On a Monday
morning, someone
bought the market.

On a Monday morning in April 2025, someone bought the market. Not in the ordinary sense. Between the opening bell and early afternoon on April 7th, the S&P 500 ETF traded $65.85 billion in dollar volume. Its baseline for that same window — the average across the five prior trading days — was $18.19 billion. The residual, the volume that cannot be explained by normal trading patterns, was $47.66 billion.

Two days later, the administration announced a 90-day pause on the sweeping global tariffs it had imposed the previous week. Markets surged. The S&P 500 posted one of its largest single-day gains in years. Anyone who had purchased broad market exposure on April 7th, when the market was still pricing in a sustained trade war, made an extraordinary return in 48 hours.

We do not know who was trading on April 7th. What we know is that they were trading at three and a half times the normal rate, in instruments that would directly benefit from a policy reversal that had not yet been announced, on a day when the only notable public event was a Truth Social post from the President of the United States telling his followers it was a great time to buy.

Market Sentinel was built to find exactly this kind of pattern. It does not identify perpetrators. It does not assert intent. It measures anomalies: moments when the market behaves in ways that normal trading activity cannot explain, clustered around policy announcements, reversals, and the social media posts of the one person in the world who knows what those announcements will be before they are made.

This is what we built, what we measured, and what we found.

How Market Sentinel Works

The core question this investigation asks is simple: did the market behave abnormally around specific policy events? Answering it requires a methodology precise enough to distinguish signal from noise, and transparent enough that any reader can challenge the findings.

Market Sentinel does not predict markets. It does not model intent. It performs a single operation, repeated across multiple events and multiple instruments: it measures how much dollar volume traded in a defined window around a policy event, compares that against the same window on the five prior trading days, and calculates the difference. That difference is the residual. A large positive residual means the market traded significantly more than normal in that window. It does not explain why. It only establishes that something happened, that it was measurable, and that it was not routine.

The data comes from two sources. Minute-level price and volume bars are pulled from Polygon.io, a public market data provider whose historical feeds are used by institutional and retail researchers alike. Options flow data — the record of call and put premium traded in a given window — comes from Unusual Whales, a market surveillance platform that aggregates and timestamps options activity. Both datasets are commercially available. Neither requires special access or proprietary feeds. Any researcher with a paid subscription to either service can replicate this analysis using the event list and methodology published alongside this article.

The events themselves are drawn from public record: archived Truth Social posts with verified timestamps, White House press schedules, and Federal Register publication times. No event in this analysis relies on an unverified or estimated timestamp. Where the precise time of a post or announcement is disputed in the public record, we have noted that dispute and used the most conservative available timestamp.

The baseline period is five prior trading days. This is a standard choice in academic event study literature, conservative enough to avoid contamination from related events but sufficient to establish a reliable average. On days where market data is unavailable due to holidays or early closes, those days are excluded from the baseline calculation and the remaining days are averaged. The methodology does not manufacture data to fill gaps.

A residual is flagged when it exceeds 2.5 times the baseline average. This threshold is not arbitrary. It represents a level of deviation that occurs in normal markets with a frequency of less than two percent — meaning that in any given window on any given day, the probability of seeing a 2.5x volume spike by chance alone is small. Across multiple instruments on the same day, the probability of simultaneous 2.5x spikes by chance approaches zero.

What We Found

Market Sentinel analyzed six events between February 1 and April 9, 2025, covering the major tariff announcements, reversals, and social media posts that defined the administration's opening trade policy posture. The full results are published in the accompanying CSV. Two dates require specific attention.

April 7, 2025: The Post

At some point on the morning of April 7th, the President of the United States posted on Truth Social. The message was brief: it was, he said, a great time to buy.

The market had been in freefall for five days. Liberation Day — the April 2nd announcement of sweeping global tariffs — had erased trillions in market value. There was no public indication that any policy reversal was imminent. The dominant narrative in financial media that morning was that the trade war was escalating, not retreating.

Market Sentinel measured the window from market open through early afternoon on April 7th and compared it against the same window on the five prior trading days.

Five things in that table deserve attention.

First, the lead instrument is not SPY. XLK, the technology sector ETF, posted the highest residual in the entire dataset at +317.4%. Not the broad market. The sector most directly exposed to the China tariff regime — semiconductors, software, hardware — traded at more than four times normal volume on April 7. That is a sector-specific bet, not a general market hedge.

Second, the bidirectional positioning. SQQQ is the 3× inverse Nasdaq ETF. It rises when the Nasdaq falls. It also traded at +191% above baseline that day. Simultaneously, TQQQ — the 3× leveraged Nasdaq — traded at +180% above baseline. Massive capital moved in both directions in the same window. This is either institutional hedging at extraordinary scale, or multiple actors with opposing views, or something more deliberate. What it is not is normal retail trading behavior during a market selloff.

Third, the individual stock concentration. AAPL, NVDA, MSFT, and AMZN all showed residuals between 76% and 163%. These are the largest companies in the world by market cap. Moving them to 1.8× to 2.6× normal volume requires institutional-scale capital. Retail investors do not generate these numbers.

Fourth, the outliers cut both ways. TSLA at +17.4% and NKE at +11.7% are essentially flat relative to baseline. Not everything was anomalous. The anomaly was concentrated in tech-exposed, tariff-sensitive, and leveraged instruments. That concentration is itself signal.

Fifth, DJT. Trump Media, the president's own company, traded below its prior baseline. Whoever generated the anomaly across the other 14 instruments deliberately avoided the most obviously connected vehicle. They moved through the most liquid, most anonymous instruments available.

April 9, 2025: The Pause

At 13:18 Eastern Time on April 9th, the administration announced a 90-day pause on the Liberation Day tariffs. Markets surged. SPY posted its largest single-day gain in years.

The April 9th numbers are large. They are partially explainable: a major policy announcement will always generate elevated volume as the market reprices. What is not explainable by the announcement alone is why April 7th — two days earlier, with no announcement — generated a larger residual in SPY than April 9th did.

What the pattern shows across both dates

Taken together, the April 7th and April 9th data describe a sequence. Capital moved into broad market instruments at anomalous scale two days before a policy reversal. When the reversal was announced, the market moved sharply in the direction that capital was already positioned for. The residual on the announcement day is large but unremarkable given the event. The residual on the pre-announcement day has no comparable explanation.

What Comes Next

The pattern documented here is not unique to April 2025. Market Sentinel's event database extends back through the first administration's tariff cycles, covering announcements, escalations, truces, and reversals from 2018 through the present. The April 2025 sequence is the most recent and most clearly documented instance. It is not the only one. Future editions of this analysis will extend the dataset, add instruments, and incorporate additional event types including Federal Reserve commentary, sanctions announcements, and executive orders with direct market consequences.

Market Sentinel is not a finished investigation. It is an ongoing instrument. It will keep running after this article publishes. The pattern it measures, if it is real and if it continues, will keep appearing in the data. We will keep reporting what we find.