Mets Geek » Blog Archive » Interview: Tom Tango

April 21, 2008

Interview: Tom Tango

Scribbled by: John Peterson @ 12:28 am | Filed under: Articles, Interviews

Tom Tango, known on the Internet as Tangotiger, is one of most well-respected researchers in baseball today. His web site, Tango on Baseball, is home to sabermetric projects such as The Fan Scouting Report and research and resources on a variety of topics. He co-wrote The Book: Playing the Percentages in Baseball, which I called “the best book on baseball statistics since Baseball Between the Numbers,” with Michel G. Lichtman and Andrew E. Dolphin, and posts frequently on the The Book’s blog.

Thanks for taking the time to answer some questions.

Glad to.

How much time do you spend every day or every week watching, researching, or writing about baseball?

Altogether? Essentially, a regular work week.

Are there any areas of analysis you wish you had included in The Book?

At the time we did it, we covered everything we wanted to. Now, with PITCHf/x coming to the forefront, there is a book you can write … for each player. Everything that is baseball is centered around the batter/pitcher matchup, and the pitch-by-pitch data will give you the convergence of performance analysis and scouting, which is the pinnacle of sabermetrics. It’s the one common language on which we can converse.

Is it disappointing that the findings in The Book have not been adapted in baseball at large?

Whether one team or all thirty teams see the light is irrelevant to me. I’ll be delighted to not see a below average hitter in the two hole, but I won’t be disappointed if it doesn’t happen. I did get excited to see Ned Yost with The Book, and talking about putting the pitcher in the eight hole.

Ten years ago fielding was thought to be overrated by statheads.

It was? You’ll need to provide evidence for that statement.

Until The Book, the sac bunt was thought to be overrated.

That’s true. The traditional thinking is much closer to the truth than the new-age thinking.

What’s the next traditional strategy that will be vindicated by proper sabermetric analysis?

If we can get the run environment down by half a run, it’ll be the stolen base. And in fact, that environment currently exists in the playoffs, and so, stolen bases should be more prevalent there.

Why do you think statistical analysis is not more widely embraced?

I don’t know that it isn’t.

Sure, front offices across the game are wising up. But you watch the games. What is one thing that you kind of hope you could see change about say, how the broadcasters present the game? No more small sample sizes? Clogging the base paths? Something must irk you.

I really, really try not to listen to them. Clogging the bases is a good one, and that is such a myth. I disproved it on my blog last year. Otherwise, when my kid is around and he’s screaming, it’s the one time I don’t tell him to not be so loud.

Why do you think the battle between believers and non-believers is so intense?

Again, I don’t know that it is. Some people are unaware of the issues, and once they see the evidence, they get excited, and start to believe. Most people have not even seen the evidence. So, that’s a huge part of the fans who are simply not in any battle. There is a small faction that is aware of the issues and have seen the evidence, and still reject the new ideas. For people who wish to remain ignorant, the best thing to do is to grant them their wish. Finally, there is an even tinier faction that simply argues for the sake of arguing. And you can’t win that kind of argument.

What should each side understand about the other?

By sides, you seem to imply that each has some truth that the other is intentionally ignoring. On that basis, most people are not on any sides. Those that are on sides, you give them the evidence, and then get out of their way.

How can stat geeks communicate analytical ideas to non-geeks in a non-geeky way?

I don’t know.

Given a team that’s in the bottom third of the league in on-field strategic decision making, by how many more wins do you think that team could improve via better strategies (platooning, lineup construction, bullpen usage, etc.)?

A few wins. Three? Five? Everything is a half-win here, a half-win there. Add them up, and you’ve got a few wins. Of course, teams pay $4-to-5 million for each of those wins on the open market. But, most teams in MLB are like most companies in corporate America: they treat education, research and development as an expense, not as an investment.

You mention the $4 million figure. This is, as I understand it, an approximation of the average price paid per “Win Above Replacement,” which is relatively easy to calculate with both hitting and fielding metrics. What are the reasons why a team ends up paying significantly more or less for a player than their WAR would suggest?

I find that teams are paying free agents extremely well. There’s really not that many free agent signings that you would say were terrible. Most signings are very consistent, and just a handful are bad. Most of the good signings by teams are given to non-Gutsy White Guys (Milton Bradley, Mike Cameron). To the extent that there are bad signings, there are a few reasons: (1) they don’t properly value his fielding, (2) they don’t properly value the positional scarcity, (3) they don’t age him properly. This is called the Carlos Lee Syndrome. Or, they are middle relievers.

You have written about the value of having cheap, young players. Such players are valuable not only because they often peak before free agency but because they cost significantly less than market price for six years. How would one calculate the average free agent value (and “free” production on top of what the team pays) for a class of young player, say, a “B” prospect in AA?

A guy’s first six years will cost you about 33.3 cents on the free agent dollar, on average. So, you can have three young players who are average for six years, and it will cost you the same as an average player who hit free agency. That’s a fantastic ROI. Now, you just have to go find them.

Marcel the Monkey Projection System, which you developed to be a simple projection system based on just weighted three year history, aging curves and regression to the mean, does not do projections for minor leaguers. Do you watch minor league games, grade prospects, or work on projection models for minor leaguers?

No, no, and no.

How do you think Marcel compares to PECOTA and other projection systems? How do you feel about black-box metrics like PECOTA?

I don’t have to think about how it compares. I know, since I have an exhaustive study on it on my blog. The fact of the matter is that PECOTA is no better than any other forecasting system out there, including Marcel. Any forecasting system that dares proclaim its superiority to another forecasting system is simply not being fair. To the extent that you need to proclaim one a winner, that would be Chone. But, that’s like saying a team that wins 84 games is better than one that wins 83. And that’s what all these forecasting systems are … they are all like teams that win 83 games. You might get one that wins 84. A few of the bad ones will win a bit less than 81. But, it’s really hard for any forecasting system to be good or bad.

As for the black box of it all, this is like a magician. If he reveals his trick, you won’t watch any more. Marcel, incredible as it sounds, is taken less seriously precisely because I have given away its secrets. This is what we are faced with, with the Fantasy Baseball market, a market that will believe whatever it wants to believe.

Do you think it’s important that baseball researchers share findings and discuss process? Or can interesting and worthwhile things happen when someone like Bill James does his own thing and isn’t really aware of what others are doing?

Sure, interesting things can get done, but also a whole lot of redundant and terrible things too. Bill James has a few pieces that are simply things that I have already done. Why he would waste his time doing that, I don’t know. But that’s the way he is, by his own admission. Sometimes, you get a pearl out of it. So be it. Researchers at The Hardball Times is the model, as far as the web is concerned. I get immense satisfaction of reading sabermetric works. You’d think all researchers would think like that.

You always seem to be working on the next great statistical project. Tell us what you’re working on right now.

I’ve got a list I posted on my site somewhere called Tango On Demand. Next up for me is creating a Fielding Database. Normally, I would be tacking the PITCHf/x data, but (a) there’s a lot of bright people already there, so I’m happy to be an interested observer at the moment, and (b) the data is still in flux, and I’d rather work on stable datasets.

What do you think about the moves the Mets made this offseason? From Lo Duca to Torrealba to Estrada to Schneider, the Milledge trade, the Castillo signing, the Santana trade, or anything else. What do you think about the general direction of the team?

Basically as good and as bad as any other team. They overpaid for Santana from an open market standpoint, but if you want to get a Lamborghini, you’ll pay whatever the manufacturer says to pay, especially if they tell you they’re only building a small number. In order to get Santana off the free agent market, they overpay for him to guarantee that the Redsox and Yankees don’t get him. I seem to remember thinking they gave Castillo at least one year too many. And I find it hard to believe that not one of the other 28 teams would have offered the Mets more for Milledge.

Do you root for any team?

Yes, the Expos. I’m checking the standings for April 11, 1994, and they’re at the bottom of the division. I’m hoping they’ll make a good run for it.

Who are your favorite players?

Tim Raines. Otherwise, I’m too old to have a favorite player.

What should we all watch for this baseball season?

Baseball. What more can someone want?

John Peterson hates old players on principle. You can read his stylized ravings regularly at Blastings! Thrilledge.

22 Responses to “Interview: Tom Tango”

Comment posted by Tim in LA on April 21, 2008 at 7:01 am (#664129)
Maybe it’s late… Is it just me, or is he being kind of a dick?
Comment posted by tangotiger on April 21, 2008 at 8:25 am (#664130)
If you are not sure, I’d appreciate the benefit of the doubt.
Comment posted by gromit on April 21, 2008 at 9:44 am (#664176)
Honestly, you do come off a bit surly which can occur in written media
Comment posted by mr.bmc on April 21, 2008 at 10:39 am (#664244)
He’s not a dick. He just doesn’t have any funny.

Tom. Be more funny.
Comment posted by jpwf13 on April 21, 2008 at 11:36 am (#664323)
Marcel, incredible as it sounds, is taken less seriously precisely because I have given away its secrets.

Or because you named it after Marcel the Monkey.

Maybe it’s late… Is it just me, or is he being kind of a dick?

If yo think he’s bad you shouls see how his pal MGL comes across.

How can stat geeks communicate analytical ideas to non-geeks in a non-geeky way?

I don’t know.

But you have to hand it to him, he’s honest.
Seriously, he says what he believes, and I guess he’s long since decided that of some think he comes off arrogant, so be it.
Comment posted by Future on April 21, 2008 at 12:54 pm (#664371)
Tango you do come off surly as heck but it doesn’t bother me. I find it entertaining.
Comment posted by Simons on April 21, 2008 at 1:35 pm (#664378)
I think John brought it out in him
Comment posted by tangotiger on April 21, 2008 at 1:40 pm (#664381)
Why not tell me exactly what it is that I said, and how it is that I said it, that makes one read it in such a negative light. Summary conclusions are fairly pointless, aren’t they?

And, is it possible for one to read that thing in a neutral or positive light? And if so, why not do so?
Comment posted by tangotiger on April 21, 2008 at 1:47 pm (#664387)
And, please, don’t try to parse my comment 8 for anything other than it is: a straight-forward, matter-of-fact question. Please read it that way.
Comment posted by Alex Nelson on April 21, 2008 at 1:53 pm (#664397)
For what it’s worth, Tango’s not a dick. A dick wouldn’t take the time out of his day to answer our questions at all.

My personal take: I think he’s just very precise in what he says, which can come off as a little cold or arrogant when really he’s just trying to be very clear.
Comment posted by John Peterson on April 21, 2008 at 3:06 pm (#664476)
Agreed. Thanks again for doing the interview, Tom. If you come off as surly, it’s only the arrogance of the great.

I’m still not sure about the basis for the claim that “teams are paying free agents extremely well.” The calculation of how much a player is worth is based on how much teams are paying for free agents. So it is not really possible for the majority of teams to be over- or under-paying free agents, since the basis for how much free agents are worth is the very contracts that major league teams give to free agents.

So I would not be declaring that there is rampant mis-payment of major league players, just wondering why some players like Juan Pierre and Luis Castillo get silly contracts. But Tom seems to want to avoid anything anecdotal or colorful, leaving the salvos to the insane bloggers of the world.
Comment posted by tangotiger on April 21, 2008 at 3:36 pm (#664504)
Alex/10: that is exactly how I intended. Truth, precision, clarity, over flowery, baseless, b.s.

I guess arrogance is like pornography.

John/11: right, exactly. On a pure value level, the average free agent is getting roughly double what he should be getting. But, this is true of most free agents. So, those teams that have figured out that free agents are way overpriced simply stay out of the bidding. And those that remain, most adhere to a fairly strong WAR/salary relationship. Every now and then, you do get some bad signings, like Carlos Lee. But, within the confines of its market class (free agency) the misses is not that much different than pre-free agency (Justin Morneau).

And to the extent that the teams misvalue players to any great degree, it’s for the reasons I cited.
Comment posted by Tim in LA on April 21, 2008 at 7:44 pm (#664680)
If you are not sure, I’d appreciate the benefit of the doubt.

I did give the benefit of the doubt; that’s why I asked, rather than allow my impression to crystallize in silence.
Comment posted by tangotiger on April 21, 2008 at 8:58 pm (#664913)
Reminds me of a Tommy Lasorda story. His pitcher is livid with a call, and Lasorda, in full earshot of the umpire yells to his pitcher: “Just tell me you didn’t call him an a–hole. Please, just tell me that you didn’t call the umpire an a–hole. You did not call him an a–hole did you?”

Asking people “is it me, or is he being a dick” is not something that you would ever say to someone’s face and think that you are being inquiring is it? I mean, would you go to someone and actually say: “Hey, is it just me, or are you being a d-ck?”
Comment posted by Tim in LA on April 21, 2008 at 9:26 pm (#665013)
No, if you can see someone’s face you probably have enough nonverbal cues at your disposal to able to tell one way or the other.

The real life analogy is more like this: A guy walks out of the room and you say to whomever’s sitting next to you, “Is that guy a jerk, or do I just not get his sense of humor?” It’s a legitimate inquiry, and I still don’t really know the answer. Not that it matters much — I can respect and appreciate your work without regard to your personality.

If I realized you’d be sifting through the comments, I wouldn’t have said anything; based on your blunt and dismissive replies to the interviewer, I didn’t think you had the time, or that you’d particularly care if you did see it. At this point it might be hard to believe, but my intention wasn’t to offend.
Comment posted by tangotiger on April 21, 2008 at 10:01 pm (#665182)
I spent at least a half hour replying to John, so I had some time. That should have been kind of clear based on my responses. And we went through a couple of iterations as he had some followup questions. And, if there are some questions to which you think I may have been dismissive, you can simply ask me to expand on the answer, and I’d be happy to do so.

Given the choice between discussing baseball and the tone to which I discuss baseball, why don’t we talk about baseball?

So, what exactly was I light on in my response that we can talk about?
Comment posted by Tim in LA on April 21, 2008 at 10:49 pm (#665258)
Well I appreciate your taking the high road. Maybe it was me that was being a bit of a dick.

If you really don’t mind answering, I was very curious about WHY you’re not interested in grading prospects and projecting from the minor leagues. It seems to me that that would be a very fun challenge for a statistician to take on. Furthermore, it’s probably the kind of evaluation that means more to the success of a franchise than any series of in-game decisions possibly could. Is it that you don’t think accurate projections are possible? Or that there just isn’t a large enough body of consistent statistics to dive into yet? It’s a very peculiar stance, and I think more than worth elaborating on.

Also, the question about your favorite player — are there any players you especially admire because of the statistics…maybe someone that’s been undervalued, or maybe a player you happen to know who pays attention to his own statistical analysis? Do favorites develop in any way, or does the beauty of scientific neutrality overwhelm all other considerations? That seems to be more the heart of the question.
Comment posted by tangotiger on April 22, 2008 at 12:32 am (#665305)
John had only asked if I do, and I said no. (To be clear, I said no to his three sub-questions, and not no three times for emphasis.) As for the reason, it’s purely one of lack of organized data.

MGL does a decent job on MLE. For example, here are his MLE for 2001-2003. Rally Monkey (Chone forecasts) also does it, as does Dan (ZiPS) and Nate (PECOTA). I won’t be able to add much in comparison, and am even less inclined if I have to spend the time to collate and organize the data. I spend so much of my time on other stuff, that I’d rather offer something new, than just be another voice. That may change as a minor league database may be upon us soon. I love working with baseball data, so my position will change in response.

In a word, you could say I’m lazy. Or efficient. Not sure which.

As for players, on a purely baseball-level, it’s Tim Raines. I even have a site dedicated to him! And he noticed (his lawyer emailed me), and we got an interview out of him. That was really cool to me. I can say that with no active player do I feel any kind of emotional attachment that could even be a smidge in comparison.

I enjoy watching Mo and Ichiro because they are really different from their peers. But, I wouldn’t say they are “favorites”. I really don’t care if they succeed or not, but rather just enjoy the battles they put up.

Brian Bannister is becoming a favorite, but that may be a May-December romance. He’s talking DIPS of all things. But, I don’t know if that’s going to wear off.

Like I said, I’m too old to have a favorite. I just love baseball.
Comment posted by Peter H on April 22, 2008 at 10:19 am (#665382)
Thanks for taking the time to do the interview, Tom. FWIW, I had no problem with your tone.
Comment posted by tangotiger on April 22, 2008 at 2:58 pm (#665694)
Thank you, I actually enjoyed it.
Comment posted by Simons on April 22, 2008 at 6:08 pm (#666371)
Rock was the man.
Comment posted by Tim in LA on April 23, 2008 at 9:26 pm (#667488)
Thanks, Tom. First answer was very interesting, second very cool to hear that about Bannister. There’s already a fondness for him around here.