ve singled out, and then track their posting trends afterwards…

It seemed off-hand, not really targeted. And to be fair, there’s plenty of reason for debate on your post.

by ohmangoAs on May 8, 2009 1:55 PM PDT up reply actions   0 recs

I can refute the first sentence rather easily,

by admitting that to a fault I can rarely remember who wrote what. In fact, if two user names are similar I can rarely remember who’s who from one comment to the next. It was a long time before I could tell Oaktown Power and Oaktown Tribesman apart, other than to remember that I knew one of them kind of annoyed me (turned out it was OT, not OP). And when I saw that I had a private email from ohmangoas, I was like, “Hmm…I wonder if we get along online.”

Honestly, until I’ve met someone in person I have the darndest time remembering who wrote what or which similarly-named user made the really nice comment to me and which one made the really snide comment to me.

All of which is to say, KMoAsFan, next time you post or comment I probably won’t remember that you even wrote this fanpost. Not that this is good or bad, just that’s it’s likely.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 8, 2009 3:21 PM PDT up reply actions   0 recs

Speaking of concession sales....

Can you figure out why the Coliseum still has concession people walking the aisles, selling Ice Tea when people are huddled in their blankets on a cold September night?

I would think they have enough statistical data that shows they’d be better off paying that guy to carry hot chocolate instead.

Thanks,
LS

by LoneStranger on May 11, 2009 8:58 AM PDT up reply actions   0 recs

Small Snapple Size?

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 11, 2009 6:16 PM PDT up reply actions   0 recs

Interesting post

I think that the general point that Moneyball and the Steroid Era are compatible is a good one. However, I have much less faith than you in the effectiveness of the drug tests. Given this, I’d say that the A’s offensive decline in recent years has more to do having less talent than juiced players being forced out of the game.

by natethesnyde on May 7, 2009 7:57 PM PDT reply actions   0 recs

Before reading your post ...
find the right mix of players

The right mix of players is all of them. When you’re talking about only a half dozen or so years, you’re dealing with a small sample size already — there is no right mix, you have to look at all of them, otherwise you will almost inevitably just succumb to confirmation bias.

That said, I know it’s a lot of work. Believe me, better than just about anyone, I know. But in statistical research, either you take the time to do it right or don’t bother … which is why I haven’t done any statistical research in a while …

"It's for your own good. Big strong Devo knows whats best for Poppy" -- Mossback

by devo on May 8, 2009 4:10 PM PDT reply actions   1 recs

WE WANT OUR STATURDAYS!!!!!

I thought screaming in a petulant manner might be useful.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 8, 2009 4:18 PM PDT up reply actions   1 recs

I want to write'em ...

but they take a long time …

"It's for your own good. Big strong Devo knows whats best for Poppy" -- Mossback

by devo on May 8, 2009 4:33 PM PDT up reply actions   0 recs

I know - just want you to know that they are much appreciated

and missed.

I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal

by Nico on May 8, 2009 4:41 PM PDT up reply actions   0 recs

Ok ... now I've read your post ...

In 2001 the A’s had Jason Giambi … in 2002 they had Scott Hatteberg …

That’s a dropoff of 74 BRAA (batting runs above average), 23 HRs and .227 Slg …

That alone accounts for ALL of the difference in runs, ALL of the difference in Slg and most of the difference in HRs …

From that point on, the offense more or less hovered around average, going up and down as the league did until 2008 … I don’t think there is any reason to think the downturn that year and anything to do with steroids.

There is (presumably) a clear correlation between PEDs and offense (though I say presumably because we don’t have any real data on PED use) — the league wide numbers support this assertion. That said, I think it’s pretty clear that the difference between 1999-2001and more recent offenses, relative to the league, isn’t steroid use, it’s talent level (and health) level. Now a study linking steroid use to health might prove interesting and, combined with anecdotal evidence and recent trends, might support your hypothesis … but this alone just reminds us of overall trends across MLB and how good of a player Giambi was in 2001 …

"It's for your own good. Big strong Devo knows whats best for Poppy" -- Mossback

by devo on May 8, 2009 4:33 PM PDT reply actions   0 recs

Devo.

Usually you get on my nerves with all the crazy stats, but this is more the type response I was hoping to this post would generate.

That said, I’m not completely sure if your references of Hatte and Giambi are really applicable. On one hand, the A’s hit more HRs in 2002 than they did in 2001. On the other they also scored 84 less runs. I have no idea what the heck to take from that. It would seem to me though, thats more than just losing Giambi would cause. Secondly, the big drop off for the league and the A’s was between 2000-2001, the year they actually started testing. The huge spike over the league average before the drugs were illegal should throw up a red flag. The fact the team immediately drops down to right around the league average could be very telling in regards to the leagues effectiveness in getting players to tone down the usage of PEDs. The recent drop off I would contribute more to Beane’s poor talent evaluation and roster moves. I know thats unsubstantiated, but I find it hard to believe the team is this bad just because the A’s were going to rely on Chavez to completely carry the offense for the second half of the decade.

What about Barry?
"Barry who?" Forst said, and I felt like I was in the middle of a knock-knock joke.

by KMoAsFan on May 8, 2009 5:42 PM PDT up reply actions   0 recs

Giambi really was THAT good in 2001 ...

I actually misread the HRs … but, given that it’s a much smaller sample size stat than Rs and Slg, we should expect it to be more volatile from one year to the next — a well hit ball at a slightly lower angle goes from being a HR to a double, loses 100% of its HR value but only 50% of its Slg value and probably about 40% of its R value …

Consider that the guys we pretty much know were roiding on the A’s from that era — Giambi and Tejada — had their best seasons after testing was instituted (Giambi in 2001, Tejada had has first big season in 2002 and was consistently pretty great through 2006).

The huge spike from 99-01 shouldn’t throw up a red flag at all. Steroids didn’t just then become available — there was no change in the PED environment. The team got better because its young stars (Giambi, Chavez and Tejada) got better.

I promise you — you may not like me referencing crazy stats — but the difference in runs from 2001 to 2002 is pretty much exactly equal to the difference in contributions from Giambi to Hatteberg. The other major change was also a significant decline — from Grieve to Damon — but that was offset by Tejada having his first big year.

Compare the common players from 01 to 02
Hernandez fell off, but he bounced back in a big way in 03 … that’s just growing pains from a young catcher, no reason to think a change in PED usage was involved …
Tejada improved significantly, winning an MVP in his walk year …
Chavez stayed the same …
Long fell off significantly on his way to becoming the worst every day player in the Majors …
Dye stayed the same …

The other four lineup spots were taken by new players (Hatteberg, Ellis, Durham and Justice) …

As you mention, perhaps Loopy Long did go straight, stop using and start sucking … other than that, none of the carry overs offer any evidence to what you are suggesting. And, for what it’s worth, Hatty, Durham and Justice all had better seasons than they did in 2001 with other teams.

The league has definitely caught up with BB quite a bit. In fact, an article by a couple of economic professors a few years ago showed that by the time Moneyball came out, OBP wasn’t actually undervalued anymore — the league had already caught up with him on that.

"It's for your own good. Big strong Devo knows whats best for Poppy" -- Mossback

by devo on May 8, 2009 8:36 PM PDT up reply actions   0 recs

Go KMO Go!

I thought this was a great, interesting provocative post, made better and sharper by the criticism. keep em coming!
I don’t think Beane sought out steroid users. I think he sought out high OBP guys with power, recognizing that at the time speed was overvalued.
during that time I think Beane was right, speed was overvalued until around 2003. Where Beane has failed was in being slow to recognize that the reason his models were successful was significantly due to the fact that his system worked when the players were juiced.
Now that they aren’t juiced. Beane needs to adjust.
I thought the firing of Macha for Geren, and especially the cavalier way he treated Macha, was indicative of Beane being set in his ways intellectually to such an extent that I felt worried for the future of the A’s.
A bit of a rambling post, but there you go.

They're called RUNS for a reason.

by connie mack on May 9, 2009 3:11 PM PDT reply actions   0 recs

Comments For This Post Are Closed


User Tools

Welcome to the SB Nation blog about Oakland Athletics.

Community Guidelines ANcillary Terms
Start posting about the Athletics »

Join SB Nation and dive into communities focused on all your favorite teams.

FanPosts

Community blog posts and discussion.

Recommended FanPosts

Ph3spec_small
Having fun with Pythagoras, or, who got lucky during 2009
Dsc00764_small
DLD - 11/07/2009 - How to Keep Yourself Occupied in the Off-Season
527918550406_0_bg_small
2010 Off-Season Blueprint
Depaulbluedemons_small
Community Prospect List #15
Super_grover_small
2010 Offseason Primer

Recent FanPosts

Depaulbluedemons_small
Zonis's Off Season Charter
Bill_king_small
AFL Rising Stars Showcase Tonight on MLBN @ 5PM
Me_at_att_park_small
Greener Grass, Episode 4: The Biggest Off Season (Potential) Decision Doesn't Involve Free Agents
Oaklandathletics_small
Hardy to Minnesota. What's with the early offseason trades?
Tyler_at_maya_school_small
Happy 6th Birthday, AN!
Oaklandathletics_small
facebook Members: Keep the A's in OAKLAND
Bill_king_small
Organizational Statistics: Winter Leagues

+ New FanPost All FanPosts >

SPONSORS


Managers

Tyler_at_maya_school_small Tyler Bleszinski

08-_the_author_small 67MARQUEZ

Baseball_small baseballgirl

Poochini-butt_in_box_2_small Nico

As_kings_cal_small louismg

Editors

Countdown_small Taj Adib

Ziegler160px_small Flashfire

527918550406_0_bg_small notsellingjeans