Zeitgeist — Monday, November 2, 2026
A weekly look (new edition every Monday morning) at where the system's head is at: the mood, what's working and what isn't, the themes, and the week ahead. It's all built from TRAI's own data, then rapped over an original boom-bap beat. New to markets? Start with the plain-words version. Screening/research, not investment advice.
What's going on this week, in plain words
The stock market ended last week unsure of itself. TRAI is a computer program that studies about 156 big U.S. stocks. It gives the market a mood score from −100 (scared) to +100 (confident). On Thursday the score was 39.7. TRAI calls that "risk-on," meaning people felt bold. By Friday's close it fell to 4.5. That's right in the middle, which TRAI calls "neutral."
The VIX is a number that shows how nervous investors are. Think of it as the market's anxiety meter. It closed Friday at 19.42. TRAI calls anything from 15 to 22 "transitional." That means uneasy, but not panicked.
Friday's big story was chip stocks. Chips are the tiny brains inside phones, computers and AI servers. Intel fell 28.58% in one day. It said its sales over the next three months would be lower than people hoped. A fund that holds many chip companies fell 6.98%.
Meanwhile, money moved to calmer places. Over the past month, energy, health care, power companies and everyday goods (like food and soap) all beat the overall market by more than 10 points. Chip stocks fell about 15 points behind. It's like riders leaving a roller coaster for a park bench.
What has worked for TRAI? It tests its ideas on past data, like a practice season. Buying stocks that were already rising strongly made about 0.2% per trade. Buying stocks the day after a big drop made about 0.17%. Both are small, and not yet proven. Betting that falling stocks keep falling lost about 0.45% per trade, so TRAI turned that idea off.
This week there's no meeting of the Federal Reserve. That's the country's central bank, which sets the basic interest rate. Its next decision is Dec. 9. Instead, there are lots of reports. The biggest is the jobs report on Friday at 8:30 AM ET. Tuesday is also Election Day.
Many tech companies share their earnings this week. Earnings are a company's report card for the last three months. Apple and Palantir go Monday after the market closes. AMD goes Tuesday. Arm and Qualcomm go Wednesday. Traders expect big jumps: about 13% up or down for Palantir, and about 4.5% for Apple.
Mini glossary
- Stock / share
- A small piece of ownership in a company. Its price goes up and down as people buy and sell it.
- Regime score (mood score)
- TRAI's one-number summary of market mood, from −100 to +100. +25 or more is 'risk-on' (confident), −25 or less 'risk-off' (scared), in between 'neutral'.
- Risk-on / neutral / risk-off
- Shorthand for confident, in-between and scared markets.
- VIX
- The market's anxiety meter. It's based on what investors pay for protection on the S&P 500. Higher means more nervous.
- Calm / transitional / stress
- TRAI's three VIX zones: calm below 15, transitional from 15 to 22, stress above 22. In stress TRAI bets half as much.
- Semiconductors / chips (SMH)
- Companies that design or make computer chips, the brains inside phones, cars and AI data centers.
- ETF / fund
- A basket of many stocks you can buy in one go. SPY holds the S&P 500, QQQ the Nasdaq-100 and SMH chip makers.
- Basis points / points
- Here, 'points' means percentage points: 10 points better means a return of, say, 12% instead of 2% over the same time. (A basis point is a hundredth of one percentage point.)
- Expectancy (per trade)
- The average gain or loss per trade. Small numbers add up over many trades, in either direction.
- Earnings (report card)
- A company's quarterly results: sales, profit and outlook. Stocks often jump or drop on them.
- Short / shorting
- A bet that a price will fall. You profit if it drops and lose if it rises.
- Signal / bucket
- A stock that matched one of TRAI's screening recipes on a given day. Each recipe is a 'bucket'.
- Long
- The normal kind of bet: buying something because you think it will rise.
- Treasury yield (10-year)
- The interest rate the U.S. government pays to borrow. The 10-year rate affects mortgages and loan costs across the economy.
- S&P 500 (SPY)
- An index of 500 large U.S. companies, often used as shorthand for 'the market'. SPY is a fund that tracks it.
- Trend / choppy / ADX
- ADX measures how strongly prices are heading one way. Under 20 is 'choppy' (zig-zagging); 25 or more is 'trending'.
- Breadth / moving average
- Breadth asks how many stocks are joining in. A 20-day or 50-day average is a stock's average price over that many days; being above it is a sign of strength.
- Volume / relative volume
- How many shares traded. Relative volume compares today with normal. 3× means three times the usual trading.
- Defensive stocks
- Businesses people keep paying in good times and bad, like food, soap and electricity. Investors often move into them when nervous.
- Correlation cluster
- A 'flock' of stocks that move together. Owning several is like one big bet, so TRAI takes at most one trade per cluster.
- Correlation (ρ, 'rho')
- How closely two stocks move together, from 0 (unrelated) to 1 (lockstep). 0.5 or more is a strong link.
- Backtest / replay / walk-forward
- Testing rules on past data as if trading back then, day by day, without peeking ahead. It's a test, not a promise.
- Momentum
- The idea that a stock moving strongly, on heavy trading, keeps going for a while.
- After costs
- Results after subtracting estimated trading fees and the price you lose getting in and out.
- t-stat
- A check on whether a result is probably real or just luck. Beyond +2 or −2 is usually taken as a real pattern.
- Mean reversion / fade
- Betting that a big one-day move snaps partly back, like a stretched rubber band.
- Hit rate
- The share of trades that made money. 50% means half won.
- Residual z-score
- How unusual a stock's move was after removing what the whole market did. Near 0 is ordinary, 2 is rare, 3+ is very rare.
- Proposal (S-0xx)
- An idea from TRAI's research notes for changing the rules. It isn't a rule until it's tested and adopted.
- The Fed / FOMC
- The Federal Reserve, America's central bank. Its committee (the FOMC) sets short-term interest rates; markets watch every decision.
- Options / implied move / straddle
- Options are contracts that pay off if a stock moves. Their prices show how big a move traders expect, the 'implied move', in either direction.
- Before open / after close (BMO / AMC)
- When a company releases earnings: before the 9:30 AM ET open, or after the 4:00 PM ET close.
- Gap / gap fill
- A gap is when a stock opens far from yesterday's close, often after news. It 'fills' if the price comes back to that close the same day.
- Risk per trade
- The most a single trade is allowed to lose if its stop is hit, as a share of the account.
- Stop loss
- A pre-set exit: if the price moves against you by a set amount, you get out automatically and take a small loss.
- Opening range / no-entry window
- The first 30 minutes after the 9:30 AM ET open, when prices jump around most. TRAI won't start trades then.
- Equity (account)
- The total value of the trading account.
- ATR (average true range)
- A stock's typical daily price wiggle over the last 14 days. TRAI sets its exits in ATRs.
- Target
- A pre-set price where you take your profit and leave.
- Time stop
- If a trade hasn't worked within a set time (60 minutes for TRAI), close it anyway.
- Position
- A trade you currently hold.
- Flat / overnight hold
- Being 'flat' means holding nothing. TRAI closes everything by 3:55 PM so nothing is held overnight.
- Daily loss limit
- If the account is down 1% for the day, TRAI stops trading until tomorrow.
- Sector
- A group of companies in the same line of business, such as energy or banks.
- Term structure / contango
- Comparing short-term and longer-term VIX. Normally short-term worry is lower ('contango'). If it flips higher, people are scared right now.
- Universe
- The list of about 150 large, heavily traded U.S. stocks that TRAI watches.
- Rotation
- When money moves from one part of the market to another instead of leaving stocks altogether.
The overview
The mood
Neutral and choppy after a risk-on week. TRAI's regime score climbed from 13.5 on Oct 23 to 45.2 on Oct 28 and was still 39.7 (risk-on) on Oct 29, then fell to 4.5 (neutral) on Friday Oct 30. ADX is 15.3, so the trend reads 'choppy'. Breadth is split: 54.5% of the universe is above its 20-day average and 48.1% above its 50-day.
On Friday's final closes the VIX was 19.42, inside the 'transitional' band (15–22), so risk per trade stays at 0.25%. VIX9D/VIX was 0.991 and VIX/VIX3M 0.935. The 4:21 PM post-close engine run itself read 20.68 ('stress') because some of its VIX inputs were stale; the regime model's final-close read (used here) and the 17:00 research report both say transitional.
Friday was a semiconductor shock: SMH −6.98% and Intel −28.58% after a weak fourth-quarter forecast, while Regeneron rose 13.33% on its earnings. Over 20 sessions SMH trails SPY by 15.33 points; energy (+13.27), health care (+13.06), staples (+12.85) and utilities (+10.85) lead. The 10-year yield ended the week at 4.94%, down 0.06.
What's working / what's not
Long momentum leads the replay at +0.216% per trade after costs (n 378, t 1.37, hit 51.85%). The mean-reversion fade-long leg follows at +0.167% (n 225, t 1.26, hit 54.22%). Neither clears |t| ≥ 2, so both are promising, not proven.
Short momentum keeps losing: −0.448% per trade (n 415, t −3.62, hit 41.69%). It stays suspended from live trading and appears only as a watchlist.
Live tracking: 107 unique picks across 13 runs, 81 scored and 26 pending. Long momentum has 19 scored live picks, too few to judge.
The themes
Data hygiene: Friday's 4:21 PM post-close run was built partly on stale Yahoo bars (22 names plus the VIX complex). The 17:00 report re-ran the buckets on final closes, and the regime model's final-close read is transitional, not stress. Proposals S-108/S-113 add a freshness guard; S-102 covers the premarket VIX timing artifact.
Crowding: Friday's corrected fade-long list had ASML, ANET, ETN and AMAT, all in correlation cluster 1 (mean pairwise ρ 0.73). S-114 proposes sizing such groups as one bet; the live rule already allows at most one position per cluster.
Earnings proximity: S-111 proposes tagging mean-reversion picks that report within two sessions (PLTR reports tonight) and half-sizing them. S-100 notes that earnings-day momentum picks were the weak subset of the replay. That applies to REGN and AMZN.
Rotation: money has been hiding in energy, health care, staples and utilities for a month while chips trail. None of these proposals is live; they need testing first.
The outlook
No Fed decision this week (next FOMC Dec 8–9). Data: ISM manufacturing Mon 10:00; JOLTS Tue 10:00 (plus Election Day); trade 8:30 and ISM services 10:00 Wed, with the Treasury refunding statement; productivity Thu 8:30; the October jobs report Fri 8:30, with the UMich survey. October CPI follows Tue Nov 10.
Earnings with options-implied moves: PLTR ±13.14% and AAPL ±4.52% (Mon after close); SHOP ±13.53% (Tue before open); AMD ±9.13% and ANET ±10.35% (Tue after close); QCOM ±10.64%, ARM ±12.47% and APP ±17.61% (Wed after close); NET ±15.53% (Thu after close). Seven cluster-1 names report Nov 2–5 (ON, ETN, AMD, SMCI, ANET, ARM, MCHP), so a single print can move the whole chip block.
Gap risk: none of Palantir's last four earnings gaps filled (average 9.61%). Fading earnings gaps stays off the table (S-006). In energy, the tightest large cluster (ρ 0.707), COP, EOG and DVN all report Thursday.
The Zeitgeist — The Cold Ledger: lyrics + plain-words translation
Original lyrics in a '90s East Coast boom-bap style. Every figure comes from the run above. Left: the bars. Right: what they mean, in plain words (open “More detail” for the full numbers).
0:00 Intro
Everything in this song comes from TRAI's own numbers about last Friday's market. TRAI is a computer program that studies stocks. It doesn't tell anyone what to buy.
More detail
Everything in this rap comes from TRAI's data. The engine run is TRAI's Friday Oct 30 post-close run (4:21 PM ET), covering the Oct 30 session; a few of its inputs were stale, so headline index and VIX numbers use final closes from TRAI's macro and regime datasets. The mood numbers come from TRAI's daily regime model on Friday's final closes. The scorecard, correlation, sector and options datasets are as of Oct 30, and the earnings calendar is as of Nov 2. This week's economic dates come from the official Fed, BLS, BEA, ISM and Treasury calendars. TRAI screens stocks from public data. It doesn't recommend anything.
0:10 Verse 1 — The mood
TRAI gives the market a mood score from −100 (scared) to +100 (confident). On Thursday it was 39.7. By Friday it fell to 4.5, right in the middle. The VIX is the market's anxiety meter. It closed Friday at 19.42: uneasy, not panicked. About half of the stocks TRAI follows are above their recent average price. So there's no clear direction.
More detail
TRAI's regime score blends volatility, the VIX, the VIX term structure, trend and breadth into one number from −100 to +100. It rose from 13.5 on Oct 23 to a peak of 45.2 on Oct 28, was still 39.7 ('risk-on') on Thursday Oct 29, then fell to 4.5 ('neutral') on Friday Oct 30. The VIX, the market's 'fear gauge' priced from S&P 500 options, closed Friday at 19.42. That is TRAI's middle 'transitional' band (15–22), so risk per trade stays at the normal 0.25%. ADX measures trend strength; at 15.3 (under 20) SPY has no clear trend, so TRAI calls it 'choppy'. Breadth is split: 54.5% of the universe is above its 20-day average and 48.1% above its 50-day.
Chip stocks had a bad Friday. Chips are the tiny brains inside phones, computers and AI servers. A fund holding many chip companies fell 6.98%. Intel dropped 28.58% in one day after it said its next three months of sales would be weaker than hoped. Drug maker Regeneron jumped 13.33% after a strong report card.
More detail
Friday, Oct 30 (final closes): SMH, the chip-stock ETF, fell 6.98%. Intel (INTC) fell 28.58% after it forecast fourth-quarter revenue below what analysts expected and warned PC shipments would shrink (per the 09:00 research report). Regeneron (REGN) went the other way: +13.33% on its earnings day, with 4.51 times its normal volume. That made it the top long-momentum name.
Over the past month, chip stocks fell far behind the overall market. Money moved to calmer places: energy, health care, everyday goods like food and soap, and power companies. It's like riders leaving a roller coaster for a park bench. Interest rates barely moved.
More detail
Over the last 20 trading days, SMH trails SPY by 15.33 percentage points. The leaders are the defensive and real-asset sectors: energy (XLE) +13.27 points vs SPY, health care (XLV) +13.06, consumer staples (XLP) +12.85, utilities (XLU) +10.85. This is money rotating to shelter, not leaving stocks altogether. The 10-year Treasury yield was 4.94% on Oct 30, down 0.06 points (6 basis points) on the week (source: US Treasury yield curve; FRED was unreachable).
Some stocks move together like a school of fish. TRAI found 24 of them, mostly chip companies. Seven of these report earnings this week. If one gets hurt, the others often follow. A 'neutral' mood doesn't mean calm. It means the market hasn't picked a direction yet.
More detail
TRAI groups stocks whose daily returns move together over 60 days. Cluster 1 has 24 names with an average correlation of 0.576: 17 semiconductors, 3 hardware, 3 industrials and 1 software name (it includes NVDA, AMD, AVGO, ARM, ON, ETN, ANET and GEV). Seven of them report earnings Nov 2–5: ON, ETN, AMD, SMCI, ANET, ARM and MCHP. One surprise can move the whole group. A 'neutral' regime score means no edge either way, not calm.
0:49 Hook — The cold ledger
TRAI trades by eight fixed safety rules, set ahead of time so it never makes decisions in a panic.
Don't start any trade in the first half hour after the market opens (9:30–10:00 AM ET), when prices jump around the most.
Each trade can lose at most 0.25% of the account ($25 on $10,000), and half that when the anxiety meter (VIX) is above 22.
Every trade gets a stop loss (an automatic exit) about one normal day's price wiggle away, and a profit target at twice that distance, so wins are bigger than losses.
If a trade hasn't worked within 60 minutes, close it anyway.
Hold no more than 3 trades at once. Only one trade from each 'flock' of stocks that move together.
If the account is down 1% for the day, stop trading until tomorrow. Close everything by 3:55 PM, so nothing is held overnight.
Everything here comes from public data. It's research, not financial advice.
More detail on the rules
The hook lists TRAI's eight risk rules from the engine run. (1) No new entries 9:30–10:00 ET; let the opening range form first. (2) Risk 0.25% of equity per trade, halved to 0.125% if the VIX is above 22. (3) Stop loss at 1× ATR (average true range, a stock's typical daily move) from entry. (4) Profit target at 2× ATR. (5) Time stop: exit if the trade isn't working after 60 minutes. (6) At most 3 positions at once, and no more than 1 per correlated cluster. (7) Down 1% on the day: stop trading. (8) Flat by 3:55 PM ET, no overnight holds. The 'cold ledger' is the rulebook: it limits damage. It doesn't predict prices.
1:10 Verse 2 — What's working, what's not
TRAI tests its ideas on past data, like a practice season. Buying stocks that were already rising strongly made about 0.2% per trade on average. Buying stocks the day after a big drop made about 0.17%. Both are small wins, and the test isn't long enough to be sure they're real.
More detail
TRAI's walk-forward replay tests each screen on 126 past signal days (May 1 to Oct 29, outcomes through Oct 30), trading next day open to close after costs. Long momentum (buying strong, unusual movers) made +0.216% per trade (n 378, t 1.37, hit rate 51.85%). The fade-long leg of mean reversion (buying after a big down day) made +0.167% (n 225, t 1.26, hit rate 54.22%). A t-statistic under 2 means the result could still be luck, so both are promising, not proven.
Betting that falling stocks keep falling lost money: about 0.45% per trade, and it was right only 42% of the time. So TRAI turned that idea off. It still watches those stocks but doesn't act on them.
More detail
Short momentum (betting against weak stocks) lost 0.448% per trade after costs (n 415, hit rate 41.69%, t −3.62). A t-statistic that far below −2 means the losses are very unlikely to be luck. That's why the bucket is suspended: the engine still lists it as a watchlist, but it isn't traded. Friday's top name on that list was Roblox (RBLX), down 18.07%.
TRAI also grades its real, recent picks. 81 have been graded so far and 26 are still waiting. That's too few to judge yet.
More detail
Live tracking scores the engine's real picks after the fact. Across 13 recent runs it logged 107 unique picks: 81 are scored and 26 still wait for their outcome. Long momentum has only 19 scored live picks, too few to judge.
The strongest stock on TRAI's list jumped after its earnings report. But past tests say jumps on report day often fade the next day.
More detail
The long-momentum screen's top name in the run is Regeneron (REGN): +13.33% on Oct 30, a residual z-score of 8.53 (how unusual the move was after removing the market's move) and 4.51× normal volume. But the move came on its earnings day. Research proposal S-100 found earnings-reaction long-momentum picks made −0.42% the next day in the replay (n 48), versus +0.27% for the rest.
TRAI's research team suggested fixes. One: be careful with bounce bets on companies about to report earnings. Two: if four picks all move together, treat them as one bet, not four. These are ideas. They aren't in use until tests back them up.
More detail
Three research proposals from last week (knowledge/strategy.md), none live yet. S-111: tag mean-reversion picks that report earnings within two sessions (Friday's corrected list had PLTR, which reports tonight) and size them at half risk. S-114: when several same-side picks share a correlation cluster, size them like one position; ASML, ANET, ETN and AMAT were all on Friday's corrected fade list, all in cluster 1, with an average pairwise correlation of 0.73. Proposals only become rules after testing.
2:10 Verse 3 — The outlook
The Federal Reserve is the country's central bank. It sets the basic interest rate, and it doesn't meet this week. Its next decision is Dec. 9. Instead, the government releases a lot of reports. The big one is the jobs report on Friday at 8:30 AM ET. Tuesday is also Election Day.
More detail
There is no Federal Reserve rate decision this week: the FOMC met Oct 27–28 and its next meeting is Dec 8–9 (federalreserve.gov). The data calendar is busy instead. Monday: ISM manufacturing survey (10:00 AM ET) and Treasury borrowing estimates (afternoon). Tuesday: JOLTS job openings for September (10:00 AM ET, BLS) and the midterm Election Day. Wednesday: U.S. trade for September (8:30 AM ET, BEA), the Treasury's quarterly refunding statement, and the ISM services survey (10:00 AM ET). Thursday: productivity and labor costs (8:30 AM ET). Friday: the October jobs report (8:30 AM ET, BLS) and the University of Michigan consumer survey. October CPI comes next Tuesday, Nov 10.
Many tech companies share their quarterly report card ('earnings') this week. Apple and Palantir go Monday after the market closes. AMD goes Tuesday, and Arm and Qualcomm go Wednesday. Traders expect big jumps. They're pricing about a 13% move for Palantir and about 4.5% for Apple, in either direction.
More detail
Earnings (TRAI's calendar, /api/events as of Nov 2) with options-implied moves (the up-or-down move option prices expect after the report; /api/options as of Oct 30): Monday after the close, Palantir (PLTR) ±13.14% and Apple (AAPL) ±4.52%. Tuesday, Shopify (SHOP) before the open ±13.53%; AMD ±9.13% and Arista (ANET) ±10.35% after the close. Wednesday after the close, Qualcomm (QCOM) ±10.64%, Arm (ARM) ±12.47% and AppLovin (APP) ±17.61%. Thursday after the close, Cloudflare (NET) ±15.53%.
After Palantir's last four reports, the stock opened about 10% away from the night before. It never came back the same day. So TRAI's research says don't bet against a big jump like that.
More detail
A 'gap' is when a stock opens far from where it closed the night before, usually after news. In TRAI's history of Palantir's last four earnings reactions, the gap averaged 9.61% and none of the four filled (the price never returned to the prior close that day). Research proposal S-006 (TRAI big-gap subset: 2.3% fill rate) says not to bet against big earnings gaps the same day.
Oil and gas stocks move together more tightly than any other big group TRAI tracks. Three of them report on Thursday. TRAI's rule: only one bet per group like this. Each trade can risk a quarter of one percent of the account. That's $25 on $10,000. It gets cut in half if the anxiety meter goes above 22.
More detail
The energy cluster (8 names: MPC, PSX, XOM, CVX, DVN, OXY, COP, EOG) is the tightest of TRAI's large clusters (five or more names), with an average correlation of 0.707. ConocoPhillips (COP) reports Thursday before the open; EOG and Devon (DVN) Thursday after the close. Because they move together, the one-per-cluster rule matters. Risk per trade is 0.25% of the account; it halves to 0.125% only if the VIX is above 22 (it closed Friday at 19.42).
3:00 Outro
That's it for this week. This is research from public data. It's not advice about what to buy or sell.
More detail
End of the edition. Next edition: Monday, Nov 9. Everything here is screening research from public data, not investment advice.
Archive
- Monday, November 2, 2026
- Friday, October 23, 2026
How this is made
A script on the research box (scripts/zeitgeist/) pulls TRAI's public JSON (/api/latest, regime, scorecard, events, correlations, macro, options). It then checks every number in the lyrics and overview against that data, and refuses to render if anything has changed. The beat (drums, dusty piano, strings, bass, vinyl crackle) is synthesized from scratch in code at 88 BPM, with no samples. A neural text-to-speech voice performs the lyrics, with tickers spelled letter by letter. Each line is aligned with speech recognition and placed on the downbeat of its bar. Slides are drawn with matplotlib from the same data, and ffmpeg encodes the H.264/AAC MP4. The lyrics are original: they don't quote any existing song or imitate any real artist. Macro dates (FOMC, GDP/PCE) come from the day's research reports. The plain-words story, glossary and translations are generated by the same script from the same data on every run, with no hand-typed numbers. A scheduled run rebuilds the page every Monday morning and frames it as the week ahead. Screening/research, not investment advice.