Seychelles Capital Flows

The tests show no consistent, material advantage over the no-change baseline.

Model: Agentic Investor TSFM (foundation model)

Accuracy vs naive

+13%Beats the no-change baseline over 8 tests (8 tests, 4 won)

Direction

59%No clear edge over a coin flip (19 of 32 calls right in 8 tests)

Turning points

9 of 18Anticipated most turns (18 turns in 8 tests, 0 without a turn)

Error vs baseline

13% lowerError ratio 0.87; 1.00 is a tie.

Tests won

4 / 8Complete rolling backtests; ties count as no win.

Test period

2018-09-01 to 2026-06-014 observations ahead per test.

80% band coverage

81%Observed outcomes inside the band across all stored test windows.

• Band coverage uses all stored test windows, including model-selection windows when present; it does not determine the confidence label.
• Evaluated 2026-11-15 against the cached history through 2026-06-01.
• This is an evaluation of the serving model; it does not establish a winner across four model classes.