Brendan Kent Profile picture
data scientist working on @ray_ban x @Meta 🕶️ • previously @lyft @DraftKings @TimbersFC @Harvard
Oct 18, 2022 47 tweets 22 min read
Looking for free sports data to kick off your sports analytics project?

Here's my updated list of 40+ free sports data sources, including pre-assembled data sets and R/Python libraries 👇 🏈 Pro Football Reference (@pfref)
pro-football-reference.com
Oct 17, 2022 19 tweets 4 min read
You've probably seen charts like this showing each team's Win Probability over the course of a game.

But how is this calculated? And is it ever "correct"?

A non-technical explanation of one of the foundational sports analytics metrics 👇 Image Win probability (abbreviated "WP") is the likelihood that a team will win a particular game, expressed as a percentage.

50% WP: If the game is played 1000 times, the team will win ~500

10% WP: If the game is played 1000 times, the team will win ~100
Oct 15, 2022 20 tweets 4 min read
Analytics is increasingly responsible for coaching and personnel decisions in sports.

Why is it such a valuable tool?

Because humans have mental and situational constraints 👇 Image Sports analytics has been able to provide value beyond traditional qualitative judgments because humans have limits that can be generally classified as:

- Mental constraints
- Situational constraints
Oct 14, 2022 18 tweets 5 min read
What is Expected Goals ("xG"), and how is it calculated?

A non-technical explanation of the increasingly mainstream ⚽️ metric 👇 xG is perhaps the most ubiquitous advanced ⚽ metric, but can be applied to structurally similar sports like 🏒 as well.

In this thread, I'll focus on its application to ⚽, but much of what can be said about xG in ⚽ applies to xG in other sports.
Oct 13, 2022 16 tweets 6 min read
Want to get into business, product, or sports analytics?

Here's how to develop your technical skillset for free: There are several key technical skills for analytics, and the importance of each depends a bit on the type of analytics you're interested in (e.g. product analytics vs. sports analytics).
Oct 12, 2022 5 tweets 1 min read
🧵 Underrated analytics skills: 1. Communication

Excellent and/or highly technical work can be lost to poor communication.

The ability to explain work to non-technical stakeholders increases impact significantly.
Apr 26, 2022 8 tweets 3 min read
New to sports analytics?

Last year I put together a series of blog posts and lists of resources to get you started.

Here they all are, in one place 👇 Sports Analytics 101 is series of blog posts that introduces the core concepts behind sports analytics in non-technical terms.
brendankent.com/sports-analyti…
Jun 23, 2021 8 tweets 3 min read
New to sports analytics?

Over the past year, I've put together a series of blog posts and lists of resources to get you started.

Here they all are, in one place 👇 Sports Analytics 101 is series of blog posts that introduces the core concepts behind sports analytics in non-technical terms.
brendankent.com/sports-analyti…
May 10, 2021 11 tweets 1 min read
Things about analytics every newcomer should know

(Feel free to add your own)

⬇⬇⬇ Raw data will almost always be messy and cleaning will be a big part of any project.

If you expect that, you'll be less frustrated with the amount of time you spend cleaning.
Mar 23, 2021 20 tweets 3 min read
Why is sports analytics valuable?

Because humans have mental & situational constraints.

A thread 👇 1/ Sports analytics has been able to provide value beyond traditional qualitative judgments because humans have limits that can be generally classified as:

- Mental constraints
- Situational constraints
Feb 11, 2021 14 tweets 3 min read
Descriptive vs. Predictive

What's the difference between a descriptive metric & a predictive metric? Why does it matter?

A thread 👇 1/ Generally speaking:

- Descriptive metrics are intended to describe what has happened in the past

- Predictive metrics are intended to provide insight into what might happen in the future
Jan 28, 2021 10 tweets 2 min read
New to sports analytics?

These are the programming languages and tools to learn & the order to prioritize learning them in.

A thread 👇 Priority 1️⃣: Get comfortable with Excel

Most high-level modeling is not done in Excel, however, it’s still important to know your way around a spreadsheet.

If you don’t have access to Excel, Google Sheets (which is free) will do the trick.
Jan 12, 2021 20 tweets 4 min read
Win Probability, Explained 📈

What is win probability? How does it work? Is it ever "correct"?

If you ever find yourself asking these questions, this thread is for you 🧵 1/ Win probability (abbreviated "WP") is the likelihood that a team will win a particular game, expressed as a percentage.

50% WP: If the game is played 1000 times, the team will win ~500

10% WP: If the game is played 1000 times, the team will win ~100
Oct 16, 2020 7 tweets 2 min read
Thread: There are a variety of online statistics, computer science, and data science courses that can be audited for free.

Here are a few I'd recommend to those interested in developing a technical skill set for sports analytics ⬇ "Intro to Statistics" from Stanford

Gotta build the foundation!

udacity.com/course/intro-t…
Oct 14, 2020 6 tweets 2 min read
#BigDataBowl 2021 is here 🏈 There aren't many better ways to get exposure to teams (that are hiring) than to perform well in this. Also, sports analytics is fun.