Copy trading NBA prediction markets on Polymarket: the 2026–27 season guide
The 2026–27 NBA season opens on Tuesday, October 20, 2026 — Boston at Detroit, Philadelphia at New York, Oklahoma City at San Antonio — and runs 1,230 regular-season games to mid-April before the play-in and a 16-team bracket that ends in June. For a copy trader that is the third-largest regular-season sample of the four major North American leagues — behind MLB's 2,430 games and the NHL's 1,312 — and it comes with the two hazards that sample sizes hide: NBA favourites are priced far shorter than baseball's, so a high hit rate is routinely just the market, and the daily repetition that makes a record readable is also what a manufactured record is made of. This guide covers where the markets live, how the season is shaped, what to check on a basketball-heavy wallet, and the graded NBA board's base rates as they stood 32 days before tip-off.
Where do NBA markets live on Polymarket?
Every NBA event carries the nba tag. In the off-season the tagged events are futures and awards — on 2026-09-18 the two largest open ones we track were the 2027 Champion market, $21.9M of volume across 36 team markets, and the 2026–27 Rookie of the Year market at $7.9M across 81 — plus a handful of player-news questions. Once games begin, each game becomes its own event under the same tag: a moneyline (the market whose slug is the event's own), point spreads, game totals and whatever period or team lines Polymarket lists, roughly six to fifteen game events a night.
CopyGrade's NBA board is built on that tag, the way the NFL board is built on nfl: a wallet qualifies with at least five trades on NBA-tagged markets and is ranked on its realized profit or traded volume on those markets alone. The same tag powers the NBA chip on the Event Hub, where every game and futures event gets a dossier listing the graded wallets active on it.
How is the season shaped, and why does it matter for copying?
The calendar below is the format, not a schedule — tip-off times and the Cup's knockout dates come from the league's published 2026–27 schedule on NBA.com, and every date here — including the three opening-night games — was checked against it on 2026-09-18. Re-verify before you trade one.
| Phase | When | What it does to a copied wallet's record |
|---|---|---|
| Opening night | Tue Oct 20, 2026 | Early-season lines are the loosest of the year — rotations, minutes and new rosters are unpriced — and early-season records are the least informative |
| NBA Cup (in-season tournament) | Group play through November; knockout rounds in December | Group games count in the standings; the semi-finals and final do not. Markets may treat them differently — check which market a wallet actually bought |
| Christmas Day slate | Dec 25 | Traditionally the regular season's showcase slate; a favourite place for a young wallet's headline win |
| Trade deadline and All-Star break | Early–mid February | Rosters change; a wallet's pre-deadline record was made against different teams |
| Regular season ends | Mid-April | 82 games per team; the sample a season-long wallet is judged on |
| Play-in and playoffs | Late April to June | A best-of-seven bracket — short series, sharp prices, and the point where a season record stops and a bracket record begins |
Two structural facts matter more than any date. First, the NBA plays almost every day for six months — like baseball and unlike football, so a wallet can accumulate a genuinely large moneyline record in a single season. Second, unlike baseball, NBA games are often lopsided: a strong home side is routinely priced at 70–85¢ before tip. A wallet that buys those favourites will post a hit rate in the seventies without beating the market by a single game, which is why the one number to read on any NBA record is the hits its prices predicted, not the hit rate — the recent-form guide walks the arithmetic.
What should you check on an NBA-heavy wallet?
The general vetting checklist applies in full. Basketball adds five emphases:
- Run the farming check first, and expect it to matter less here than elsewhere. On 2026-09-18 the NBA board had the second-lowest farming-flag share of our 26 category boards (40.8%; only Formula 1 was lower). That is still above the 35.3% site-wide rate — every one of the 26 boards is, because a board requires five trades in one category and the site-wide figure counts every graded wallet — so read the flag rate as lower than other boards, never as clean. A farming-risk flag is a veto whatever the board's base rate.
- Price every hit. NBA moneylines make favourite-buying look like skill. A record of 30–10 built at an average price of 74¢ was expected to go about 29.6–10.4; it beat the market by nothing. Where a wallet's recent form shows its expected hits, or its trade ledger shows the prices it paid, read the record against that line.
- Separate the season's games from the futures. Champion, conference and award markets run for nine months and pay out once. A wallet whose NBA profit is one futures hit from last spring has a different record from one that ground out 300 moneylines — the verdict's category-edge breakdown shows where the profit sits. One longshot resolving is variance wearing a costume.
- Check when it trades. Late scratches, rest days and back-to-backs move NBA lines sharply in the last hour before tip and continuously in play. A wallet whose fills cluster after tip-off has a live-trading record — the kind a bot copying it seconds later cannot reproduce, because the price it got is gone. Pre-tip positioning is the only part of a record a copier inherits at a comparable price.
- Demand tenure. A season is long enough to build a record, so there is no reason to copy one that is three weeks old. Of the 22 positive-verdict wallets on the board on 2026-09-18, two had been under observation for 60 days. A score is only as old as its wallet.
How do you use the board and the tools during the season?
- The NBA board ranks graded wallets by NBA-attributed profit or volume, with each wallet's score, verdict and farming level beside it. Sort by volume first: it surfaces the wallets with sample, which is what a season is for.
- Scout filters the whole graded set by verdict, farming level, real edge, drawdown, minimum trades and volume, and by sport through its Topic filter — the shortlisting step before any single verdict is opened.
- Event dossiers on the Event Hub show, for one game or one futures market, which graded wallets are in it, on which side, at what price — the pre-tip positioning that a board cannot show.
- Board watch (Pro) fires when a tracked wallet you do not already follow opens a buy above your threshold on NBA markets — checked hourly, so a pre-tip position reaches you within the hour. Off until you switch it on, like every notification here.
- The Graded Slate and the per-wallet recent form card currently cover the NFL, college football and MLB — the leagues with a slug-matched moneyline per game. Their rules are published, and any extension to a new league would follow them.
Nothing on any of those surfaces is a pick. CopyGrade never executes a trade; every surface describes what a set of graded wallets did.
The base rates, dated
All wallet figures are a point-in-time pull from CopyGrade's live scored set via the read-only database on 2026-09-18 — the off-season, so these are wallets graded on last season's games and this season's futures. The board recomputes with every sync and will grow when games start; treat every figure as a dated level, never as a trend, and compare it only with figures carrying the same date.
| NBA board, 2026-09-18 | Wallets | Attributed volume | Realized profit | Profitable on NBA | Median real edge |
|---|---|---|---|---|---|
| Passes farming check | 1,741 | $93.2M | +$527,106 | 208 (11.9%) | −0.50% |
| Watch-level flag | 372 | $44.7M | +$2,176,695 | 87 (23.4%) | −1.15% |
| Severe farming flag | 829 | $68.9M | +$2,355,577 | 170 (20.5%) | −2.00% |
| Board | 2,942 | $206.8M | +$5,059,377 | 465 (15.8%) | −0.70% |
- Flagged share: 40.8% (1,201 of 2,942; 28.2% severe) — second-lowest of the 26 boards on that date, against 35.3% site-wide across 10,553 graded wallets.
- Negative real edge: 86.3% of the board — third-highest of the 26, tied with the NFL and behind only Formula 1 and golf. The clean cohort is underwater more often (89.1%) than the severe cohort (81.7%) and by far less (median −0.50% against −2.00%). Quote either half alone and the finding inverts; the NFL board showed the same shape.
- The clean cohort is net positive on this board — the 59% of wallets that pass the farming check traded 45% of the volume and kept about 10% of the profit. On the graded MLB board the same cohort was net negative on the same date; sport by sport, the clean cohort's take is a level to check, not a rule.
- Positive verdicts: 22, two of them observed for 60 days or more. 457 wallets had 50 or more NBA trades — the population where a season record can actually be read.
- Median CopyGrade Score on the board: 50.
"Real edge" is our post-fee, post-latency estimate of what a copier would have kept across a wallet's whole record, not its NBA record alone, per the published methodology; the NBA-attributed volume and profit are the board's own. Board attribution is per category, so a wallet active elsewhere appears on other boards too. A farming-risk flag is an algorithmic risk assessment, not an accusation of conduct, and a positive verdict is where diligence starts, not a recommendation — our own validation of whether the score predicts copier returns is still null.
CopyGrade is analysis-only — it never executes trades, holds funds, or custodies keys. A CopyGrade Score is a documented research opinion. Not financial advice.