The Eye Test and the Algorithm

I am assuming you have all seen the movie “Moneyball”, correct? Early on in that movie, Billy Beane (Brad Pitt) is sitting around with his regional scouts talking about players that the cash strapped Oakland A’s can use to fill out their roster. See the full scene below:

Now, there is a lot to unpack here and me being the inarticulate fucker that I am will probably leave some massive holes in this argument. Yesterday was an interesting day on college basketball twitter, which is a fascinating place by the way. Eye test guys were trashed by everyone, including some of the most respected people in analytics who may have received so much twillatio over the years that they are believing their own legend a bit too much. Then, in what seemed like art imitating life in some sort of political comedy illustrating the left consuming itself, the analysts turned on each other. There were questions challenging baselines, metric integrity and which algorithm holds more water. I laughed, I cried, I rolled my eyes. The only thing I was missing was the popcorn.

Now, I bring the “Moneyball” video to your attention because for years, I was Grady, the analysts were Pete and the NCAA is Billy Beane. Fast forward a bunch of years, the analysts are now Grady, I am Pete, and the NCAA is still Billy Beane except they aren’t in this fight, they’re just trying to not pay anyone.

Think they aren’t Grady?

Now, let me say, I use KenPom literally every single day. I use it for metrics, I use it for my wagering previews, I use it for the podcast and I use it for show prep when I guest on other shows. For Ken to be completely dismissive of the eye test in order to rank or judge teams is incredible to me. I get that there is a brand associated with Ken and he is very well respected, but you have to admit that watching Baylor succeed and simply knowing that they are succeeding this season are two completely different things. Watching Baylor may show you that, even as the number one team in the nation, they aren’t exactly dominant. Also, no offense to Ken, Virginia was in his top 15 until nearly Christmas when they clearly were not a top 15 team.

Why even play the games? Why not just simulate the tournament and be done with it. But, isn’t 90% of the fun actually watching the basketball games? The stubbornness by the analysts in college basketball circles is equal to the stubbornness in rebuffing advanced metrics when they were introduced in baseball and college basketball in the early 2000’s. The roles have reversed, and just when I was starting to embrace them. Now I feel like a man without a country.

Yesterday, everyone wanted to argue about Stephen F. Austin being a bubble team because of their win on the road at Duke. It has something to do with WAB. A new term that I had no idea even existed until yesterday and will probably forget by tomorrow. I don’t know what it means and I don’t care. Stephen F. Austin is a Southland Conference school that is 19-3 with 19 of their games coming against quadrant three and four teams. Their strength of schedule is 327. Win your conference tournament, that is how you get in. You’re not an at large team and you’re not a bubble team. If you are considering SFA, you should consider North Carolina State, who has 7 wins against Q1/2 teams (SFA has 1) and no Q4 losses (SFA has 1). Of course, North Carolina State has three quadrant three losses, which is why they aren’t really in the conversation in the first place. So what are we even talking about?

I don’t know what the answer is, but I do know that algorithms are only as good as the people write them. I also know that there is a need to actually watch basketball in order to judge teams. To dismiss one or the other is simply irresponsible and ridiculous. Something else that I know is, we shouldn’t be fighting over who is the best at predicting an outcome that is decided by an inherently corrupt organization that is really only in it for the dollar dollar bills. Remember when we just loved college basketball because it was awesome? Let’s get back to that.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s