How Signals work

One simple rule, applied to everyone's calls at once — replayed through history to see if it made money.

~4 min read · no finance degree needed · rebuilt nightly after US close

Key takeaways
  • A signal strategy is a rule that turns each day's calls into a 10-stock book, every day, automatically.
  • Every card answers one question: $100 on day one → how much now, vs random picks and the S&P 500?
  • A strategy that can't beat the random line has no edge — that's luck.

On this page: Reading a card · The strategies · Baselines · The numbers · FAQ

Reading a strategy card#

Every strategy starts with $100 on day one. The card shows where that $100 ended up — for the strategy, for random picks, and for the S&P 500:

Example
This strategy $202 · Random picks $130 · S&P 500 $113 → the rule doubled the money and finished $72 ahead of random picks. That last line is the one that matters.

Press How it works ↓ on the card for the rule in one paragraph. The left column lists every variant — click to focus it, click several to compare.

The strategies, one by one#

All of them: up to 10 stocks, equal money in each, rebuilt daily, and all of them stick to liquid mega- and large-cap names (roughly $10B and up).

Alpha Signalsthe 10 strongest calls of the day

We compare bullish and bearish calls on every stock. Better-rated speakers count a little more. The 10 strongest stocks enter the book — bullish picks are bought, bearish picks are shorted.

  • top / all speakers — count only the day's 50 best-rated voices, or everyone.
  • 1d / 2d / 7d — how far back “who's bullish right now” looks. 1d is twitchiest, 7d is calmest.
Consensusonly when 3+ top speakers agree

A stock enters only when at least 3 top speakers agree to buy or short it. Fewer, stronger picks — but some days almost nothing qualifies, so one or two stocks can swing the whole curve. High highs, wild ride.

The baselines: what counts as good?#

S&P 500— the boring bar: $100 in an index fund, walk away.
Random-10 · luck check— 10 random stocks from the same talked-about pool, fresh picks daily. The trap detector: lose to this and the “edge” was luck.
Most talked— buys whatever gets mentioned most, any direction. Pure attention: does hype alone make money?
Buzzberg Top 50— the house view: the 10 stocks the day's 50 best-rated speakers agree on most, re-ranked each morning with no look-ahead.

The chart and the numbers#

The stats columns
Total+81% = $100 → $181.
vs SPYMinus what the S&P 500 did. Negative = an index fund won.
Beta3.0 = a 1% market day is a 3% day for this book — amplified market, not skill.
SharpeReturn per unit of shaking. Above 1 decent, above 2 very good.
Max DDWorst peak-to-bottom fall. −28% = a quarter of the money gone before recovery — the “could you have held on?” number.
Avg #posNear 10 = well filled; 2–3 = single stocks can swing everything.
The event study below the chart

A different question: after a call, what does the stock do next, on average? Tens of thousands of calls, checked 5 and 20 trading days later. win rate = how often the call was right (50% is a coin flip); vs S&P 500 = beat the market, not just moved; t above 2 = unlikely to be luck. Gold rows — the day's top-50 speakers — are the headline. The same calls are also split by platform — X, YouTube, Reddit and Substack — so you can see where the edge actually lives.

Warning
Backtests and paper portfolios — not track records, not investment advice. No trading costs, one market regime, and history is replayed with today's cleaned data.

FAQ#

Huge return but a scary Max DD — why?

Concentration. Consensus on a 1-day window sometimes holds one or two stocks — when they rip, the curve rockets; when they dump, it craters. Total return is where it ended; Max DD is what it cost emotionally to get there. Sharpe balances the two.

Why are shorts so rare?

Speakers are long-biased, and each day's book is the 10 strongest leans — bullish leans dominate, so shorts almost never make the cut (typically under 1% of positions).

Why doesn't today's move show up?

Everything rebuilds once a night, after the US close. Yesterday is the newest day you'll see.

Can I trade these strategies?

Not from this page — it's descriptive. The point is measuring whether the voices are worth listening to, and which rule extracts the most from them.