Skip to content
mapperlab
← Dashboard
Backtest

The disclosure follower

Buy a fixed dollar amount of every position each marquee fund discloses (new opens and adds), at the filing-date price. Sell the moment the fund's next filing no longer shows it. Drag the slider to size each signal differently and see how the dollar figures rescale.

124,176 buy signals across 25 funds · 50,816 closed · 73,360 marked-to-market · avg hold 328 days

Position size per signal
$10,000
$500$250K

Slide to size every individual buy signal differently. Percentages (returns, alpha %, win rate) don't change — only the dollar figures rescale, since each signal's P&L is linear in position size.

Slippage assumption
Gross (no slippage)

Headline figures assume frictionless execution at the filing-date price.

Headline

Total invested
$1.24B
124,176 signals × $10k
Follower ending value
$1.46B
+17.2% on capital
SPY ending value
$1.45B
+17.1% on same dollars
Alpha vs SPY
+$1.53M
+0.1% of invested capital
Beat SPY (win rate)
41%
signals with positive alpha
Avg signal return
+17.2%
Median signal return
+3.0%
Avg hold
328d
3.6 quarters

Realized P&L curve

Cumulative profit · follower vs SPY
$0$10.0M$20.0M$30.0M$40.0M$50.0M$60.0M$70.0M2015201720192021202320252026
Follower P&L (realized)SPY P&L over the same windows

Per-fund contribution

FundSignalsInvestedFollowerSPYAlphaReturnWin
Two Sigma Investments12,536$125.36M$148.36M$144.29M+$4.06M+18.3%42%
Millennium Management14,391$143.91M$171.76M$168.87M+$2.89M+19.4%39%
Citadel Advisors19,120$191.2M$222.92M$220.29M+$2.63M+16.6%40%
Renaissance Technologies14,032$140.32M$164.82M$162.33M+$2.49M+17.5%40%
Coatue Management1,788$17.88M$23.06M$21.44M+$1.62M+29.0%41%
D.E. Shaw11,132$111.32M$130.5M$129.15M+$1.34M+17.2%41%
Appaloosa534$5.34M$8.15M$7.28M+$865.32K+52.6%47%
Point72 Asset Management6,829$68.29M$75.33M$74.59M+$736.12K+10.3%43%
Duquesne Family Office1,119$11.19M$13.19M$12.53M+$664.53K+17.9%47%
Maverick Capital6,912$69.12M$78.82M$78.2M+$624.8K+14.0%45%
Lone Pine Capital725$7.25M$9.45M$8.83M+$617.37K+30.3%45%
Viking Global Investors1,612$16.12M$19.43M$19.05M+$388.02K+20.6%47%
Scion Asset Management172$1.72M$2.14M$1.87M+$267.51K+24.2%52%
Pershing Square77$770K$1.27M$1.21M+$61.56K+64.5%55%
Greenlight Capital264$2.64M$3.47M$3.46M+$12.03K+31.4%41%
JANA Partners268$2.68M$3.03M$3.11M$74.34K+13.2%47%
Elliott Management125$1.25M$1.23M$1.35M$116.56K-1.6%42%
Tiger Global685$6.85M$9.68M$9.83M$151.46K+41.3%37%
Glenview Capital873$8.73M$11.81M$11.98M$169.15K+35.3%47%
Soros Fund Management723$7.23M$7.68M$7.88M$192.22K+6.3%41%
Third Point696$6.96M$7.87M$8.16M$286.82K+13.1%39%
D1 Capital Partners574$5.74M$6.2M$6.6M$397.1K+8.1%44%
Berkshire Hathaway298$2.98M$5.29M$5.75M$456.25K+77.5%32%
AQR Capital Management15,100$151M$176.32M$176.79M$470.27K+16.8%40%
Bridgewater Associates13,591$135.91M$153.81M$169.24M$15.43M+13.2%39%

Biggest alpha winners

Biggest alpha losses

Caveats
  • Survivorship bias. The nine marquee funds are well-known and still operating; this sample excludes funds that blew up. Aggregate alpha is biased upward versus a non-curated universe; comparisons within the sample are still informative.
  • Filing-date pricing is the theoretical earliest moment a follower could have acted. Real-world execution is T+1 to T+5; the disclosure-day alpha decays fast. (T+N sensitivity table is on the roadmap.)
  • Heavy funds (Renaissance Technologies, Citadel Advisors, Soros Fund Management, Two Sigma Investments, D.E. Shaw, AQR Capital Management, Millennium Management, Point72 Asset Management) use 8-quarter snapshots; the rest use full history. The aggregate return reflects this uneven sampling.
  • All buys priced at the 13F filing date (45 days after period-end — the earliest a follower could act). Exits priced at the next filing's date.
  • Prices are Yahoo's adjusted close— adjusted for both splits and dividends — so total return (price appreciation + reinvested dividends) is captured. SPY's ~2%/yr dividends are properly counted in the benchmark.
  • No transaction costs, taxes, or slippage modeled. Yahoo's adjusted-close has minor quirks for very old (10+ yr) historical data; per-signal returns within a typical hold window are accurate but year-over-decade returns can have ±5–10% noise.
  • Open positions are marked-to-market at the most recent close. Their sell-side P&L is unrealized.
  • Signals with unmapped tickers (small/foreign/delisted) are excluded. Options and short positions are excluded throughout.