Knowledge · Research · Trump-post event study: simulated trading ROI (9 BTC events, 2025)

Trump-post event study: simulated trading ROI (9 BTC events, 2025)

Market research 2026-05-10 3 strategies analyzed 8 sources
Event study: what would directional BTC trading after the 9 biggest Trump posts of 2025 have yielded? Entry 5 min after the post, trailing stop 1-3%, max hold 6h. Simulation on real Binance 1m data. Result: +19.5% total P&L (9 trades, 89% win rate with a 1.5% trail). But with an important selection-bias disclaimer.
  • 9 simulated events: 89% win rate (8/9) with a 1.5% trailing stop, avg P&L +2.17% per trade, total P&L +19.5% (compound +21.2%).
  • Best trailing stop: 1.5% - highest win rate (89%), solid avg P&L. 2% and 3% have somewhat more raw P&L but a worse win rate (78%).
  • Tariff/risk-off posts (5 events) are the most consistent short signal - all 5 profitable with a 1.5% trail. Average +2.5% per short trade.
  • Crypto-positive posts (3 events): 2/3 profitable. Strategic Reserve post (Mar 2) best long trade (+6.3% with a 2% trail). Jan 23 EO was the only clear losing event (wrong move detection within the window).
  • October 2025 China tariff: biggest single winner - short +3.88% (3% trail) on a BTC drop from $122k to $103k (-16%).
  • 5-minute entry delay works: not too early (avoids buying the spike directly), not too late (the first big move is not yet over).
  • SELECTION BIAS: only events that are known because they moved BTC. In reality ALL Trump posts would have to be classified - hundreds of irrelevant posts would produce false positives.
  • BTC data available Jan 2020 - Apr 2026 (Binance 1m futures). All 9 events fully covered.
P1 Do not implement directly as a Botty strategy
Without a reliable, low-cost Trump-post feed this is not feasible. The Twitter/X API costs $5k/month for real-time. Truth Social has no official API. Unofficial scrapers are unstable and legally grey.
Implementation: Do not implement.
Evidence: 9 events in 10 months = 0.9 trades/month. Even if all perfectly classified: too little frequency for a standalone strategy.
P2 Use tariff events as a manual stop-loss warning
If Peter reads a tariff post while an open long position is running: manually tighten stops. 5/5 tariff events were profitable shorts - the signal is consistent.
Implementation: Manual. Peter reads the post -> opens Botty -> lowers the ATR multiplier or closes the position manually.
Evidence: Tariff events in the test: all profitable as shorts. Avg -7.6% BTC move. Even with a 1% trail: positive.
P3 Document the event-classification logic for a later expansion stage
If a Trump feed ever becomes cheaply available (e.g. third-party aggregators at ~$20-50/month): the classification rules are clear: tariff/trade -> short, crypto-positive -> check for long, immigration/other -> ignore.
Implementation: Document in product/memory/modules/briefings.md. The briefings module can handle Trump-post classification on manual input.
Evidence: Keyword sets from Gjerstrad/Molnar: 'tariff', 'products', 'trade' for negatives. 'bitcoin', 'crypto', 'reserve' for positives.
Trump-event directional trading (1.5% trail)
2/5 4/10
Trump post classification (topic + sentiment) BTC price momentum 5 min after the post Trailing stop loss 1.5% off the peak
period
Jan 2025 - Oct 2025 (9 events)
max hold
6 hours
n events
9
win rate
89% (8/9)
data type
Simulation on real Binance 1m futures data
best trade
+4.76% (Liberation Day short, Apr 2)
entry delay
5 minutes after post detection
worst trade
-0.18% (Jan 23 EO long)
total pnl linear
+19.49%
avg pnl per trade
+2.17%
total pnl compound
+21.15%
The numbers look good, but they are highly selective. We only took the 9 best-known Trump-BTC events of 2025 - all guaranteed to be known because they moved the market. In reality hundreds of posts would have to be classified continuously, and most ('immigration', 'geopolitics in general') produce no BTC move. The real edge is in the classification, not in the trailing stop.
Trump-event directional trading (1.0% trail)
2/5 3/10
Trump post classification Trailing stop 1.0%
max hold
6 hours
win rate
89% (8/9)
entry delay
5 minutes
avg pnl per trade
+1.03%
total pnl compound
+9.58%
1% trail triggers too early - exit at normal volatility, before the actual move is done. Best trade on the Reserve post: +3.77% instead of +6.33% (with a 2% trail). Too tight for post-event volatility.
Trump-event directional trading (3.0% trail)
2/5 3/10
Trump post classification Trailing stop 3.0%
max hold
6 hours
win rate
78% (7/9)
entry delay
5 minutes
avg pnl per trade
+2.31%
total pnl compound
+22.47%
More raw P&L but a lower win rate. Jan 23 EO and the Liberation Day follow-up day become losers because the stop is too wide and BTC reverses. Worst trade: -1.64% instead of -0.18% with a 1.5% trail.

Overview

This study simulates what directional BTC trading after the 9 best-known Trump posts/events of 2025 would have yielded. The question: is it worth monitoring Trump and trading immediately?


Method

Data basis: Binance 1m BTCUSDT futures (Jan 2020 - Apr 2026, 3.3M bars)

Event detection: For each documented event a time window is defined (e.g. 20:00-04:00 UTC for Liberation Day). Within the window the start point of the strongest directional move is found automatically (60-bar forward algorithm).

Entry: 5 minutes after T0 (simulates post-detection delay + NLP classification).

Exit: Trailing stop (1.0%, 1.5%, 2.0%, 3.0% off the peak) or after 6 hours (half-life of the signal effect per the literature: 36-48h, but the simulation conservatively capped at 6h).


Results (1.5% trailing stop)

Event Date Direction P&L
Crypto-national-priority EO Jan 23 Long -0.18%
Strategic Reserve Truth Social post Mar 2 Long +3.25%
Canada/Mexico/China tariffs Mar 4 Short +2.64%
Strategic Bitcoin Reserve EO (disappointment) Mar 6/7 Short +3.11%
Liberation Day 185-country tariffs Apr 2 Short +4.76%
Liberation Day follow-up day Apr 7 Short +0.33%
'GREAT TIME TO BUY' + tariff pause Apr 9 Long +2.77%
401k crypto executive order Aug 7 Long +0.76%
100% China tariff threat Oct 10 Short +2.05%
Total +19.49%

Win rate: 89% (8/9) | Avg +2.17% | Compound +21.15%


What the numbers mean

The 89% win rate and +21% in 10 months sound strong. But two fundamental caveats:

1. Selection bias: We only took events that are known because they moved the market. Trump posts 10-50 times daily. 95%+ of his posts are irrelevant to BTC (immigration, politics, sports commentary). In reality one would have to classify all posts - and the false positives would massively dilute performance.

2. Data sourcing: A reliable, real-time Trump feed costs $5k/month (Twitter Pro API) or is unofficial and unstable via Truth Social. 9 trades in 10 months = 0.9 trades/month - too little frequency for a standalone strategy.


Practical conclusion

The attempt to trade the pattern is not worthwhile as a standalone Botty strategy, but very much as a manual risk filter:

  • Tariff post detected -> flag open longs as at risk, tighten stops
  • Crypto-positive post -> no blind buy, but a momentum check: is EMA/MACD currently bullish? Then possibly hold
  • Everything else -> ignore

Botty's ATR-based position sizing and trailing stops already automatically absorb the elevated volatility regime after Trump posts - without any need for Trump monitoring.