Polymarket Bot Tutorial · Chapter 15 of 32

Sports microstructure bots on Polymarket: in-game edge, scoreline-driven mispricing, the NBA tag (745) and Tennis tag (864), live data sources, and execution patterns for high-frequency sports markets.

By Harley Young, lead writer at Polymarkets.co.il. Last reviewed: May 2026.

What this chapter covers

This is chapter 15 of our 32-part series on building a Polymarket trading bot. We cover the topic in depth across the sections below. Body content for each section is being written and rolled out chapter-by-chapter; FAQ answers and references are already complete and reflect production experience from running our own trader.

  • Why sports markets are tradeable
  • Pre-game vs in-game (different bots)
  • Verified tag IDs (745 NBA, 864 Tennis)
  • Data sources: ESPN, official APIs, on-screen
  • Latency budget for in-game
  • The 0.99 / 0.01 trap
  • Code: subscribe to a games book and react

Why sports markets are tradeable

This section is in active development. Want to be notified when it goes live? Contact us or watch the authors page.

Pre-game vs in-game (different bots)

This section is in active development. Want to be notified when it goes live? Contact us or watch the authors page.

Verified tag IDs (745 NBA, 864 Tennis)

This section is in active development. Want to be notified when it goes live? Contact us or watch the authors page.

Data sources: ESPN, official APIs, on-screen

This section is in active development. Want to be notified when it goes live? Contact us or watch the authors page.

Latency budget for in-game

This section is in active development. Want to be notified when it goes live? Contact us or watch the authors page.

The 0.99 / 0.01 trap

This section is in active development. Want to be notified when it goes live? Contact us or watch the authors page.

Code: subscribe to a games book and react

This section is in active development. Want to be notified when it goes live? Contact us or watch the authors page.

Frequently asked questions

What sports tags are most active on Polymarket?
NBA (tag_id 745), Tennis (tag_id 864), and soccer (varies by competition) lead 24h volume during their seasons. NFL spikes weekly during the regular season and playoffs. We have verified the NBA and Tennis tag IDs in production - others should be checked via the gamma /tags endpoint before relying on them.
Can I bot in-game sports markets profitably?
Possibly - but it is hard. The edge is real (live scoreline often mispriced for 30-90 seconds) but other bots are watching too. Best results we have seen come from combining a fast live-score data source with simple rules ("opponent scored, market hasnt moved yet, buy"). Pure stat-arb without the data source loses to faster competitors.
Where do I get live sports data?
ESPN.com has unofficial JSON endpoints that return live scores - good enough for many strategies. Official APIs (NBA Stats API, NFL public endpoints) are more reliable but slower. Twitter accounts of beat reporters give text but require LLM parsing. None are HFT-grade; all are "fast enough" for retail.
What is the 0.99 / 0.01 trap?
When a sports market is at 99 cents YES (very likely won), there is almost no upside left and a 1-cent move can wipe a months expected gain. Many bots get caught buying at 0.99 chasing the last cent and getting smacked when an unexpected event drops the price to 0.85. Hard rule: do not buy above ~0.95 unless your expected value math is bulletproof.
How does Polymarket sports compare to traditional sportsbooks?
No house edge on the spread (vs ~5-10% vig at FanDuel/DraftKings), but liquidity is thinner and spreads can be wider. Polymarket excels for events that traditional books underprice - international tournaments, esports, niche markets. For mainstream NFL/NBA, traditional books are more liquid but cost more in vig.
Can my bot trade across multiple sports markets concurrently?
Yes - and you should. Sports microstructure works best as a portfolio of 5-20 simultaneous games. Per-game position cap (e.g. 50 USD), portfolio cap (e.g. 500 USD), and uncorrelated exposure across games. Concentrating on one game maxes out variance.