Jeff Green > Kevin Durant

Plus, something's rotten in the state of Massachusetts

Welcome back to Take Week on The F5. If you didn’t read yesterday’s newsletter then here’s what you need to know:

Every day this week I’m publishing takes from two anonymous friends of The F5.

Here’s the menu for today:

• First, a data-driven look at the hidden impact of Jeff Green and other counterintuitive findings about the NBA.

• Then, there is something rotten in the state of Massachusetts.

- Owen Phillips

---

Jeff Green > Kevin Durant

All in one metrics are great because they strip away context, insight, and nuance from player performance and replace it with a single number that you can use to drive engagement and farm clicks.

I have created my own all-in-one metric, using a complicated formula painstakingly refined and honed to match what I see when watching games. The formula has three main components, which are added together:

• Component 1: [Field Goal %] x [Field Goal Attempts] x 2

• Component 2: ([Usage Rate] x (([Team Field Goal Attempts] + [Team Free Throw Attempts] x 0.44 + [Team Turnovers]) x ([Minutes Played] x 5 / [Team Minutes Played]) - [Field Goal Attempts] - [Turnovers]) x (25/11) x [Free Throw %]

• Component 3: [Three Point Shots Made]

I call this metric Player Overall Impact to Net Team Success, or POINTS.

Jeff Green is my favorite NBA player of all time. And it just so happens I can use my POINTS metric to prove he’s the best NBA player to emerge from his draft class, even better than Kevin Durant. I have summed up cumulative career POINTS for both players.

Sometimes if you want to tell a story, you need to visualize your numbers. Here are cumulative POINTS by season for Jeff Green and Kevin Durant:

Charts don’t lie. Green > Durant.

Three Point Shooting is Beta

Analytics has ruined the game I grew up with. Today’s soft NBA players forgo exciting, physical post up play and midrange artistry in favor of becoming three point merchants. The nerds will argue that three point shots help teams win. But I have done my own research. And that is false.

Let’s focus on when games really matter, the 4th quarter. Contrary to what the blog boys will tell you, teams that shoot more threes in the 4th quarter are not winners. In fact, winning teams are less likely to shoot threes, according to the chart above.

Home Court Advantage in the Playoffs - A Just So story

I have one more exciting discovery to share. This one is regarding home court advantage in the playoffs.

Everyone knows home court has heightened importance in the playoffs, but did you know that it is strongest, by a fair margin, in the first game of a playoff series?

Here is the average scoring margin for the home team in NBA playoffs series, split between the 1st game and all subsequent games.

After the first game of a playoff series, the impact of home court seems to wear off considerably. Two straightforward conclusions emerge from this data:

1. Playoffs are a stressful environment. Away teams are clearly rattled at first by having to play for such high stakes in unfamiliar and hostile territory

2. As the series progresses, the away team becomes more accustomed to their opponent’s venue, and they play with more confidence

It’s just so.

- Anonymous Koala

---

Boston is for Cheaters

My favorite analytics professional is Warren Sharp. I think he’s one of the smartest people working in sports. Sharp writes about football, but my favorite sport is basketball. So I trained an A.I. model on Sharp’s writing to help me reveal interesting insights about the game of basketball.

And wouldn’t you know it, I think I’ve also found evidence that a professional sports team in Boston is guilty of cheating. Just like Sharp, when he proved beyond a shadow of a doubt that the New England Patriots were guilty of deflating footballs over a decade ago.

But first, some context.

Analytics experts mostly agree that NBA defenses have almost no control over opponent three-point shooting. They say it’s luck that determines who has the best three-point defense vs. who has the worst.

“The vast majority of work done on the subject strongly suggests that teams do not have much control over opponent shotmaking on jump shots,” is what Seth Partnow, the former director of basketball research for the Milwaukee Bucks, wrote about opponent three-point shooting in an article for The Athletic.

According to Partnow and other experts, most threes in the NBA are taken when the closest defender is more than six feet away. That means most threes are wide open. So it has to be luck that separates the best three-point defenses from the worst. (Although some people who are really into Star Trek like to joke that the best defenses must use mind control, which they refer to as “Jedi Defense”)

It turns out, the Boston Celtics are one of the best — or luckiest — teams when it comes to defending the three. This season, the Celtics rank third in opponent three-point percentage. Opponents are shooting only 34.8 percent from three against the Celtics, which is 1.2 percentage points lower than league average.

But that’s just the beginning.

The Celtics have had a top five three-point defense in 16 of the last 18 NBA seasons. No other team has had more than six top five finishes during that time period.

This is very suspicious.

If opponent three-point percentage is largely the result of chance, then where you rank in this metric should be largely random. That means a team should be just as likely to finish 1st as 30th in any given year. But Boston’s never finished below 22nd since 2008. Meanwhile, they have finished first four different times.

So how do we explain this anomaly?

We can’t. Therefore they must be cheating, according to my Sharp A.I. model.

Let’s use some math to show how improbable it is for the Celtics to have finished top five in opponent three-point percentage 16 times in 18 years.

Imagine you have a 30-sided die. The odds of you rolling between a one and five is 16.6 percent. That’s the same as your chances of finishing in the top five in opponent three-point percentage if it is really just luck — which again, the experts say it is.

So all we need to do is calculate the odds of rolling less than five with a 30-side die 16 or more times in 18 rolls. I asked my Sharp A.I. model to crunch those numbers for me below:

“Almost zero,” according to my Sharp A.I. model.

The only plausible explantation is that the Celtics are cheating. Even though I don’t how, I’m certain of this. What else could possibly explain such “an extremely rare event?”

Maybe they deflate basketballs in Boston, too

- Anonymous Walrus

---

See you tomorrow (April 2nd) for more takes