By Market Recap: July 2, 2026, The Day in Numbers
Holiday-eve rotation: green breadth under a falling Nasdaq, tech last of the eleven sector funds, memory-rout day two, and the split that faked a crash.
Thursday, July 2, 2026, the last session before the Independence Day closure, was a rotation day wearing a selloff's headline. QQQ printed -1.71% while DIA rose 1.04%, and breadth was POSITIVE: 3398 liquid names rose against 2758, 54.6% of the tape green while the growth index fell. Eight of eleven sector funds closed higher; the selling sat in tech and the memory complex that broke the day before.
The scoreboard
The exact SQL behind every number
WITH prior AS (
SELECT ticker, argMax(close, window_start) AS prior_close
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker IN ('SPY', 'QQQ', 'DIA', 'IWM')
AND window_start >= '2026-07-01 13:30:00' AND window_start < '2026-07-01 20:00:00'
GROUP BY ticker
),
sess AS (
SELECT ticker,
argMin(open, window_start) AS day_open,
argMax(close, window_start) AS day_close,
max(high) AS day_high,
min(low) AS day_low,
round(toFloat64(sum(volume)) / 1e6, 1) AS shares_traded_m
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker IN ('SPY', 'QQQ', 'DIA', 'IWM')
AND window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00'
GROUP BY ticker
)
SELECT
s.ticker AS ticker,
round(toFloat64(p.prior_close), 2) AS prior_close,
round(toFloat64(s.day_open), 2) AS day_open,
round(toFloat64(s.day_close), 2) AS day_close,
round((toFloat64(s.day_close) / toFloat64(p.prior_close) - 1) * 100, 2) AS pct_change,
round(toFloat64(s.day_high), 2) AS day_high,
round(toFloat64(s.day_low), 2) AS day_low,
s.shares_traded_m AS shares_traded_m
FROM sess s
JOIN prior p ON s.ticker = p.ticker
ORDER BY s.tickerDIA at 1.04% against QQQ at -1.71% is the day in one line, industrials and growth nearly three points of daily return apart. SPY split the difference at -0.12%; IWM closed -0.59%.
Was the day unusual?
Two lenses in one panel: SPY's open-to-close move and QQQ's close-over-close move, each ranked against the trailing month by absolute size (rank 1 = biggest).
The exact SQL behind every number
WITH per_day AS (
SELECT ticker,
toDate(toTimeZone(window_start, 'America/New_York')) AS d,
(argMax(toFloat64(close), window_start) / argMin(toFloat64(open), window_start) - 1) * 100 AS oc_pct,
argMax(toFloat64(close), window_start) AS rth_close
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker IN ('SPY', 'QQQ')
AND window_start >= toDateTime('2026-06-02 00:00:00')
AND window_start < toDateTime('2026-07-03 00:00:00')
AND (toHour(window_start) * 60 + toMinute(window_start)) BETWEEN 810 AND 1199
GROUP BY ticker, d
),
with_prev AS (
SELECT ticker, d, oc_pct,
lagInFrame(rth_close) OVER (PARTITION BY ticker ORDER BY d) AS prev_close,
(rth_close / lagInFrame(rth_close) OVER (PARTITION BY ticker ORDER BY d) - 1) * 100 AS cc_pct
FROM per_day
)
SELECT
round(anyIf(oc_pct, ticker = 'SPY' AND d = toDate('2026-07-02')), 2) AS spy_open_to_close_pct,
arrayCount(x -> x > abs(anyIf(oc_pct, ticker = 'SPY' AND d = toDate('2026-07-02'))),
groupArrayIf(abs(oc_pct), ticker = 'SPY' AND d != toDate('2026-07-02'))) + 1 AS spy_abs_move_rank,
countIf(ticker = 'SPY') AS spy_sessions_compared,
round(anyIf(cc_pct, ticker = 'QQQ' AND d = toDate('2026-07-02')), 2) AS qqq_close_over_close_pct,
arrayCount(x -> x > abs(anyIf(cc_pct, ticker = 'QQQ' AND d = toDate('2026-07-02'))),
groupArrayIf(abs(cc_pct), ticker = 'QQQ' AND d != toDate('2026-07-02') AND isFinite(cc_pct) AND prev_close > 0)) + 1 AS qqq_abs_move_rank,
countIf(ticker = 'QQQ' AND isFinite(cc_pct) AND prev_close > 0) AS qqq_sessions_compared,
toString(minIf(d, ticker = 'SPY')) AS first_session
FROM with_prevAt the index level, no. SPY's -0.35% open-to-close ranks 15 of 22 trailing sessions, the bottom half. QQQ was louder but still not extreme: its -1.71% close-over-close ranks 9 of 21 sessions with a defined prior close, back to 2026-06-02, a mid-pack bad day for the growth index. The single-name tape is where July 2 was loud.
Breadth: green tape, red growth index
The exact SQL behind every number
WITH per_ticker AS (
SELECT
ticker,
toFloat64(argMaxIf(close, window_start, window_start < '2026-07-02 00:00:00')) AS prior_close,
toFloat64(argMaxIf(close, window_start, window_start >= '2026-07-02 00:00:00')) AS day_close,
sumIf(toFloat64(close) * toFloat64(volume), window_start >= '2026-07-02 00:00:00') AS day_dollar_volume
FROM global_markets.delayed_stocks_minute_aggs
WHERE (window_start >= '2026-07-01 13:30:00' AND window_start < '2026-07-01 20:00:00')
OR (window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00')
GROUP BY ticker
)
SELECT
countIf(day_close > prior_close AND day_dollar_volume >= 1000000) AS advancers,
countIf(day_close < prior_close AND day_dollar_volume >= 1000000) AS decliners,
countIf(day_close = prior_close AND day_dollar_volume >= 1000000) AS unchanged,
countIf(day_dollar_volume >= 1000000) AS liquid_tickers,
count() AS tickers_traded_both_sessions,
count() - countIf(day_dollar_volume >= 1000000) AS dropped_by_liquidity_filter,
round(100.0 * countIf(day_close > prior_close AND day_dollar_volume >= 1000000)
/ countIf(day_dollar_volume >= 1000000), 1) AS advancer_pct,
reverse(arrayStringConcat(extractAll(reverse(toString(countIf(day_close > prior_close AND day_dollar_volume >= 1000000))), '[0-9]{1,3}'), ',')) AS advancers_fmt,
reverse(arrayStringConcat(extractAll(reverse(toString(countIf(day_close < prior_close AND day_dollar_volume >= 1000000))), '[0-9]{1,3}'), ',')) AS decliners_fmt,
reverse(arrayStringConcat(extractAll(reverse(toString(count() - countIf(day_dollar_volume >= 1000000))), '[0-9]{1,3}'), ',')) AS dropped_by_liquidity_filter_fmt,
reverse(arrayStringConcat(extractAll(reverse(toString(count())), '[0-9]{1,3}'), ',')) AS tickers_traded_both_sessions_fmt
FROM per_ticker
WHERE prior_close > 0 AND day_close > 03,398 advancers, 2,758 decliners, 63 unchanged, 54.6% of the liquid tape rose while QQQ fell. Cap-weighted indexes and equal-count breadth answered differently; days like this are why both get a panel. The filter drops 5,317 of 11,536 dual-session tickers under $1 million traded.
Sector by sector: where the green tape sat
Breadth counts names; it does not say which kind. The eleven SPDR sector funds cut the session by industry, and the gap between best and worst is the day's dispersion in one number.
The exact SQL behind every number
WITH per_etf AS (
SELECT
ticker,
toFloat64(argMaxIf(close, window_start, window_start < '2026-07-02 00:00:00')) AS prior_close,
toFloat64(argMaxIf(close, window_start, window_start >= '2026-07-02 00:00:00')) AS day_close,
round(sumIf(toFloat64(close) * toFloat64(volume), window_start >= '2026-07-02 00:00:00') / 1e9, 2) AS day_dollar_bn
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker IN ('XLB', 'XLC', 'XLE', 'XLF', 'XLI', 'XLK', 'XLP', 'XLRE', 'XLU', 'XLV', 'XLY')
AND ((window_start >= '2026-07-01 13:30:00' AND window_start < '2026-07-01 20:00:00')
OR (window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00'))
GROUP BY ticker
)
SELECT
ticker,
round(prior_close, 2) AS prior_close,
round(day_close, 2) AS day_close,
round((day_close / prior_close - 1) * 100, 2) AS pct_chg,
round((day_close / prior_close - 1) * 100 - min((day_close / prior_close - 1) * 100) OVER (), 2) AS pts_above_worst_sector,
day_dollar_bn
FROM per_etf
ORDER BY tickerHealth care (XLV) topped the board at 2.63%, then utilities 2.23%, staples 2%, materials 1.94%. Technology (XLK) finished last at -2.71%, the only sector down more than a point; consumer discretionary (-0.81%) and communication services (-0.13%) were the other red funds, the remaining eight green. Best-to-worst dispersion: 5.34 percentage points. The DIA-up/QQQ-down split ran across the market, not just four megacaps, which is why a green advance-decline line under a falling Nasdaq is no contradiction.
The day's highlight: memory rout, day two
The complex that broke Wednesday fell harder Thursday, magnitude and co-movement, not cause.
The exact SQL behind every number
WITH per_name AS (
SELECT
ticker,
toFloat64(argMaxIf(close, window_start, window_start < '2026-07-02 00:00:00')) AS prior_close,
toFloat64(argMaxIf(close, window_start, window_start >= '2026-07-02 00:00:00')) AS day_close,
maxIf(toFloat64(high), window_start >= '2026-07-02 00:00:00') AS day_high,
minIf(toFloat64(low), window_start >= '2026-07-02 00:00:00') AS day_low,
argMinIf(window_start, toFloat64(low), window_start >= '2026-07-02 00:00:00') AS low_bar,
argMaxIf(window_start, toFloat64(high), window_start >= '2026-07-02 00:00:00') AS high_bar,
round(sumIf(toFloat64(close) * toFloat64(volume), window_start >= '2026-07-02 00:00:00') / 1e9, 2) AS day_dollar_bn
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker IN ('MU', 'SNDK', 'STX', 'WDC')
AND ((window_start >= '2026-07-01 13:30:00' AND window_start < '2026-07-01 20:00:00')
OR (window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00'))
GROUP BY ticker
)
SELECT
ticker,
round(prior_close, 2) AS prior_close,
round(day_close, 2) AS day_close,
round((day_close / prior_close - 1) * 100, 2) AS pct_chg,
round(day_high, 2) AS day_high,
formatDateTime(toTimeZone(high_bar, 'America/New_York'), '%H:%i') AS day_high_et,
round(day_low, 2) AS day_low,
formatDateTime(toTimeZone(low_bar, 'America/New_York'), '%H:%i') AS day_low_et,
round((day_high / day_low - 1) * 100, 2) AS range_pct,
day_dollar_bn
FROM per_name
ORDER BY tickerSanDisk printed -14.32%, Seagate -10.38%, Western Digital -9.92%, MU -5.57%, on $51.4 billion of MU turnover, roughly one-and-a-half times SPY's. MU and SanDisk printed their lows late (15:26, 15:26 ET), Seagate and Western Digital earlier (14:23, 13:59). Context: Wednesday and the MU deep-dive.
The rotation's other half, in the megacaps:
The exact SQL behind every number
WITH per_name AS (
SELECT
ticker,
toFloat64(argMaxIf(close, window_start, window_start < '2026-07-02 00:00:00')) AS prior_close,
toFloat64(argMaxIf(close, window_start, window_start >= '2026-07-02 00:00:00')) AS day_close,
maxIf(toFloat64(high), window_start >= '2026-07-02 00:00:00') AS day_high,
minIf(toFloat64(low), window_start >= '2026-07-02 00:00:00') AS day_low,
argMinIf(window_start, toFloat64(low), window_start >= '2026-07-02 00:00:00') AS low_bar,
argMaxIf(window_start, toFloat64(high), window_start >= '2026-07-02 00:00:00') AS high_bar,
round(sumIf(toFloat64(close) * toFloat64(volume), window_start >= '2026-07-02 00:00:00') / 1e9, 2) AS day_dollar_bn
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker IN ('AAPL', 'MSFT', 'NVDA', 'TSLA')
AND ((window_start >= '2026-07-01 13:30:00' AND window_start < '2026-07-01 20:00:00')
OR (window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00'))
GROUP BY ticker
)
SELECT
ticker,
round(prior_close, 2) AS prior_close,
round(day_close, 2) AS day_close,
round((day_close / prior_close - 1) * 100, 2) AS pct_chg,
round(day_high, 2) AS day_high,
formatDateTime(toTimeZone(high_bar, 'America/New_York'), '%H:%i') AS day_high_et,
round(day_low, 2) AS day_low,
formatDateTime(toTimeZone(low_bar, 'America/New_York'), '%H:%i') AS day_low_et,
round((day_high / day_low - 1) * 100, 2) AS range_pct,
day_dollar_bn
FROM per_name
ORDER BY tickerAAPL rose 4.75% in a one-way line, low at 09:30 ET, high at 15:57, three minutes before the close, while TSLA ran the mirror image at -7.65%. MSFT added 1.41%; NVDA closed -1.55%. Same index, opposite days.
Where the money traded
The exact SQL behind every number
SELECT ticker, leaderboard, dollar_volume_bn, if(dollar_volume_bn < 1, dollar_volume_m, NULL) AS dollar_value_m, shares_m,
round(100 * if(leaderboard = 'by dollars traded', dollar_volume_bn, shares_m)
/ max(if(leaderboard = 'by dollars traded', dollar_volume_bn, shares_m)) OVER (PARTITION BY leaderboard), 1) AS pct_of_board_leader
FROM (
SELECT
'by dollars traded' AS leaderboard,
ticker,
round(sum(toFloat64(close) * toFloat64(volume)) / 1e9, 2) AS dollar_volume_bn,
round(sum(toFloat64(close) * toFloat64(volume)) / 1e6, 0) AS dollar_volume_m,
round(sum(toFloat64(volume)) / 1e6, 1) AS shares_m
FROM global_markets.delayed_stocks_minute_aggs
WHERE window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00'
AND ticker NOT IN ('SPCX')
GROUP BY ticker
ORDER BY dollar_volume_bn DESC
LIMIT 6
UNION ALL
SELECT
'by shares traded' AS leaderboard,
ticker,
round(sum(toFloat64(close) * toFloat64(volume)) / 1e9, 2) AS dollar_volume_bn,
round(sum(toFloat64(close) * toFloat64(volume)) / 1e6, 0) AS dollar_volume_m,
round(sum(toFloat64(volume)) / 1e6, 1) AS shares_m
FROM global_markets.delayed_stocks_minute_aggs
WHERE window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00'
AND ticker NOT IN ('SPCX')
GROUP BY ticker
ORDER BY shares_m DESC
LIMIT 4
)
ORDER BY leaderboard ASC, if(leaderboard = 'by dollars traded', dollar_volume_bn, shares_m) DESCMU's $51.4 billion led the tape for a fourth straight session (Monday, Tuesday, Wednesday) against SPY's $32.7 billion, and SanDisk's $26.57 billion put a second memory name in the top four. SOXS, the 3x-inverse semiconductor ETF, topped the share board at 748.2 million: cheap shares dominate a share count, expensive ones a dollar count, and relative volume compares either against a name's own norm. Basis: July 2 regular hours, one reused-symbol listing excluded (receipts).
The exact SQL behind every number
SELECT
formatDateTime(toStartOfInterval(toTimeZone(window_start, 'America/New_York'), INTERVAL 30 MINUTE), '%H:%i') AS et_time,
round(sum(toFloat64(volume)) / 1e9, 2) AS shares_bn,
round(100 * sum(toFloat64(volume)) / max(sum(toFloat64(volume))) OVER (), 1) AS pct_of_biggest_bucket
FROM global_markets.delayed_stocks_minute_aggs
WHERE window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00'
GROUP BY et_time
ORDER BY et_time2.13 billion shares in the opening half hour, a 0.81 billion trough at 14:30, 2.3 billion into the holiday-eve close, the final bucket the day's biggest, as the closing auction pulls resting orders into one print.
The options tape: the shifted weekly lands
The exact SQL behind every number
WITH
(
SELECT (any(underlying_symbol), any(toFloat64(strike_price)), any(option_type),
any(toDateOrNull(concat('20', substring(ticker, length(ticker) - 14, 6)))),
sum(size), count(), round(avg(toFloat64(price)), 3))
FROM global_markets.options_trades
WHERE sip_timestamp >= '2026-07-02 00:00:00' AND sip_timestamp < '2026-07-03 00:00:00'
GROUP BY ticker
ORDER BY sum(size) DESC
LIMIT 1
) AS top_contract,
(
SELECT round(toFloat64(argMax(close, window_start)), 2)
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker = 'SPY' AND window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00'
) AS spy_regular_close
SELECT
round(count() / 1e6, 2) AS option_prints_m,
round(toFloat64(sum(size)) / 1e6, 2) AS contracts_m,
round(100.0 * sumIf(size, option_type = 'C') / sum(size), 1) AS call_pct_of_volume,
round(100.0 * sumIf(size, substring(ticker, length(ticker) - 14, 6) = '260702') / sum(size), 1) AS same_day_expiry_pct,
round(toFloat64(sumIf(size, substring(ticker, length(ticker) - 14, 6) = '260702')) / 1e6, 2) AS thu_jul2_expiry_contracts_m,
countIf(substring(ticker, length(ticker) - 14, 6) = '260703') AS fri_jul3_expiry_prints,
round(toFloat64(sumIf(size, substring(ticker, length(ticker) - 14, 6) = '260710')) / 1e6, 2) AS jul10_weekly_contracts_m,
round(toFloat64(sumIf(size, substring(ticker, length(ticker) - 14, 6) = '260717')) / 1e6, 2) AS jul17_monthly_contracts_m,
round(toFloat64(sumIf(size, underlying_symbol = 'SPY')) / 1e6, 2) AS spy_contracts_m,
round(toFloat64(sumIf(size, underlying_symbol = 'QQQ')) / 1e6, 2) AS qqq_contracts_m,
top_contract.1 AS top_contract_underlying,
top_contract.2 AS top_contract_strike,
top_contract.3 AS top_contract_type,
top_contract.4 AS top_contract_expiry,
top_contract.5 AS top_contract_volume,
reverse(arrayStringConcat(extractAll(reverse(toString(assumeNotNull(top_contract.5))), '[0-9]{1,3}'), ',')) AS top_contract_volume_fmt,
round(top_contract.7, 3) AS top_contract_avg_price,
round(top_contract.2 - spy_regular_close, 2) AS top_strike_minus_spy_close,
round(spy_regular_close - top_contract.2, 2) AS spy_close_minus_strike
FROM global_markets.options_trades
WHERE sip_timestamp >= '2026-07-02 00:00:00' AND sip_timestamp < '2026-07-03 00:00:00'Options traded 80.96 million contracts across 13.15 million prints, and 47.4% of that volume expired the same Thursday. No contract with a July 3 expiration code printed all day (0 prints), so Thursday carried the daily expiry and the week's shifted weekly at once. The busiest contract was the same-day SPY $740 put, 540,403 contracts at an average $0.499 premium, SPY closing 4.8 dollars above the strike: out of the money, and a put atop a board where Tuesday and Wednesday had calls. Calls still took 58.4% of volume; the next weekly drew 9.64 million contracts, the July monthly 8.82 million (expiry mechanics).
The quote tape: what it cost to trade
Prices say what happened; quotes say what it cost. The bid-ask spread is the toll on every round trip, in basis points of the mid-price (a basis point is a hundredth of a percent). We measure it every session, ordinary or not, which is what makes "spreads blew out" a falsifiable claim.
The exact SQL behind every number
SELECT
round(countIf(toDate(sip_timestamp) = toDate('2026-07-02')) / 1e6, 2) AS jul2_updates_m,
round(countIf(toDate(sip_timestamp) = toDate('2026-07-01')) / 1e6, 2) AS jul1_updates_m,
round((countIf(toDate(sip_timestamp) = toDate('2026-07-02')) / countIf(toDate(sip_timestamp) = toDate('2026-07-01')) - 1) * 100, 1) AS day_over_day_pct,
round(countIf(toDate(sip_timestamp) = toDate('2026-07-02') AND ticker = 'SPY') / 1e6, 2) AS jul2_spy_updates_m,
round(countIf(toDate(sip_timestamp) = toDate('2026-07-02') AND ticker = 'QQQ') / 1e6, 2) AS jul2_qqq_updates_m,
round(countIf(toDate(sip_timestamp) = toDate('2026-07-02') AND ticker = 'MU') / 1e6, 2) AS jul2_mu_updates_m
FROM global_markets.cache_stocks_quotes
WHERE sip_timestamp >= '2026-07-01 00:00:00' AND sip_timestamp < '2026-07-03 00:00:00'The national best bid and offer, the top of the consolidated book, was rewritten 597.22 million times on July 2 against 449.15 million on July 1: a 33% jump into the closure. QQQ took 7.56 million, above SPY's 5.42 million; MU 1.01 million.
The exact SQL behind every number
SELECT
ticker,
round(med_bps, 2) AS median_spread_bps,
round(med_dollars * 100, 1) AS median_spread_cents,
round(med_bps / min(med_bps) OVER (), 1) AS times_the_spy_spread,
round(quote_updates / 1e6, 2) AS rth_updates_m,
invalid_quotes_dropped
FROM (
SELECT
ticker,
quantileExactIf(0.5)((toFloat64(ask_price) - toFloat64(bid_price)) / ((toFloat64(ask_price) + toFloat64(bid_price)) / 2) * 10000,
toFloat64(bid_price) > 0 AND toFloat64(ask_price) > toFloat64(bid_price)) AS med_bps,
quantileExactIf(0.5)(toFloat64(ask_price) - toFloat64(bid_price),
toFloat64(bid_price) > 0 AND toFloat64(ask_price) > toFloat64(bid_price)) AS med_dollars,
count() AS quote_updates,
countIf(NOT (toFloat64(bid_price) > 0 AND toFloat64(ask_price) > toFloat64(bid_price))) AS invalid_quotes_dropped
FROM global_markets.cache_stocks_quotes
WHERE ticker IN ('SPY', 'QQQ', 'AAPL', 'TSLA', 'NVDA', 'MU', 'SNDK', 'WDC')
AND sip_timestamp >= '2026-07-02 13:30:00' AND sip_timestamp < '2026-07-02 20:00:00'
GROUP BY ticker
)
ORDER BY median_spread_bps ASCSPY quoted a median 0.27 basis points wide: about 2 cents on a $744.8 ETF. QQQ came in at 0.83 bps, NVDA 1.03. The names doing the falling were the expensive ones to trade: MU 5.52 bps, SanDisk 10.4, Western Digital 10.79, 40x SPY's spread. Crossing a basket of memory names cost multiples of crossing the index, before any price impact. Invalid quotes (one-sided, crossed) are counted per name, not hidden.
The exact SQL behind every number
WITH per_day AS (
SELECT toDate(sip_timestamp) AS d,
quantileExact(0.5)((toFloat64(ask_price) - toFloat64(bid_price)) / ((toFloat64(ask_price) + toFloat64(bid_price)) / 2) * 10000) AS med_bps
FROM global_markets.cache_stocks_quotes
WHERE ticker = 'SPY'
AND sip_timestamp >= '2026-06-02 00:00:00' AND sip_timestamp < '2026-07-03 00:00:00'
AND (toHour(sip_timestamp) * 60 + toMinute(sip_timestamp)) BETWEEN 810 AND 1199
AND toFloat64(bid_price) > 0 AND toFloat64(ask_price) > toFloat64(bid_price)
GROUP BY d
)
SELECT
round(anyIf(med_bps, d = toDate('2026-07-02')), 3) AS jul2_median_spread_bps,
round(quantileExact(0.5)(med_bps), 3) AS trailing_median_bps,
round(anyIf(med_bps, d = toDate('2026-07-02')) - quantileExact(0.5)(med_bps), 3) AS jul2_minus_trailing_bps,
arrayCount(x -> x > anyIf(med_bps, d = toDate('2026-07-02')), groupArrayIf(med_bps, d != toDate('2026-07-02'))) + 1 AS wider_rank,
count() AS sessions_compared,
round(max(med_bps), 3) AS widest_session_bps
FROM per_dayAn ordinary day for liquidity, and that is the finding: SPY's median spread of 0.27 bps sits 0 bps from the trailing month's median (0.27 bps), ranking 11 of 22 sessions by wideness, nowhere near the month's widest at 0.409 bps. A concentrated rout under a green tape did not stress the plumbing.
Rates: the July 2 print, landed
The exact SQL behind every number
SELECT
(SELECT count() FROM global_markets.treasury_yields WHERE date = '2026-07-02') AS jul2_rows,
(SELECT count() FROM global_markets.treasury_yields WHERE date = '2026-07-01') AS jul1_rowsThe treasury feed runs a day or two behind the tape: at first publication the July 2 close had zero rows on file, and this page said so instead of guessing. The print has since landed, 1 row for July 2, 1 for July 1, so the panel below carries the session's own curve.
The exact SQL behind every number
SELECT
t.1 AS curve_point,
round(t.2, 2) AS jul2_yield_pct,
round((t.2 - t.3) * 100) AS one_day_change_bp
FROM (
SELECT arrayJoin([
('1 month', toFloat64(d.yield_1_month), toFloat64(p.yield_1_month)),
('3 month', toFloat64(d.yield_3_month), toFloat64(p.yield_3_month)),
('1 year', toFloat64(d.yield_1_year), toFloat64(p.yield_1_year)),
('2 year', toFloat64(d.yield_2_year), toFloat64(p.yield_2_year)),
('5 year', toFloat64(d.yield_5_year), toFloat64(p.yield_5_year)),
('10 year', toFloat64(d.yield_10_year), toFloat64(p.yield_10_year)),
('30 year', toFloat64(d.yield_30_year), toFloat64(p.yield_30_year)),
('2s10s spread', toFloat64(d.yield_10_year - d.yield_2_year), toFloat64(p.yield_10_year - p.yield_2_year))
]) AS t
FROM (SELECT * FROM global_markets.treasury_yields WHERE date = '2026-07-02') AS d,
(SELECT * FROM global_markets.treasury_yields WHERE date = '2026-07-01') AS p
)At the close the 10-year sat at 4.49%, the 2s10s spread at 0.35 points.
The calendar behind the day
The exact SQL behind every number
WITH
(
SELECT (count(), uniqExact(publisher))
FROM global_markets.stocks_news
WHERE toDate(toTimeZone(published_utc, 'America/New_York')) = '2026-07-02'
) AS news,
(
SELECT (argMax(t, n), max(n))
FROM (
SELECT t, count() AS n
FROM (
SELECT arrayJoin(tickers) AS t
FROM global_markets.stocks_news
WHERE toDate(toTimeZone(published_utc, 'America/New_York')) = '2026-07-02'
)
WHERE t != 'SPCX'
GROUP BY t
)
) AS top_news,
(
SELECT (count(), countIf(split_to > split_from), countIf(split_to < split_from),
arrayStringConcat(groupArray(concat(ticker, ' ', toString(split_to), '-for-', toString(split_from))), '; '))
FROM global_markets.stocks_splits
WHERE execution_date = '2026-07-02' AND ticker NOT IN ('SPCX')
) AS splits
SELECT
(SELECT count() FROM global_markets.stocks_dividends WHERE ex_dividend_date = '2026-07-02') AS ex_dividend_records,
splits.1 AS splits_executed,
splits.2 AS forward_splits,
splits.3 AS reverse_splits,
splits.4 AS split_records,
(SELECT count() FROM global_markets.stocks_ipos WHERE listing_date = '2026-07-02') AS ipos_listed,
(SELECT uniqExact(accession_number) FROM global_markets.stocks_sec_edgar_index WHERE filing_date = '2026-07-02') AS sec_filings,
(SELECT uniqExactIf(accession_number, form_type = '4') FROM global_markets.stocks_sec_edgar_index WHERE filing_date = '2026-07-02') AS insider_form4_filings,
(SELECT uniqExactIf(accession_number, form_type = '8-K') FROM global_markets.stocks_sec_edgar_index WHERE filing_date = '2026-07-02') AS filings_8k,
(SELECT arrayStringConcat(groupArray(concat(ticker, ' — ', issuer_name)), '; ') FROM (
SELECT ticker, issuer_name FROM global_markets.stocks_ipos WHERE listing_date = '2026-07-02' ORDER BY ticker
)) AS ipo_names,
news.1 AS news_articles,
news.2 AS news_publishers,
top_news.1 AS most_covered_ticker,
top_news.2 AS most_covered_articles,
(SELECT any(split_from) FROM global_markets.stocks_splits WHERE ticker = 'CRWD' AND execution_date = '2026-07-02') AS crwd_split_from,
(SELECT any(split_to) FROM global_markets.stocks_splits WHERE ticker = 'CRWD' AND execution_date = '2026-07-02') AS crwd_split_to,
(SELECT round(toFloat64(argMax(close, window_start)), 2) FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker = 'CRWD' AND window_start >= '2026-07-01 13:30:00' AND window_start < '2026-07-01 20:00:00') AS crwd_prev_close,
(SELECT round(toFloat64(argMax(close, window_start)), 2) FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker = 'CRWD' AND window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00') AS crwd_day_close,
(SELECT round(sum(toFloat64(close) * toFloat64(volume)) / 1e6, 2) FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker = 'CRWD' AND window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00') AS crwd_dollar_m,
(SELECT count() FROM (
SELECT ticker FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker IN (SELECT ticker FROM global_markets.stocks_splits WHERE execution_date = '2026-07-02')
AND window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00'
GROUP BY ticker
)) AS split_names_with_bars322 dividend records went ex-dividend, 2 new listings arrived (MIACU — Meridian3 Industrials Acquisition Corp.; VIIU — Viking Acquisition Corp. II), and the SEC index logged 5203 filings into the holiday, 2109 Form 4s, 258 8-Ks. The news feed carried 201 articles from 3 publishers (NVDA most-covered, 18 of them).
The splits trap anyone reading raw closes. CRWD ran a 4-for-1 forward split, so the unadjusted tape shows $772.45 on Wednesday and $193.67 on Thursday, a fake decline that is really four new shares for each old one. It was not alone: 8 split records executed, 5 forward and 3 reverse (the split_records column lists them), and a reverse split fakes the opposite artifact, a raw price that leaps overnight. Only 2 of those names traded on our tape; CRWD's $1473.51 million of turnover is the one big enough to contaminate a screen, and every mover screen here excludes it.
On deck
Calendar fact, not forecast, what the tables held about the next sessions.
The exact SQL behind every number
SELECT
(SELECT count() FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker = 'SPY' AND window_start >= '2026-07-06 13:30:00' AND window_start < '2026-07-06 20:00:00') AS jul6_spy_regular_bars,
(SELECT count() FROM global_markets.stocks_dividends WHERE ex_dividend_date = '2026-07-06') AS exdiv_records_jul6,
(SELECT arrayStringConcat(groupArray(ticker), ', ') FROM global_markets.stocks_dividends
WHERE ex_dividend_date = '2026-07-06'
AND ticker IN ('AAPL', 'MSFT', 'JPM', 'JNJ', 'XOM', 'KO', 'PG', 'WMT', 'CVX', 'HD')) AS household_exdivs_jul6,
(SELECT count() FROM global_markets.stocks_splits WHERE execution_date = '2026-07-06') AS splits_jul6,
(SELECT toString(min(date)) FROM global_markets.stocks_market_holidays WHERE date > '2026-07-02') AS next_scheduled_closure,
(SELECT any(name) FROM global_markets.stocks_market_holidays
WHERE date = (SELECT min(date) FROM global_markets.stocks_market_holidays WHERE date > '2026-07-02')) AS next_closure_name,
(SELECT any(status) FROM global_markets.stocks_market_holidays
WHERE date = (SELECT min(date) FROM global_markets.stocks_market_holidays WHERE date > '2026-07-02')) AS next_closure_status,
(SELECT toString(max(settlement_date)) FROM global_markets.stocks_short_interest
WHERE settlement_date <= '2026-07-02') AS latest_si_settlement,
(SELECT dateDiff('day', max(settlement_date), toDate('2026-07-02')) FROM global_markets.stocks_short_interest
WHERE settlement_date <= '2026-07-02') AS si_settlement_age_daysMonday July 6 reopened with a full 390-bar session, verified from its own bars. It carried 119 ex-dividend records, one household name among the ten we probe (JPM), and 15 split executions. Next scheduled closure: Labor Day, 2026-09-07 (closed). Short interest was old news, as always, newest settlement on file 2026-06-30, 2 days back, publication trailing settlement by about two weeks (why).
The session, verified, and the Friday that wasn't
The exact SQL behind every number
SELECT
formatDateTime(min(toTimeZone(window_start, 'America/New_York')), '%H:%i') AS first_spy_bar_et,
formatDateTime(max(toTimeZone(window_start, 'America/New_York')), '%H:%i') AS last_spy_bar_et,
count() AS spy_minute_bars,
countIf(window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00') AS regular_session_bars,
uniqExactIf(toDate(toTimeZone(window_start, 'America/New_York')), window_start >= '2026-07-02 13:30:00' AND window_start < '2026-07-02 20:00:00') AS day_sessions,
(SELECT count() FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker = 'SPY' AND window_start >= '2026-07-03 00:00:00' AND window_start < '2026-07-04 00:00:00') AS jul3_spy_bars
FROM global_markets.delayed_stocks_minute_aggs
WHERE ticker = 'SPY' AND window_start >= '2026-07-02 00:00:00' AND window_start < '2026-07-03 00:00:00'July 2 was a full session, not an early close: SPY's bars run 04:00 to 19:59 New York time (pre-market and after-hours bars included), with exactly 390 regular-window bars. Friday July 3 printed 0 SPY bars, a full closure for Independence Day, July 4 falling on a Saturday. The four-session week: the week recap.
FAQ
Why did the Dow rise while the Nasdaq fell on July 2, 2026?
The two indexes hold different companies: DIA closed 1.04%, QQQ -1.71%. The sector board carries the same split, health care, utilities and staples on top, technology last at -2.71%, a 5.34-point gap best to worst.
Was the stock market open the Friday of Independence Day week?
No. Independence Day fell on a Saturday, and the exchanges observed it with a full Friday closure the day after this session: our tape shows 0 SPY minute bars for it.
What does a forward stock split do to the share price?
It multiplies the share count by four and divides the price by four; the position's value is unchanged. CRWD, July 2: an unadjusted $772.45 close on Wednesday, $193.67 on Thursday, a fake decline on any screener that skips the split.
How wide were bid-ask spreads on July 2, 2026?
SPY's median quoted spread was 0.27 basis points of the mid-price in regular hours, 11 of 22 trailing sessions by wideness, an ordinary day. Single names ran wider: MU 5.52 bps, WDC 10.79.
Data notes
- Dollar volume is a per-minute proxy, close × volume, summed per bar.
- The July 2 treasury print landed after first publication, the original note disclosed its absence with a row-count receipt; this revision carries the print, receipt shown.
- CRWD's raw close change is a split artifact, excluded from mover screens.
Full data notes
- The sector board is the eleven SPDR Select Sector funds (XLB, XLC, XLE, XLF, XLI, XLK, XLP, XLRE, XLU, XLV, XLY), a fixed, disclosed basket, not a vendor sector field.
- Quote-tape counts bucket by the UTC date of the SIP timestamp; a summer session falls inside one UTC day.
- One reused-symbol June listing is excluded from the leaderboards (receipts); forensic tick work lives in the deep-dives.
Methodology
- The period is one trading session (1 session, verified from observed bars). Timestamps are stored in UTC and converted to New York time inside the queries; "close" means the last regular-session minute bar, and day changes compare July 2 with July 1. The July 3 closure was verified from bars, never assumed.
- Spreads are quoted (ask minus bid) in basis points of the mid-price, median across regular-hours NBBO updates, on a deterministic quantile. Decimals are cast to floats before ratio arithmetic; option expiries are re-parsed from the OCC ticker. Every panel is read once, at authoring time, through the gated read-only path.
Chart, table, and SQL are one object. Paste any panel into the Strasmore terminal and make it your own. Previous session: July 1. The week: the four-session holiday week.