A public, pre-registered experiment

The Honest Bot Experiment

Five trading bots, rules frozen in advance, results published win or lose.

DRY RUN: pipeline shakeout. Ledger resets on July 6, 2026.Read the pre-registration

Scoreboard

Trendline

LONG

$101,618

+1.62%

vs Benchmark: 0.00 pp

Benchmark

LONG

The bar to clear

$101,618

+1.62%

Snapback

CASH

$100,627

+0.63%

vs Benchmark: -0.99 pp

Coinflip

LONG

$99,012

-0.99%

vs Benchmark: -2.61 pp

Oracle

CASH

$98,805

-1.20%

vs Benchmark: -2.81 pp

Equity curves

$97.8k$99.3k$100.7k$102.1kJun 11Jun 16Jun 18Jun 24Jun 26Jul 1
TrendlineSnapbackOracleBenchmark(the bar)Coinflip

All five sleeves start from $100,000 in simulated cash. Fills are at official opening prices with a 0.01% cost per trade.

The numbers

BotTotal returnSharpe (ann.)Max drawdownTrades
Trendline+1.62%2.132.52%2
Snapback+0.63%4.630.00%2
Oracle-1.20%-3.011.94%2
Benchmark+1.62%2.132.52%2
Coinflip-0.99%-1.253.72%7

Backtest, for the record (Jan 2000 to Jun 2026, QuantConnect)

Trendline: 413.67% · Sharpe 0.263 · maxDD 24.2% · 173 orders

Snapback: 287.87% · Sharpe 0.183 · maxDD 34.5% · 605 orders

The forward test exists to measure the gap from these numbers. Oracle has no backtest, by design.

Oracle's corner

Its answer, verbatim

"CASH"

Resulting signal:CASH
The frozen prompt
You are deciding a one-day position in SPY (the S&P 500 ETF) based solely on this morning's headlines. Below are the ten most recent headlines from a major financial news feed, fetched at 9:00 AM Eastern on {DATE}.

{HEADLINES}

Based only on these headlines, decide whether to be LONG SPY for today's session or to hold CASH. Reply with exactly one word: LONG or CASH.

Daily log

Most recent: July 2, 2026

TrendlineLONG
close
745.76
sma200
691.8671
SnapbackCASH
rsi2
75.9499
OracleCASH
source
cnbc-markets
headline_count
10
BenchmarkLONG

always LONG

CoinflipLONG
draw
2767461878
DateSignalsActivity
Jul 2TLSCOCBLCL·
Jul 1TLSCOCBLCLCoinflip BUY @ 745.00
Jun 30TLSCOCBLCCCoinflip SELL @ 741.29Snapback SELL @ 741.29
Jun 29TLSLOCBLCLCoinflip BUY @ 736.52Snapback BUY @ 736.52
Jun 26TLSCOCBLCC·
Jun 25TLSCOCBLCCOracle SELL @ 738.91
Jun 24TLSCOLBLCCCoinflip SELL @ 735.17
Jun 23TLSCOLBLCLBenchmark BUY @ 733.81Trendline BUY @ 733.81
Jun 22T·S·OLB·CLCoinflip BUY @ 747.70Oracle BUY @ 747.70
Jun 19TLSCOLBLCC·
Jun 18TLSCOCBLCCCoinflip SELL @ 747.76Trendline div +$257Benchmark div +$257Coinflip div +$257
Jun 17TLSCOCBLCL·
Jun 16TLSCOCBLCL·
Jun 15TLSCOCBLCL·
Jun 12TLSCOCBLCLTrendline BUY @ 740.71Coinflip BUY @ 740.71Benchmark BUY @ 740.71
Jun 11T·S·O·B·C··

Each chip: bot letter, then stance (L long, C cash). Bots: T Trendline · S Snapback · O Oracle · B Benchmark · C Coinflip

Changelog

No amendments. The rules have not changed since pre-registration.

44 pipeline runs, 11 errors. All public, including the errors.

Every table behind this page is publicly readable: prices, signals, trades, the daily ledger, the run log, and this changelog.

A bot wins only by beating Benchmark on both return and Sharpe over the full window. Six months can't distinguish skill from luck; that's what Coinflip is for.

Nothing here is investment advice. All money is simulated.

Follow by email

A machine-composed reading of the ledger every two weeks, win or lose. Double opt-in; unsubscribe in one click.