The NBA board six days before opening night: 3,488 wallets, 43% flagged, 31 positive verdicts with $7,565 of NBA volume between them

Six days before the NBA's opening night, the graded NBA board holds 3,488 wallets, and 43.0% of them carry a farming-risk flag — the fourth-lowest share of the 26 category boards we grade, and still eight points above the 34.9% site-wide rate. 85.2% of the board has a negative real edge, the fourth-highest share of the 26. The wallets that pass the farming check are net positive — +$481,032 on $95.8M of NBA volume — but they keep less than a tenth of the board's profit, and the top ten wallets by NBA profit, eight of them flagged, hold 84% of it. Thirty-one wallets carry a positive copy verdict. Between them they have traded $7,565 of NBA markets.

And the preseason — 48 game events since September 28, $17.3M of volume — was traded almost entirely by one cohort: 97% of the fills graded wallets placed on those games came from wallets with a severe farming flag, and 78% of those fills landed after tip-off.

All wallet figures are a point-in-time pull from CopyGrade's live scored set via the read-only database, dated 2026-10-14; market figures are from our Gamma sweep the same day. The NBA board, like every tag board, recomputes with every sync, so every number here is a dated level — the kind the NFL board got before kickoff — and not a trend. "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.

What is the NBA board carrying before a game counts?

Polymarket's nba tag covers 86 events in our hub with $101.0M of lifetime volume: 26 still open ($34.7M traded, $1.22M of open interest, 612 markets) and 60 settled or closed ($66.2M). The largest NBA event we have ever tracked was not a game and not a futures market: "LeBron James Next Team" took $46.2M across 36 markets and is now closed. The largest open event is the 2027 Champion market — $25.4M traded against $714k currently held, roughly thirty-six dollars turned over for every dollar of open position — followed by the 2026–27 Rookie of the Year market at $7.97M across 81 candidates.

That ratio matters for the same reason it did on the NFL board in August. A futures market that trades thirty-six times its open interest is being flipped, not held, and a wallet's NBA "record" earned in it is a record of flipping futures, which says very little about how it will trade a Tuesday night moneyline.

The preseason gave the board its first games:

Forty-eight preseason game events, 968 markets, $17.3M. The biggest were Thunder–Hawks on October 12 ($830k) and Warriors–Lakers on October 13 ($802k) — game events already carrying 24 and 37 markets each, the moneyline, spreads and totals that the regular season will list six to fifteen times a night.

Who is on the board?

The flag rate first. 43.0% of the board is flagged (1,500 wallets; 29.2% severe), against 34.9% across all 13,039 graded wallets. Only Formula 1 (40.9%), the crypto price board (41.4%) and golf (42.8%) sit lower, and every one of the 26 boards sits above the site-wide rate, because a board requires five trades in one category and concentration on one market family is exactly the repetition the farming detectors read. Read the NBA figure as lower than most boards, never as clean — and note that among wallets active enough to judge at all (20 or more trades in the last 90 days), the site-wide flag rate is 57.3%, not 34.9%.

The last three columns carry the pattern the NFL board taught us and the NBA season guide repeated from the September read: they have to be read together. Clean wallets are underwater more often than severe ones (87.6% against 81.2%) and by far less (a median −0.50% against −2.20%). Clean wallets lose small and often; flagged wallets win often on paper and lose large in the estimate of what a copier keeps. Quote either half alone and the finding inverts. The board's negative-edge share, 85.2%, is the fourth-highest of the 26 — behind Formula 1 (89.5%), golf (86.6%) and the NFL (85.4%), and just ahead of the Premier League, MLB, the Champions League and the NHL. The boards whose wallets look best by headline measures keep turning out to be the ones with the least copyable edge.

The clean cohort is net positive — and where the profit actually sits

The NBA board is one of the places where the wallets that pass the farming check finish in the black: +$481,032 on $95.8M, a take of half a percent. Fifty-seven percent of the board's wallets, forty-seven percent of its volume — and under 10% of its realized profit. The watch and severe cohorts took $2.29M and $2.18M on less volume each.

Where the money is, by wallet rather than cohort:

The top fifty wallets by NBA profit hold more than the board's entire net result, which is another way of saying that everyone outside the top fifty is, collectively, underwater on NBA markets — including most of the clean cohort. Forty-six of those fifty carry a flag. A profit leaderboard built from this board would be a list of flagged wallets with one or two clean names near the bottom of the first page, which is the finding the leaderboard study keeps returning at every scale. We are not naming any of them: the ratios are the point, not the wallets.

Thirty-one positive verdicts, and what they have traded

Every one of the 31 also clears the canonical four-part copyable bar (clean check, positive real edge, 20 or more trades in 90 days), so for this board on this date "positive verdict" and "worth starting diligence on" are the same 31 wallets. Read the table against that. These are good whole-record wallets that have, between them, put $7,565 through NBA markets — a median of sixteen NBA trades each, four of them with fifty or more. They earned their verdicts elsewhere and have an NBA footprint. None of them is a basketball record.

On 2026-09-18 the season guide read 22 positive verdicts on a board of 2,942, two of them observed for 60 days. Today it is 31 on 3,488, fifteen past 60 days. Those are two dated levels on a board that grew by 546 wallets in between, not a trend — but the shape is the one the wallet-age post predicts: a positive verdict is a hypothesis a season has not yet tested, and the ones that survive to 60 days are worth more than the ones that are three days old, of which the board has several.

The preseason tape: who traded the 48 games

Preseason games are the board's first basketball in six months, so we read every fill graded wallets placed on the 48 nba-2026 game events since September 28, split by cohort and by whether the fill landed before tip-off (the game-start time Polymarket publishes for each market):

97.3% of the graded fills on preseason games came from the severe cohort, and 78% of those fills were placed in play. The clean cohort's entire preseason is 251 fills from 50 wallets — a hundred thousand dollars, 38% of it before tip. This is the college football pattern in a different sport: the largest cohort by volume is not forecasting games, it is trading them live, and live trading is the churn the detectors read. It is also the kind of record a copy bot cannot inherit — the price a live trader got is gone seconds later.

One more number from the same tape. Across 2,552 pre-tip moneyline buys by graded wallets, the median price paid was 47¢; 40.7% were between 40¢ and 60¢ and only 12.0% were at 70¢ or above. Preseason lines are not the lopsided 70–85¢ favourites the regular season will produce, which cuts two ways: a preseason hit rate is closer to honest than an in-season one, and it is also built on rotations and minutes that mean nothing in a week. Neither reading makes 48 exhibition games evidence.

What to do with this before Tuesday

• Vet on the whole record, not the board column. The 31 positive verdicts are whole-record verdicts; their NBA samples are tiny. The season guide sets out the five basketball-specific checks, the first of which is the farming check and the second of which is pricing every hit — a 30–10 record at 74¢ beat the market by nothing.
• Expect the board to change shape fast. Six to fifteen game events a night from October 20 will turn 48 preseason events into hundreds inside a month and hand every live-trading wallet a fresh record. A board figure quoted in November will not match one quoted today, which is why each carries its date.
• Use the surfaces for what they show, not for picks. The NBA board sorted by volume surfaces the wallets with sample; Scout filters the graded set by verdict, farming level and real edge; each game's Event Hub dossier shows which graded wallets are in it, on which side, at what price; and board watch (Pro, hourly) fires when a tracked wallet opens an NBA buy above your threshold. CopyGrade never executes a trade.

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. Not financial advice.