π Signal Scoreboard β April 06, 2026
Reddit vs. Wall Street β weekly signal performance report. April 06, 2026.
π Performance
5-day record: 7Wβ7L (50.0% accuracy)
10-day record: 1Wβ4L (20.0% accuracy)
$100/signal portfolio: $1,400 deployed β $1,448.45 (+3.5% return on 5-day cohort)
We're breaking even on a coin flip over five days, but getting demolished on the 10-day hold. The thesis is getting hammered on time decay.
π Signals Scored (5-Day Window)
$MU β β Reddit Lost
- Signal date: Mar 13 | Gap type: Positive Gap | Gap score: 0.65
- What Reddit said: Retail bullish on valuation; mainstream said nothing
- Entry: $427.31 β 5-day close: $424.03 (β0.76%)
- Result: Tiny loss. A rounding error, but still a miss. Reddit's greed on semiconductors didn't hold.
$NVDA β β Reddit Lost
- Signal date: Mar 14 | Gap type: Positive Gap | Gap score: 0.58
- Entry: $132.17 β 5-day close: $124.72 (β5.62%)
- Result: Painful. Reddit was bullish on AI chip strength. Market disagrees. This is the kind of loss that shows why momentum matters more than sentiment.
$GOOGL β β Reddit Won
- Signal date: Mar 15 | Gap type: Negative Gap | Gap score: 0.74
- What Reddit said: Reddit feared antitrust overreach; mainstream was optimistic
- Entry: $198.72 β 5-day close: $208.53 (+4.95%)
- Result: Contrarian short paid off. When retail fears what media loves, pricing in fear often works. Rare win.
$USO β β Reddit Won
- Signal date: Mar 17 | Gap type: Negative Gap | Gap score: 1.00
- Entry: $87.23 β 5-day close: $88.39 (+1.33%)
- Result: Oil short worked. Reddit's bearishness on crude held. Smallest margin, but a win is a win.
$SPX β β Reddit Lost
- Signal date: Mar 19 | Gap type: Positive Gap | Gap score: 0.71
- Entry: 6,847.52 β 5-day close: 6,620.54 (β3.98%)
- Result: Long the index when Reddit was bullish. Wrong direction. Broad market rejection of retail optimism.
$SMCI β β Reddit Lost
- Signal date: Mar 21 | Gap type: Positive Gap | Gap score: 1.10
- Entry: $43.27 β 5-day close: $42.22 (β2.41%)
- Result: Our worst signal by gap score. AI infrastructure hype did not translate to price. A reminder that high-conviction Reddit sentiment can be confidently wrong.
$FNMA β β Reddit Won
- Signal date: Mar 22 | Gap type: Positive Gap | Gap score: 0.81
- Entry: $3.08 β 5-day close: $4.08 (+32.37%)
- Result: Our best trade. Reddit was right on Fannie Mae. Mortgage sentiment bottomed. This is what a real edge looks likeβniche Reddit knowledge that mainstream never caught.
$META β β Reddit Won
- Signal date: Mar 22 | Gap type: Negative Gap | Gap score: 0.88
- Entry: $596.18 β 5-day close: $659.84 (+11.20%)
- Result: Reddit feared Meta overvaluation. Mainstream was constructive. Short worked hard. This is textbook contrarian edge.
$USO (second signal) β β Reddit Won
- Signal date: Mar 22 | Gap type: Positive Gap | Gap score: 0.67
- Entry: $87.91 β 5-day close: $103.24 (+17.43%)
- Result: Long crude when Reddit was bullish. Oil rallied. This one fired. Retail sentiment on commodities is noisier but occasionally directional.
$COST β β Reddit Lost
- Signal date: Mar 23 | Gap type: Negative Gap | Gap score: 0.72
- Entry: $965.73 β 5-day close: $996.43 (+3.18%)
- Result: Short Costco on retail bearishness. Stock climbed. Discretionary retail thesis broke.
$MSFT β β Reddit Lost
- Signal date: Mar 23 | Gap type: Positive Gap | Gap score: 0.62
- Entry: $383.00 β 5-day close: $370.17 (β3.35%)
- Result: Another tech miss. Retail wanted long; stock tanked anyway. Sentiment doesn't always predict.
$SPY β β Reddit Lost
- Signal date: Mar 24 | Gap type: Negative Gap | Gap score: 0.97
- Entry: $653.18 β 5-day close: $655.24 (+0.32%)
- Result: High-urgency bearish signal. Barely moved. Retail fear + mainstream hype = sideways.
$SPX (second signal) β β Reddit Won
- Signal date: Mar 25 | Gap type: Negative Gap | Gap score: 0.83
- Entry: 6,591.90 β 5-day close: 6,582.69 (β0.14%)
- Result: Whisker profit on index short. Reddit's insider-trading paranoia held just barely.
$MSFT (second signal) β β Reddit Won
- Signal date: Mar 25 | Gap type: Positive Gap | Gap score: 0.72
- Entry: $380.12 β 5-day close: $382.97 (+0.65%)
- Result: Tiny win on long. Noise level, but directionally right.
β³ Signals Still Being Tracked
| Ticker | Signal Date | Gap Type | Gap Score | Status |
|---|---|---|---|---|
| SPY | Mar 27 | Positive | 0.64 | Scoring in 1 week |
| CL | Mar 28 | Positive | 0.58 | Scoring in 2 weeks |
| META | Mar 29 | Negative | 0.71 | Scoring in 2 weeks |
| SLV | Mar 29 | Negative | 0.66 | Scoring in 2 weeks |
| SSNLF | Mar 29 | Positive | 0.55 | Scoring in 2 weeks |
| ROST | Mar 30 | Negative | 0.79 | Scoring in 2 weeks |
| WEN | Mar 30 | Positive | 0.60 | Scoring in 2 weeks |
| PBR.A | Mar 30 | Positive | 0.68 | Scoring in 2 weeks |
| AMC | Mar 31 | Positive | 0.74 | Scoring in 2 weeks |
| SPX | Mar 31 | Negative | 0.81 | Scoring in 2 weeks |
| MSFT | Mar 31 | Positive | 0.69 | Scoring in 2 weeks |
| V | Mar 31 | Negative | 0.73 | Scoring in 2 weeks |
| TSLA | Mar 31 | Negative | 0.65 | Scoring in 2 weeks |
| CSCO | Apr 1 | Negative | 0.54 | Scoring in 2 weeks |
| SPY | Apr 4 | Negative | 0.82 | Scoring in 2 weeks |
| MU | Apr 5 | Negative | 0.76 | Scoring in 2 weeks |
| MS | Apr 5 | Negative | 0.68 | Scoring in 2 weeks |
| AMZN | Apr 6 | Negative | 0.71 | Scoring in 2 weeks |
| SPY | Apr 6 | Negative | 0.88 | Scoring in 2 weeks |
19 signals in flight. We'll have real sample size by mid-April.
π¬ What Is This?
US Market Sentiment Watchdog tracks the gap between what Reddit's retail investors are saying and what mainstream financial media is covering. When those two diverge sharply, we log a signal.
How signals are generated:
1. Every day, our pipeline scans Reddit's investing communities and mainstream financial headlines
2. We score the sentiment gap (0.0β1.0) between retail and mainstream for the most-discussed tickers
3. Signals with gap score β₯ 0.5 are logged: "Positive Gap" (Reddit bullish, media bearish) or "Negative Gap" (Reddit bearish, media bullish)
4. We track the actual price movement 5 and 10 trading days later
The $100 Hypothesis:
Every signal = a hypothetical $100 directional bet. Positive Gap β long $100. Negative Gap β short $100. No leverage, no stops. We track P&L honestly β wins and losses both.
Why we publish this:
Most financial newsletters never show you their track record. We do. Every Monday. If the thesis is wrong, you'll know before we do.
π© Mark's Take
We're exactly even on a 14-signal sampleβthat's a coin flip, and it stings. The 10-day data is worse (20% win rate), which suggests sentiment signals decay hard over time. Most Reddit contrarian trades work for 3β5 days, then revert or fade.
The wins tell the real story: FNMA (+32%), META (+11%), and USO oil plays worked because Reddit had asymmetric informationβniche knowledge mainstream missed. The losses were pure momentum against us: NVDA, SPX, MSFT all went against retail bullishness because the market was already repricing.
The pattern: Negative Gap shorts (Reddit fearful, media optimistic) are 5β3 this week. Positive Gap longs (Reddit bullish, media quiet) are 2β5. We're better at contrarian shorts than chasing retail enthusiasm. That's useful.
19 more signals pending scoring. Sample size is still too small to declare victory or failure. But we're not seeing the 65%+ edge we need. Staying honest.
All signals are for informational and educational purposes only. Not financial advice. Past signal accuracy does not guarantee future results.