Working with Real Sports APIs
Read API docs like a developer, then pull your first real data from MLB, the NBA, StatsBomb, nflverse, the NHL and ESPN.
Every sport hides its data behind a slightly different door. This short course teaches the skill of opening them: first how to read API documentation at all, then one first-pull tutorial per major source — pybaseball for Statcast, nba_api, StatsBomb's free event data, nflverse, and the NHL's keyless public API.
Each pull tutorial also ships bundled sample data, so everything runs offline when an API is down, rate-limited, or being redesigned — which, you will learn, is often.
The finale puts the whole skillset to work on ESPN's undocumented scoreboard JSON: a complete Premier League season in one date-range call, plus the defensive habits — spotting silent response caps, refusing to overwrite good data with a bad refresh — that separate a script that ran once from a dataset you maintain.
What you'll be able to do
- Read an API's documentation and find the endpoint you actually need
- Pull real data from six major free sports data sources
- Handle the etiquette: rate limits, retries, caching, polite headers
- Know each source's quirks before they cost you an afternoon
- Keep a saved dataset safe from truncated or failed refreshes
The bundle contains each part's finished script, the shared helpers, and the sample data they read, with a README listing the run order - the whole course works offline.
The syllabus
Pace: Part 1 first, then any order - each source is its own door; part 7 closes the loop. A week of evenings.
0 of 7 parts marked complete on this device.
Every part of this course is marked complete on this device. The capstone below is the victory lap - and the quiz will tell you if anything slipped.
- How to Read API Documentation: A Sports Data Field GuideThe meta-skill: any API, read calmly, yields its endpoint.
- Pull Your First MLB Data with pybaseballStatcast is the deepest free feed in sports; open it first.
- Pull Your First NBA Data with nba_apiPowerful and picky - pull it politely and it gives you everything.
- Pull Your First Match Data with StatsBomb Open DataProfessional event data, free - learn its shape once.
- Pull Your First NFL Data with nfl_data_pyEvery NFL play for decades, one import away.
- Pull Your First NHL Data from the Public NHL APINo key, no signup - the friendliest first live pull in sports.
- Pull Live Soccer Results from ESPN's Public Scoreboard APIThe finale: a whole season from one call, and a dataset you guard.
Check yourself
Five questions, graded on the page, nothing recorded anywhere - pick an answer and the explanation appears.
1. Rate limits exist to…
A public API is shared infrastructure. Backing off and caching is both etiquette and self-interest - blocked scripts pull nothing.
2. The first thing to locate in an API's documentation is…
Everything else is decoration around 'which URL, with which parameters, returns which shape of data'.
3. Caching a pull to a local file matters because…
You'll re-run your analysis fifty times while writing it. The data changed zero of those times - pull once, read locally.
4. An HTTP 429 response means…
429 is the rate-limit status. The polite response is an exponential backoff, not an immediate retry loop.
5. These tutorials bundle sample data so that…
APIs break, move and throttle. The bundled CSVs keep every tutorial runnable - and honest, since the sample is real retrieved data.
Where next: Baseball Analytics with Statcast · Basketball Analytics with the NBA API · Soccer Analytics with StatsBomb & xG · NFL Analytics with nflverse.