A chat with my smart and funny friend Patrick who makes his own tourney brackets
How ‘bout those Jayhawks? [THEATRICAL PAINED GROAN]
We’ll get to a post mortem at some point—we always do. In the meantime, let’s talk NCAA Tournament, which is fun for most people.
Patrick Sullivan has been mentioned in the newsletter before, and for good reason. He’s one of my oldest friends and one of the best thinkers I know. He’s always built his own brackets, which is something that fascinates me. While he’s taking a slight break this year—he has two male children aged 3 and 5, the prime ages for terror—I called him up to talk about the process of determining who’s worthy and who isn’t.
The morning after the [shudders] Cincy loss, I called Patrick up to talk about the puzzle of the NCAA Tournament seeding, what he would change and most importantly for the gambling community, which teams he likes based on his research.
Patrick is a teacher, another thing I admire, teaching high school chemistry and physics in Kansas City. Recently, tragedy struck a family at his school and their community is raising money to support them. I’ll be contributing this week’s subscription income to that. Join me if you can!
First of all, how are the brackets looking? How do you put them together? What’s your approach?
PATRICK SULLIVAN: I have a spreadsheet that I've made that I have to enter a lot of the data into myself. I tried to [have data update/load automatically…however in Google Sheets, you can only extract so much information before it cuts you off. And so I've had to do it on Excel and that has its own problems. I've taught myself how to do all the things on Excel, so that's part of the issue is it just takes up a lot of time. And so because of that, I haven't been able to keep it updated for this upcoming year. I just haven't had the time.
Tell me historically when, how much were you taking in? What kind of things were you consulting and what were your resources?
PATRICK SULLIVAN: My main thing is… sometimes the analytics get a little too specific. Yeah, sure. And so I wanted to make one that was just a ranking system based on your actual wins. So when I first started doing this, I just tried to build on that and I just started building it and I just started going for it. My spreadsheet became incredibly convoluted. But I came up with a system that was good enough that most years when I kept track and then the actual brackets would come out, it'd be pretty good. I'd have a lot of the teams in there that were supposed to be in there.
So that gave me confidence to continue with it. I basically tried to get a team's schedule in the spreadsheet, and then I wanted to give a score for each result, based on opponent and location. I remember doing this before they started doing the quadrant thing, because I thought, What could we all universally accept as like a more valuable win? That would be obviously on the road. So road wins count more, neutral sites are neutral, and then home games don't count as much. I also had the teams broken up into 4 categories: 1-50, 51-150, 151-250, 250+...based on where an opponent ranked into those categories, there would be a multiplier of a sort, depending on if it was a win or a loss. So, wins against a 250+ team contribute positively to your score, but only a little… Whereas a loss against a 250+ team is a significant negative.
I just needed something just to generally rank teams. And then I think on top of that, I also have several different ranking systems that involve that point total, as well as how well have you done in the last 12 games, and other factors like that that are important going into the tournament. I just play around with those, putting varying weights on different factors until I get an output I feel good about. Then I’ll see how that does the next year, make more iterations and try to keep track which one does the best year-to-year.
What about the puzzle of the brackets is intriguing to you? Are you trying to solve a problem? Basically… why do you do it?
PATRICK SULLIVAN: So for one, I mean, I just love puzzles in general. So part of it trying to figure things out.
But I think also just because so many times, even before I started doing this, just like I watched Selection Sunday every single year growing up, my parents, we would have a blank bracket and we'd fill it out as they announced the teams. And so I think that obviously that started out with that, but then there'd always be teams that's like, why didn't this get team in? And there's always a big debate and I'm like, well, let me try see if there's a way I could figure it out, because I'd always root for those mid-major teams that would get in. Unless they're playing KU, I'm always rooting for the underdog. And so I love seeing those teams get in, and so I just was like, I bet I can come up with a pretty solid ranking system on my own.
So, now questions for the gambling community… what sleepers do you like?
PATRICK SULLIVAN:I like McNeese State a lot. I've watched 'em a few times in theSouthland. I’ve been a firm believer in… it doesn't really matter your schedule, if you win 30 games in a season, you’re a good team.
Colgate…This is their 5th straight NCAA tournament appearance. Experience matters a lot in these games.
Stetson… it's their first NCAA tournament. You just never know. They could lose by 50, but I’m watching out for them. Sure. But sometimes anybody could lose by 50. A team that I think should make it, but might not, is Indiana State.
I don’t want KU to draw James Madison. Who else should we watch out for?
PATRICK SULLIVAN: I don't want to see Oakland, South Dakota State, Charleston. Don’t give us a mid-major.. Give us somebody like Syracuse who always sneaks into the tournament. This year, teams on the bubble that would be ideal are like Colorado, or Virginia. Regardless, we need to put the ball in the basket to win games.
Top to bottom, the SEC kind of rivaled the Big 12. normally I'm like, okay, SEC, blah blah blah. But this year they really were good. I don’t want to see an SEC team, even if it’s someone like Texas A&M, or Florida, because they played in a good conference this season.
Now for the big question… what happened at KU this year, in your opinion?
PATRICK SULLIVAN: I mean, for one, I think Dajuan not playing his best affected us a lot. If we were going to be successful, we would need him to play consistently good. We might have needed Dewan to be more of a scorer, and he can do that sometimes, but I think he needs that guy, that other guy that can spread out the floor, open things up for Dajuan, because I don't think he's very good at creating offense on his own.
And then I think also, I think just because of the way our team is structured, we were easier to guard. I just think that we just never figured it out on defense. Furphy showed improvement on defense as the season went on. But I mean, Dickinson's a liability on defense. Timberlake's a liability on defense, although he showed he didn't play great on Wednesday night, but he showed some defensive plays that I was like, okay, okay, he’s a good athlete.
It was probably just bad luck to be very frank with you.I mean sometimes that’s how you win the NCAA tournament, right?
This interview had been edited for brevity and clarity.