Koraki

κοράκι — the raven watches, and keeps the record

An analyst publishes a twelve-month call. A trend rule overrides it in weeks — adding on strength, cutting on weakness, both measured against SPY. Those two things disagree, and the disagreement is expensive. Koraki runs five rule books over one set of picks and measures which one the research actually survives.

OPEN THE WATCH →

Scroll

The Method

Four stages, one measurement

Every pick enters all five books on the same day at the same price. Nothing else is held constant, and nothing else needs to be — identical inputs mean the difference between the books is the rules and nothing but the rules.

The Pick

A Top Pick appears in the morning digest. Ticker and digest date go in. The entry fills at the next open, and the SPY level is stamped alongside it.

Five Books

The same pick opens a position in all five books at once. Four apply trend rules. The fifth just holds, full size, for twelve months.

The Divergence

Books add, stop, and exit at different moments on the identical price path. One books a gain near the high while another rides it back to the stop.

The Difference

Each book is measured against the one that just held. Same pick, same path — so what's left is what the rules cost or earned.

The Lineup

Five ways to hold the same stock

One pick. Five books. Hollow marks are exits — the moment a book stopped believing.

FIXEDStop at −7% relative to entry, and it never moves. Trades live.

RATCHETSame stop, recomputed from blended cost each time it adds.

TRAILINGStop follows the peak down by 7 points. Exits first, usually.

HYBRIDFixed until the position clears +15%, then it starts trailing.

HOLDFull size at entry. No stop, no adds, twelve months. The baseline.

Written before anyone was invested in the answer

What would make us stop

A test nobody will abandon isn't a test. These four were written into the README before the first line of code, which is the only time they can be written honestly.

ATTENTION

Thirty days without opening the dashboard. Not looking is the cheapest kill signal there is, so days-since-viewed sits on the front page.

THROUGHPUT

Fewer than fifteen completed positions by month twelve. Too few observations to conclude anything, and no amount of patience fixes it.

NULL RESULT

At forty positions, every book's confidence interval containing zero and tight enough to matter. That's a real finding: the rules don't move the needle.

BROKEN RECORD

Missed entries, wrong timestamps, unadjusted splits. Better to restart clean than to publish a number built on a corrupt log.

Not on the list: a losing paper account. A negative result is a successful experiment with an unwelcome answer, and it gets published either way.