On this episode, Mark Simon (@markasimonsays) is joined by 38-time Jeopardy champion and baseball aficionado Matt Amodio (@AmodioMatt) for a lengthy, fun conversation about Jeopardy strategy and analysis, Matt’s favorite players, stats, and baseball interests, and the work he’s currently doing in data science.

A transcript, lightly edited to shorten some of the questions, is below.

 

Mark Simon: On today’s show we’re joined by Matt Amodio, who just recently concluded a 38 day run as Jeopardy champion, where he won $1.5 million. He’s currently a doctoral candidate in artificial intelligence at Yale university. And what makes him particularly pertinent for us? His hobby is studying baseball stats!

 

Hey Matt, how you doing?

 

Matt Amodio: Fantastic. How about you?

 

Baseball Talk

Mark Simon: We’re great here. So today, we talk the ins and outs of Jeopardy, Matt’s work in the field of big data and we talk about baseball.

And we start there, let’s establish some of your baseball credentials. People who listen to this podcast regularly know that when we have players on, we always ask the same first question.

Can you tell us the story of a great defensive play you made as a kid?

Now, I saw a picture of you at an Akron River Ducks game with a very nice baseball glove. So I’m going to guess that there’s a play in your past that you made, that you can tell us about.

Matt Amodio: Absolutely. Yeah, so I was primarily a pitcher, but I played some third base too.

I was never good at hitting. And so anytime I was in the lineup and playing third, I felt very nervous. And I had to make my value with my glove. And I remember one of those pop-ups in short left field, down the line, you have to run backwards and Omar Vizquel made them look so easy, but they’re actually really hard because you’re worried and holding your head back to see the ball behind you. And I made an over the shoulder basket catch. I think I’ve missed about five or six of them  exactly like it, but there was one I made and that one I’ll always remember.

Mark Simon: Very nice. So give us your baseball origin story.

Matt Amodio: I have been watching baseball since before I can remember, an Indians fan who grew up around Cleveland. Growing up in the nineties was a very fun time to be an Indians fan. I didn’t know what it was like to go to a stadium that wasn’t sold out until I was like 14. And so it was just the right time and the right place and I watched winning teams over and over again.

Mark Simon: So nineties, there are a number of different ways that you could’ve gone for a favorite player. Who was yours?

Matt Amodio: Kenny Lofton.

I didn’t properly appreciate him at the time. So I love that he was always good with the glove, but he also carried himself with a flair that appealed to me as a kid.

So anytime he would see Ball 4, a fastball just off the plate outside, he would flip the bat enthusiastically and then take his gloves off right in front of the umpire. And he got called on strike three looking a couple of those times. But when he just predicted the call ahead of time, it was so fun.

Mark Simon: People who studied the numbers will tell you that Kenny Lofton actually has a pretty good Hall of Fame case, particularly when you compare it to Tim Raines. Make the case for Kenny for the Hall of Fame.

Matt Amodio: Yeah. So I, this is where I was saying I could not appreciate him properly at the time because I loved those singles that he would leg out with his amazing speed.

What I did not know is how valuable his walks were. And when you have somebody who is walking close to a hundred times in a season who also has the speed to leg those into stolen bases. He has more value when you look at the on-base percentage than I ever thought before. So he’s putting up the OPS+ that put him as a well above average hitter, and then also playing a great defensive center field.

At the same time, the WAR numbers don’t lie. He belongs in the Hall.

Mark Simon: Who’s your favorite obscure Indian/Guardians player?

Matt Amodio: Yeah. So I really can’t explain it, but I always loved watching Einar Díaz. So he didn’t do particularly well, but he always looked in control of the bat and he’s got a big body type. And so it just looked like the ball would carry off of his bat. It was always fun to watch.

Mark Simon: Were you at Game 7 of the 2016 World Series?

Matt Amodio: I was not, I was at Games 1 and 2, which were the right ones to be at there if you’re an Indian fan.

Mark Simon: You also said that your favorite ballpark was Progressive field. The I’m curious why you would say that.

Matt Amodio: You should have heard the screaming that was going on inside my head as soon as I said Progressive Field, I was like, no, can we take that back and retape?

I should’ve said Jacobs Field, I never call it progressive field to my friends or family, but just, I felt like being PC in the moments that those traditional names never die.

Mark Simon: Have you ever sat in left field with the drummer, John Adams?

Matt Amodio: Yeah. So the bleachers are a great seat, but then also John Adams is so welcoming to the random strangers who just want to shake his hand and get to meet him. And so it’s a cross off the bucket list type item.

Mark Simon: You also mentioned that you’re a ballpark traveler.

What’s your favorite ballpark?

Matt Amodio: Yeah, I think the Giants park is my favorite non hometown one to see because there are so many great views. They have a lot of aisleways where you can watch. When I go, I like getting a cheap seat up in the upper deck and then not spending a single second at it and walk around.

I love the ones where you can get a great view from the aisle somewhere. I watched several lenings from from the right field porch there and it’s fantastic.

Mark Simon: You list baseball data as a hobby. So just give us the brief overview. We’ll come back to it, but I want to just get a brief overview.

Matt Amodio: Yeah. So this has been like a long-term thing in my life, and you can trace my growth through the baseball stats as well. I remember when I was in high school, all I would do is look at Baseball-Reference and look for black ink and sort by leaders and the main stats.

And I thought that I was doing baseball stats when I did that. And then gradually I got a little more comfortable on a computer. And so I would get some Pitch FX data and look at a sequencing. And I was also studying probability at the time. So I liked saying, okay how do probabilities of this pitch change with nobody on versus with runners on and that kind of thing.

And now I read stuff that’s more advanced than I could do. That’s for sure. Super interesting. And I’ve always been most interested in pitching aspects as a pitcher myself.

 

Jeopardy

Mark Simon: Let’s segue to Jeopardy.

On your first show, they go to the first commercial break come out of the break and you get interviewed.

And your fun fact was that you self-identified as a coward because you were afraid to take the escalators in Washington, DC, where they’re very steep. But your game, for the most part, is anything but cowardly.

And I’m gonna read you a scouting report that I’ve compiled, that I think describes it pretty well.

As a pitcher,  you’re Cliff Lee. You worked down in the zone, you finish up top, you take your time and think things out, but you can change the tempo of the game quickly. Like changing categories. There’s not much muss or fuss to his game (thanks Alex Vigderman for helping on that one).

As a hitter, you’re José Ramírez and Jim Thome. José Ramírez, because of the high batting average. You hit on something like 90% of the clues that you buzzed in on. If you buzz in, you get it right.

Thome, because you like to take big swings. That’s intimidating. You’ll live with the strikeouts. You don’t get a lot wrong. The home runs will be more valuable over time. And you hit a lot of home runs.

Defensively, you’re Franklin Gutierrez with great range. You’ve covered the whole field of questions from baseball to the Bible, geography to music.

And as a game manager, you’re Terry Francona or, as we wrote, Gabe Kapler.

When the situation calls for it, you manage smartly. You know the rule book inside and out. You’re well-prepared thanks to Wikipedia. Fair?

Matt Amodio: That’s better than fair. That is the highest praise I’ve heard in a long time. And I especially loved the Franklin Gutierrez pull. That’s a little deep. I approve.

Mark Simon: Explain how you played and manage the game for us.

Matt Amodio: Yeah. So I think you put it well in a lot of those cases.

So I think that you really have to view the wagering chances in the game as your opportunities, and either you can use the opportunities to put the game out of reach and that might seem risky, but actually just as risky to see an opportunity and in the face and pass it up.

So a lot of people come on and take a Daily Double that they land on a, in a late game situation and say ‘What if I don’t know?’ And they sit on it. They bet a couple thousand and they get it right. And later on in the game, they lose control a little bit. Maybe they get Final Jeopardy wrong and they end up losing the game.

All of it hinged on getting a Daily Double and betting too low. So I went in saying, I’m not going to let that happen. I am going to take an opportunity and take a swing. I might go down, that might happen, but I’m not going to look at a fastball down the plate and go down that way.

And so I think that’s just the right way to do it. And it seemed to work.

Mark Simon: And I’ll run down, just list the different approaches that all relate to that one. You started from the bottom row and you worked across, and then occasionally I noticed you would skip one, but I noticed that you always worked across. What was the value in that?

Matt Amodio: So a couple of things. One thing I think when you go in there, you’re going to be a little intimidated and you might be a little off your game and then gradually get more comfortable.

I wanted the most expensive clues, the highest leverage stuff to go when my opponents are uncomfortable. Let’s make most of the money when they’re rough and then make the cheap clues when they’re in the groove. So that was a big thing.

And then also going across the board, I found from practice, I can bounce from Bible to geography, to sports, to movies pretty quickly. And other people, I think that’s something other people struggle on.

And so I view it as if I’m a .350 hitter and somebody else is a .300 hitter, I’m okay taking this down to being a .330 hitter, if their drop is even worse. So, if I can make the terrain rough for everybody, but I acclimate a little bit better, even if it’s hard on me, I’m still going to be better off for having done it.

Mark Simon: With Daily Doubles, if you got them in the opening round, you went all-in automatically. Even if the category was – cringe – motorcycles. In Double Jeopardy you basically took the approach that if you could end the game right there, you go big. If you had a very comfortable lead, you went relatively small.

I want to get to one other thing. You buzzed in sometimes not knowing the answer. What was that about?

Matt Amodio: I buzzed in a lot of times not knowing the answer. So I would always read the clue in my head and then just say ‘Do I recognize it? Are there clues I recognize?’

I’m a really visual thinker. So if I could picture the answer, then I would say ok, time to buzz in. And sometimes I would picture the answer and immediately the word would come to me. And then that’s 10 seconds before I have to buzz in, no problem. But other times I’ve pictured it and said, ah, the word isn’t coming to me, but it will, so I press the buzzer.

And then I have five seconds. Hopefully the word will come to me. A lot of times it did. Sometimes it didn’t. Including in my last game, several of those where I know that wasn’t spawn, but venom just wasn’t popping into my head.

Mark Simon: We’re going to talk some Jeopardy situational stuff, just cause I think it’s pretty cool.

In Game 10, you’re playing and it’s a tough game. You have $13,000. Your opponent has a little more than $10,000. You hit a daily double on Australian History and you said, ‘Oh my goodness.’

And you bet everything. That’s living on the edge. What were you trying to do?

Matt Amodio: Yeah. So I think if I remember correctly, this is the first time that I had to actually put my money where my mouth was on this.

So I played this situation out in my head prior to going on the show and said, ‘If you have the chance to do this, you’re not going to lay down and I hope that the rest of the game goes your way. You’re going to try to put the nail in the coffin here. But this is not just a strategy. Now, this is actually doing it.

And so I thought the situation was pretty good for me because it’s a geography question or a history question and then it’s also somewhere in the middle of the board. So it isn’t even at the bottom level of difficulty. And then also one of my brothers lives in Australia now.

And I said, okay if I underbet on this, and then I lose because I didn’t bet enough, I’m not going to be able to visit him ever again. And so I just gotta go for it.

Mark Simon: That’s awesome. And you got it right.

You could feel the air was let out of the balloon right there. And you won handily. There was no catch-up.

Another game: This was much later in the run. You had a game where you were negative going into Double Jeopardy. Thanks to, I think it was motorcycles and a little bit of a rough go on Muhammad Ali.

You go into Double Jeopardy and in three clues, you undid all of the damage that had previously been done. The people that played you, one of them commented on Reddit about the intimidation factor of playing you and the stress of playing you. And all of a sudden you had this momentum and it was boom, up the mountain and you were gone.

Is there a momentum component to Jeopardy?

Matt Amodio: I think there’s gotta be a momentum component, especially with the buzzer. So that’s one big aspect of the game that often goes untalked about. There’s a rhythm to it. And if you get in the right rhythm, you’re going to get in on 1. 2 3, 4 in a row.

And that builds momentum, builds confidence. It makes other people think they’re doing something wrong. And so they might overcompensate and get themselves out of their own rhythm. And so there’s a lot to that momentum.

Mark Simon: The buzzer requires the right timing. And in some cases, a timing mechanism, sometimes you can see it You appeared to have the timing mechanism of what I call – the hip pivot – where you twisted as you were buzzing in trying to time it to the last word that the host was saying.

This seems like coming up with a batting stance. How did you come up with your timing?

Matt Amodio: Absolutely. It’s like a batting stance because I was imagining myself in the cage as I was doing it.

I remembered I would always have a couple of touchpoints on when I was swinging. You do a toe tap to start things. You bring the hands up. But my big analogy here is that you have to be ready for a fastball, but also be ready to adapt to a curve ball if you identify it mid-pitch.

The buzzer is not done by a computer. it’s done by a human who presses a switch and that turns on lights.

And then you can buzz in. So you can’t be ready for a 95 mile an hour fastball because what if the guy is just a little bit later than usual, that one turn. And so I would anticipate it, but then be ready to adjust. Having a little bit of curl and physical movement really helped me get primed, but also adjust.

Mark Simon: Do you think that being a baseball player helps? And I asked that because a couple of years ago at the SABR Analytics Conference, they did comparisons of people buzzing in to light as a study and they found that baseball players were considerably better at pressing a buzzer than your average human beings.

Matt Amodio: I’m not sure it helped. I think they’re at least strongly correlated. So you need to have good vision, fast-twitch muscles and the ability to coordinate the two. And it doesn’t surprise me that baseball players are better at that because it’s necessary.

Mark Simon: Unquestionably.

You hit a final jeopardy in mythology because you remembered a school project that I think you said you had done in fifth grade. And I’m curious about the memory retention skills of a Jeopardy player.

And again, I’m thinking baseball, thinking about pitchers that you faced over time and things of that sort and how your brain compartmentalizes knowledge.

Matt Amodio: Yeah. So I think that this was a very memorable thing to me. I don’t like to think about all the stuff that I should have remembered.

So who knows what I read in seventh grade that would have made me $2,000 on Jeopardy if I had just remembered it.

But it’s always great when you can tie something to a physical memory. So that, often helps with me whenever I try to remember stuff for school, as well as trivia for general purposes – that I wouldn’t just try to remember the fact – that I would try to tie it to something around me.

I listen to audiobooks. A lot. And I walk around when I do it and it’s amazing to me, five years later, I don’t just remember that scene in the book. I remember the tree and my neighborhood that I was walking past when I listened to that. It’s something that’s always been helpful to me.

Mark Simon: It’s amazing. How memory like sticks with someone.

Was there a clue or moment where you were like, wow this thing’s meant to be? Like Tolkien?

Matt Amodio: So that’s a good one. So I am very conservative in terms of getting things right, where I don’t trust myself actually knowing that until I hear confirmation that is right.

I can be really confident in something and it still happens. I could have sworn that guy’s name was JD Ballenger, not Salinger. So I’m always worried about something that I’m sure of, being wrong. The one time in Final Jeopardy where I wrote it down and said that I am so sure of this, I’m going to give a smile to the camera as it passes by was the Tolkien question. There are bigger Tolkien nerds than me, but I want to meet them.

Mark Simon: You’re the ultimate Tolkien fan. One of your fun facts was that you watch all of the Lord of the Rings movies in succession in a given day.

Matt Amodio: And the hardest thing was keeping it to just one fun fact. I could have done all 38 on Tolkien.

Mark Simon: I want to review some stats here. You had one game where you swept the bottom row in Double Jeopardy on consecutive clues. That website J-archive.com said that had never been done in their records.

You had a game where you got 44, right. That was one shy of what is the known record by Ken Jennings. You pitched two perfect games.

You had a 40-question perfecto, which if I’m not mistaken was your 15th game in three days. Your first 19 wins, you averaged $34,000. Your last 19 wins, you averaged $46,000. You got smarter over the break time in-between seasons.

There was another where your 10th game in two days was probably your most exciting game, where you had to hit a daily double for everything. And even still, you wound up at the end of Double Jeopardy with exactly twice as much money as your opponent (so it was still a game). Nicolle Neulist.

At the end of Double Jeopardy,  you looked like you were ready to pass out.

So explain how Jeopardy is a game of stamina, especially as a pitcher, I think you can appreciate it.

Matt Amodio: I was ready to pass out. Yeah. One thing that I didn’t even really realize until I went on, is you don’t just tape a show and then 24 hours later that show airs.

So you’ll do a Monday night show and then on Tuesday night, the next game’s played. But that’s not how they’re taped. They’re taped in rapid succession. So you play a game. Sometimes you have a lunch break or something, but often it’s just 15 minutes later, you’re playing the new game and you might have won by a ton.

You might’ve just squeaked by, but you start a 0, 0, 0 tie with the other two contestants in the next game. And so I had some embarrassing final Jeopardy’s where I lost more money than I make in a year on a trivia question. And I had to be ready to beat two other smart people. 15 minutes later, I had games where I was close to the record for the highest game. And then 15 minutes later had to be ready to start 0, 0, 0 with two other opponents. And so a big part of the stamina is knowing that you can’t coast on the previous day, or you can’t obsess over your mistakes from the previous day. You always have to focus on the one question at hand and then they just keep coming for 12 hours straight in a day sometimes.

Mark Simon: Yeah that’s insane. It’s like if a reliever had to pitch in three consecutive doubleheaders.

Matt Amodio: I think that the inning break is a great analogy because you might get out of a basis loaded situation and just squeak by, you held your one run lead. That’s great.

Now you have to come back and somebody could take you out of the park if you leave a fast ball right down the middle to start the next inning. So you have these stress situations and then a short break. But then any one pitch could do you in in the next inning. And so you have to, you can’t just give a get-me-over curve ball at any point, or you’re going to be done. You always have to be on your game

Mark Simon:. And we go back to your very first game in the 38 straight wins. This is super interesting to me and this combines my baseball nerdery and what amounts to my Jeopardy nerdery. Matt trailed his opponent by nearly $11,000 with 17 clues left and no daily doubles left on the board, which is an important fact.

I don’t know if you realize the score at that point, but it looked bleak. But like it was for the 1986 Mets, the 2004 Red Sox and the 2011 Cardinals. And I guess we could even say the 2016 Indians in the eighth inning of Game 7 in 2016, this one wasn’t over till it was over. And not only did you come back, but it required something unusual to happen at the end.

And I’m fascinated by little things that happen along the way. There were a couple in this game. If a traffic engineer in Boise, Idaho, who is a very sharp opponent, doesn’t confuse Andrew Lloyd Webber and Stephen Sondheim at the end of the Double Jeopardy, he’s winning at the end and going into Final Jeopardy. And your run never ever happens.

What do you remember about your first win?

Matt Amodio: One, one part where the analogy breaks down to baseball though, is that if you’re on a roll and even if it’s the bottom of the ninth, two outs, you get to keep playing until they get that last out. So doubles, homers in a row, keeps everything going.

There’s a clock in Jeopardy. There are only X number of questions to play, but then there’s also a literal clock that keeps the time going. And we don’t know exactly what that is, but you have a sense of it. And you got a one-minute warning. So you have a sense of when exactly is one minute to play. And so I remember frantically playing to the clock in a way that as a baseball player just felt very foreign to me.

Mark Simon: So relating this all to baseball, I like a stat called win probability. What are the team’s chances of winning at any point during a game. You can do it for baseball, football, basketball, presidential elections.

You can do all those things with win probability and with Jeopardy there’s a website, TheJeopardyFan.com run by Andy Saunders. That after each victory, it list, the winners chance of winning the next game, 2, 3, 4, et cetera. But I like to go deeper than that. I like in-game win probability.

And I would guess Matt, that your in-game win probability was comparable to what the Indians win probability was when Rajai Davis faced Chapman in Game Seven. I’m thinking something like 5%. That led to a conversation I had with a colleague about what factors should go into a Jeopardy in-game win probability.

And I’m curious if you have a thought on that.

Matt Amodio: Absolutely. So one factor has gotta be part of the board left. So you have more innings to play. You have more chances to come back. So how many questions are there, but how many Daily Doubles are left to reveal. Now you could get super intricate in this because Daily Doubles are going to be more likely at the bottom of the board.

And so if you picked off bottom of the board clues and not found them, that means that one $1,200 square is almost certain to be the location. And so you can adjust this win probability matrix or the Daily Double location matrix square by square each time, but then you also have to keep in mind your chance of getting it right. And where in the game you are when you got it.

So if I’m in a position where I need to bet it all, and then I get it wrong. I might be done for. If I get that same Daily Double late in the game where I only bet a little, because I already have a cushion, I can get it wrong and still win.

And so there’s so much game dynamics because your actions are going to change based on the game play, which is different from baseball, because you always want to hit grand slams, whether it’s the seventh and you’re up by one, or it’s the second and you’re up 10 already.

You always want to just score more and give up less.

Mark Simon: One of the things that we were trying to ascertain was the idea of knowledge base, meaning that if you had gotten three right in a category, let’s say it was Muhammad Ali, and you got the four, the eight and the 12. Does that mean that you have a higher probability of getting the 16 and the 20, because it’s established that you know Muhammad Ali?

Matt Amodio: Ooh, that’s a good question. I I actually think that it might get a little more complicated because they put some questions in the top that I think are designed for the kiddie pool. So if you’re just dipping your toes into Muhammad Ali, maybe you’ve never seen a boxing match in your life, but they’ll ask you about his famous catchphrase. And so somebody who doesn’t know anything about boxing at least knows the catch phrase. So I think that there might be some component to that, but I wouldn’t be surprised if it’s a smaller relationship than we would expect.

Mark Simon: also want to give a shout out to the question writers at Jeopardy, who I know you like. I love writing trivia questions as much as I like answering them. So I appreciated that you said that, and I think that they do a fantastic job.

Matt Amodio: Absolutely.

And the only sad part is they read them so quickly. And then you move off that a lot of the intricate intricacies of language and everything, you just get flown over again.

Baseball Stats

Mark Simon: So segueing to baseball stat topics.

One that you brought up to me when I asked you what you’d like to talk about is pitch tunneling. And as a former pitcher I’m curious for your appreciation of it. I was reading an article on Pitcher List the other day about Shane Bieber and Jacob deGrom’ usage of it, and the different stats that you can find on it on a place like Baseball Prospectus, or in our Bill James Handbook. We’ve got an essay on Pitch Mix Index, about which pitchers best mix their pitches. I’m curious for a pitcher/statistician take on the subject of tunneling.

Matt Amodio: I was a pitcher in school and this was something I was obsessed over before I even knew the proper word for it. So I wanted to make my fastball and curveball look exactly the same and then fork into different directions.

And you not only need the pitches to do that, but you need the location. So like a fastball up and in, and a curve ball in the dirt, aren’t going to tunnel, but a fastball on the outside corner and a curve ball off the outside will tunnel.

And so coming up with pairs of these and then working them into a whole sequence, it made me feel like I was playing chess except instead of chess, I was successful.

Sometimes I’m a terrible chess player.

Mark Simon: Can I ask what your ERA was?

Matt Amodio: It was good. In the 2s for my career.

I never threw hard and my other teammates were throwing fastballs like 15 miles an hour faster than me. And so I had to work a lot on deception there. One of my favorite things is seeing somebody like Aaron Civale, who not only has two directions that he can tunnel with a fast ball and a cutter, but what I could never do is make the baseball go in the opposite direction as well.

So his two seamer/cutter and straight four-seamer just form a three-prong tunnel.

Mark Simon: And he had a much improved season this year. Another topic that you seem to have an interest in is defensive metrics and the different things that you can tell from them.

Matt Amodio: Yeah, absolutely. With the shift, it has made this whole problem a lot harder to keep track of, because one thing I think that people would realize quickly is that the defensive shift makes some balls easier to get to than others just in terms of where they’re located on the field.

But one thing that’s interesting to me is also momentum, which you had talked about before.

So with the shift you’re also going to be moving a little bit, and on a ball towards the line, f your shading yourself up the middle is going to look only five, six feet away, but be impossible to get to cause your weight’s in the wrong position. And so defensive strategies are getting so much more complicated.

Mark Simon: That’s something that you’d like to study on the baseball side, but because you’re dissertating and doctoring that you haven’t had the chance to really look at.

Matt Amodio: A lot of the stuff that is really getting close to physics is really interesting to me, but so far above what I’m capable of, understanding that would just be a deep dive just to get myself set.

But a lot of the spin rate stuff where you can analyze spin rate as just an arbitrary column and a dataset, and then see what it predicts. But then to have something deeper and understanding like how to get a better spin rate out of your fastball, or whyfastballs up in the zone are more effective than high spin rate fastballs elsewhere in the zone, with everything else held constant. That stuff’s really interesting to me, but I haven’t delved in yet.

Professional Career

Mark Simon: After you complete your doctorate, which we’ll get to in a second, I certainly recommend the SABR Analytics Conference, Saberseminar, all these different events at which you can certainly learn more about.

Let’s get to your professional career and your professional experience. You’ve built predictive models for massive data sets in fields, such as social media, networking, natural language processing, spacial routing, cyber security, and computational advertising. Tell us about that.

Matt Amodio: So one of the cool things about my type of research is we build the models and they’re largely dataset independent.

So I know a lot of math and the math is about extracting information and patterns out of the data and really what the data is on doesn’t play that much into the models themselves. And so that allows me to move from one area to one vastly different area quickly, which is helpful.

Mark Simon: There are some differences between what I just read and what you’re currently working on, which is in the medical field in immunology.

Matt Amodio: Yeah. And again I don’t know that much about any of the things that you listed that I worked on. I learned a little bit on each, enough to get by, but don’t consider myself an expert in those at all.

Mark Simon: So how do you approach working on these things.

Matt Amodio: So I think it requires a lot of just comfort with the unknown.

So one of the things I go in to any dataset with is the expectation that I don’t understand most of the interesting stuff about it. I start looking at what you can visualize. I use common techniques to embed things in two dimensions, so you can plot them and you just try to understand it from a numbers perspective, as well as you can.

Mark Simon: What languages do you work in? I know that my colleagues are gonna wanna know that one.

Matt Amodio: Almost exclusively Python. I did other stuff in research or in my work careers beforehand, but now that I’m on my own, I don’t force myself to work with any nasty languages.

Mark Simon: Two fun questions for you. As we near the close here, you won in, I think it was April initially. And your first episode aired in July (and you couldn’t talk about it for months), and then you had to keep it from people for a month that you lost.

You are in the elite class of secret holders.

What’s the secret to keeping a secret.

Matt Amodio: Yeah. So there were different secrets there. So for the first season, I had three months knowing that I won a whole bunch of games and my streak wasn’t over yet. So that was great where I could just be ecstatically happy that I won my first game. And the people didn’t know that I’m actually ecstatically happy, because there were 17 more that you don’t know about yet.

So keeping that secret was pretty easy. The harder one was the loss. Because I am a perfectionist. No matter how many games I won, if the most recent thing I did was lose, I would feel like a loser. And so it took a little bit of skill to cheer on the wins, knowing the inevitable was coming.

But it’s like a team sport. I actually felt quite a bit like when I was in high school and I wanted my high school team pitchers to do just a little poorly so that I would get to be brought on in relief or that I would get the ball in Game 1 of our playoffs. So I wanted us to win, but there was also some little bit of jealousy. And so I know how to project a happiness that’s different when I feel on the inside.

Fun Jeopardy Stats

Mark Simon: I wanted to do something special for your appearance. We wanted to invent a stat and we’re a company that loves details. We chart everything. So with the help of a woman named Lily, who goes by the Twitter handle @OneEclecticMom, I’ve got one for you.

It’s called WAR-  Wardrobe Above Replacement. The intersection of UniWatch and jeopardy. So about an hour after you said yes to being interviewed, I went to this woman, Lily, she’s basically done in an encyclopedic work of what you’ve worn when you’ve been on Jeopardy. She’s done charts, she’s done images.

And I asked her if he could look into any connections between clothing worn and money won. I have the information in front of me.

We start with the idea of sweater jacket tie.

When you wore a sweater on jeopardy, you averaged $40,000 (11 times). When you wore a jacket on jeopardy, you average $33,900 per win. So sweater is greater than jacket. However, when you wore a tie, you averaged $46,000 per win in the nine episodes in which you wore a tie..

Matt Amodio: This is actionable information that I need to keep track of. So I have more games to play in my future. I’m going to be wearing those ties.

Mark Simon: This is silly stuff, but you give the people that they want. And on Twitter, people seem to like this stuff.

And especially in baseball, like you look at like, when the Mets went to their black jerseys, people went nuts and then they were terrible in them. People are very into apparel and I’m fully embracing of that. And we can even get into this. It’s the sabermetrics of apparel color.

You wore a pumpkin orange shirt once and won $70,000. You wore a blue plaid shirt with a white color twice and won $57,000. Those are your two highest shirt wins.

Among things that you were frequently: A maroon shirt, worn eight times, was a little over $40,000, and then a purple shirt, worn nine times was also about $40,000.

Thank you, Lily, @OneEclecticMom on Twitter.

Matt Amodio: Thank you, Lily. I love it. And we can even take a little learning lesson out of this because this is the dangers of extrapolation. One might assume that I could then wear a tie with the orange shirt underneath it, and then a sweater on top, and now I’ve got James Holzhauer’s single-day record in hand right now. You can’t necessarily stack these things,

Mark Simon:. So to wrap up, I do want to note that you won over a million dollars for yourself. You won considerable money for a number of charities

Be The Match – a national bone marrow donor program

Reading is Fundamental – literacy

Kidsmart – a charity for providing school supplies

The Robin Hood Foundation – poverty.

Alex Trebek, in his last episode, made comments to the effect of kindness should always win out. You have been first-class with people on social media to an extreme. I am very impressed with how you have handled social media.

As Andy Warhol once said in the future, everybody will be famous for 15 minutes. I know that you know that, because that quote was in a question you answered during your run.

How has celebrity been and what has it been like to spread a little kindness?

Matt Amodio: Yeah. So I’m a shy, introverted person. I didn’t know how the celebrity was going to go with me at all. I was shocked at how mostly nice people were on the internet.

That just is not at all what I expected and it made it easier to be nice myself. If I was dealing with a hundred percent haters and trolls, it might’ve brought out some negative aspects of me, but other people give support, gave me kindness to feel, and that’s what I’m trying to spread to other people. So maybe you give one little nice comment you give will make that person brighter and then spread it. And so it can easily just pass on.

Mark Simon: I recommend Matt on Twitter. @AmodioMatt. And also you can find him on Wikipedia. There’s a Wikipedia bio for Matt. Thank you for joining us.

Matt Amodio: My pleasure.