Daily Pulse — Options Flow + Dark Pool, 9 September 2026

Daily Pulse · Wednesday, 9 September 2026 · EOD options flow & dark-pool research · AZTMM HLDGS LLC

9 September 2026 — EOD read. The session that mattered was the one inside the options tape, not the one on the price chart. Net call premium ran to a market-wide +$189M by 9:55 in the morning and then bled away all day to finish at −$71M, while SPY gave up 0.46% and set its low at 760.94 before drifting sideways for four hours. Against that fade, one desk paid roughly $350M for 50,000 SPX calls expiring 31 December, in two lines printed in the same second at 2:14 pm. That is the day: a session tide that turned defensive, and a single position that did not.

SPY opened at 764.09 and spent the first fifteen minutes inside a 763.23 to 764.38 range. It reached the session high of 764.47 at 11:00 am, then broke straight down to the low of 760.94 at 11:25, and never traded above 763.41 again. The close was 762.40. Fifteen SPY dark-pool prints cleared inside the regular session and all fifteen carried the vendor’s contingent sale condition, each one 125,633 shares for $96.2M at 765.8875, a price above the day’s entire range because it was struck off a derivative leg rather than the tape. The figure draws that level dotted and marks it ×15. Of the five heaviest premium shelves, three sat inside the day’s range: 763 at $23.6M net put premium, 762 at $18.5M and 761 at $16.0M. Two sat off the price scale below it, 760 at $47.8M net put premium and 755 at $15.4M.

The tide opened at +$36.5M and climbed for twenty-five minutes to its peak of +$188.6M at 9:55 am. From there it fell for the rest of the day, crossing into negative territory just after noon and reaching its trough of −$79.0M at 3:00 pm. It finished the session at −$71.3M. Of the 78 five-minute readings, 41 sat above the zero line and 37 below it, and the series crossed zero three times. The shape is the whole story: everything positive happened in the first half hour.

8 September’s radar — OI confirm

Verdict describes the open-interest update only, never the contract.

Eight of the ten contracts I flagged on Tuesday came back as new open interest this morning. The cleanest were the ones with the smallest prior volume: CRCL March 2027 $150 calls turned 2,770 contracts into 2,737 of fresh open interest, and SKHY December $170 calls kept 980 of 1,047. TQQQ September $75 calls added 23,859 contracts on 25,263 of volume, which took that line from 1,506 open contracts to 25,365. The one outright miss was the SPY October $767 call line, where 26,516 contracts of volume produced only 1,960 of new open interest. Most of that volume was bid-side and most of it closed something rather than opening it. MU’s September 11 $1010 puts came in partial at 0.22, which is what a two-day contract usually looks like.

What happened

The first half hour was the only constructive part of the day. Net call premium built to +$188.6M by 9:55 am as SPY held its opening range, and then the bid simply stopped. By 11:25 the index ETF had printed 760.94, three and a half points below its high, and the remaining four and a half hours were a narrow drift between 761.32 and 763.41 that resolved into a 762.40 close. The move was 0.46% lower. QQQ lost 0.29% to 716.31 and NVDA fell 0.91% to 223.67, so the weakness was broad rather than concentrated in the large-cap semis.

Premium told the same story more plainly. SPY closed with $812.3M of put premium against $771.8M of call premium, and QQQ was more lopsided still at $702.7M of puts against $585.0M of calls. VIX rose from 15.00 to 16.00, a 6.7% move, on a session low of 15.57 and a high of 16.68. None of these are large numbers on their own. Taken together they describe a tape where hedges got paid for and upside did not.

The exception was loud. At 2:14 pm two SPX lines expiring 31 December printed in the same second: 25,000 contracts at the 8140 strike for $181.0M, and 25,003 contracts at 8160 for $168.8M. The 8140 line went up against open interest of 20,379 and the 8160 line against 70,017. That is roughly $350M paid for index upside almost 7% above where SPX was trading, with sixteen weeks to run. It is the largest single premium on today’s flagged-events screen by a factor of four.

Off exchange, the boards ran to $20.19B across 74 prints at or above $100M, in 45 distinct names. Both ends of the list were index products. The largest single print was 1,402,927 shares of IVV for $1.07B at $765.84, crossed at 4:05 pm and 0.03% below that fund’s 766.10 close, and it was one of six IVV prints inside a tenth of a percent of each other that together carried $4.03B. SPYM added $2.13B in three prints. Away from the index complex, AVGO led the single names with $1.07B in three prints at $364.38.

Why it matters

A tide that peaks in the first twenty-five minutes and then declines for six hours is a different animal from one that simply opens negative. It means the premium that came in early was met and absorbed, and that the marginal buyer through the middle of the day was buying protection rather than exposure. The 41-to-37 split between positive and negative readings understates that, because the positive readings were front-loaded and shrinking while the negative ones deepened into the close.

The strike ladder shows where that protection went. Every one of the six heaviest SPY shelves within a percent of spot carries more put premium than call premium, and the heaviest of all is 760 at $47.8M net. That is one dollar and forty cents below where SPY closed and four cents below the session low. The first shelf where calls lead is 745, at $8.7M net, which is 2.3% lower. There is no call-side shelf above the money at all inside the band.

Set against that, the $350M SPX position is not a hedge and is not short-dated. Someone is paying real money for a year-end move the rest of the tape is not positioned for. Both of those readings can be true at once, and they usually are. What resolves them is which one shows up in tomorrow’s open-interest update, and the December 31 lines are on the radar table below for exactly that reason.

What to watch into Thursday’s session

• 760.94 — today’s low, and the level to watch. A close below it would mark the first break of a session low since the 8 September reversal.
• 764.47 — today’s high, set at 11:00 am. Reclaiming it would undo the entire afternoon fade in one move.
• The 760 put shelf — $47.8M more put than call premium across all expiries, the heaviest single strike on the board and directly beneath the close.
• The tide’s first hour — today it peaked at 9:55 and never recovered. A tide that holds its opening build past 11:00 am would read differently.
• SPX 8140 and 8160, 31 December — whether roughly $350M of premium turns into new open interest, or was a roll of something already there.
• VIX 16.68 — today’s high. The index has closed under 17 every session this month.

Names on our radar

Every row below is frozen at publication and scored mechanically in the Accountability Ledger at +5 and +21 sessions — misses stay on the page. Options prints are checked against the next morning’s open-interest update in the following session’s post.

Positioning into Thursday

I go into Thursday reading this as a market that is neither breaking nor building. The regime line still reads Bull · early at MPI 65 as of Tuesday’s close, and nothing today argues for a change of state. What today did was drain the cushion. SPY sits one dollar and forty-six cents above its session low with the heaviest put shelf on the board sitting directly underneath, the tide finished negative for the second straight session, and the leveraged-long, inverse and volatility ETFs contributed nothing to either screen. That last reading is worth stating plainly rather than treating as a blank: no one reached for leverage in either direction today, which reads as positioning being trimmed rather than reversed. The single loud exception is a year-end call position four times larger than anything else on the flagged screen, and until the open-interest update lands tomorrow it is a claim rather than a fact.

The boards — data appendix

Today’s $20.19B is the smallest board since 4 September and 29.0% below Tuesday’s $28.43B, on 74 prints against 93. Distinct names fell to 45 from 52, and the ETF share of notional dropped 15.0 points to 53.6% from 68.6%. Fewer prints, fewer names and a smaller ETF share together are what a session looks like when the heavy index flow that dominated Tuesday is not repeated. Across the five sessions on record the board has ranged from $18.09B to $28.43B, so today sits in the lower half of a narrow band.

The heaviest strike on the board is 760, where put premium leads call premium by $47.8M across all expiries. The nearest shelf on the other side is 745, where calls lead by $8.7M. Everything between those two levels leans to the put side, including all four strikes within a dollar of the close.

Universe. Every board below is computed on the complete set of dark-pool prints at or above $100M premium for the 9 September session, obtained by paginating the vendor feed to exhaustion: 74 prints in 45 names totalling $20.19B, after dropping canceled rows and exact duplicates. No trailing baseline is available in this environment, so no comparison against a trailing average is offered.

Relative size

Relative size is each print’s share count divided by that name’s 30-day average daily volume.

The two screens behind this edition disagree by design and it is worth naming why. The boards above are a $100M premium screen and take every issue type; the relative-size table is a $10M screen limited to regular-hours common stock. LXP’s $34.6M print is the largest print of the day relative to its own 30-day average, at 0.90 times, and never reaches the $100M board. The $1.07B IVV print that tops the board is 0.22 times IVV’s 30-day average and never reaches the relative-size screen, because it is an ETF. The headline notional figure of $20.19B comes from the $100M board.

Cluster watch

Twelve names printed two or more times inside a tenth of a percent of one price today. The three above account for $7.08B of the $20.19B board between them, all of it in index products or a single mega-cap, and all of it in the closing hour.

Notable data points

1. SPY closed at 762.40, down 0.46% from 765.96, on a session range of 760.94 to 764.47.
2. Market-wide put/call volume ratio finished at 0.76, on 35.70M call contracts against 27.25M puts.
3. SPY put premium of $812.3M exceeded call premium of $771.8M; QQQ put premium of $702.7M exceeded calls of $585.0M.
4. VIX closed at 16.00, up 6.7%, on a range of 15.57 to 16.68.
5. The market tide crossed zero three times and spent 37 of 78 five-minute readings below the line.
6. Two SPX December 31 call lines totalling 50,003 contracts and roughly $350M printed within the same second at 2:14 pm.
7. The $100M+ dark-pool board totalled $20.19B across 74 prints in 45 names, of which ETFs were 53.6% of notional.
8. All fifteen regular-session SPY dark-pool prints carried a contingent or derivative-priced sale condition.

Method note

Flow and dark-pool data sourced from Unusual Whales. MPI score and regime classifier are our internal composite; daily synthesis is AI-assisted from those inputs. The flagged-events screen shown here is the 50 most recent flagged events at or above $1M premium for the session, and every statement about it is bounded by that depth. Volume-over-OI activity is unconfirmed as new positioning until the next morning’s open-interest update; confirmations are published in the following session’s post. Relative size is a print’s share count divided by that name’s 30-day average volume as reported with the print, computed by AZTMM; it is not any vendor’s relative-size or relative-strength score. Option premium by strike is aggregated across all expiries. The session map is drawn from 5-minute bars; the opening range is the first fifteen minutes; dark-pool prints marked contingent carry the vendor’s contingent or derivative-priced sale condition. Board baselines are trailing-session averages computed from our own archived pulls; derived indicators are computed by AZTMM from licensed market data.

This is research, not advice. Nothing here is a recommendation to buy, sell, or hold any security.

AZTMM HLDGS LLC is not a registered broker-dealer, investment adviser, or FINRA member. All content is retrospective research published for general circulation — not personalized advice, not trade signals. Options involve substantial risk, including losses that may exceed the initial investment. Full disclaimer.

New here? Start Here · Pulse Lab · Accountability Ledger · Trading Academy · Subscribe for daily updates