A controlled test of trend rules against analyst conviction
Six books

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 six books over one set of picks and measures which one the research actually survives.

OPEN THE WATCH →
Scroll

Koraki

Six books, one set of picks. Watching what the rules do to the research.
↑ TOP   RESEARCH LIBRARY →
POSITIONS
HOW TO READ THIS
TICKER
The pick. TRADED means you own it for real and the sixth book is live. HALT means a bar looked wrong — nothing triggered, it’s waiting for you.
2 HOUSES
The same name was flagged independently by a second research house. Only the first appearance is logged as a pick — a second position on an identical price path would not be a second observation.
ADDED
The date the pick was published. Not the day you typed it in.
DAYS
Trading days held. The time stop fires at 50 if the position never added.
DIVERGENCE
Six lines leaving one entry point. Dashed is hold. A hollow circle is the day that book exited; a filled dot means it’s still in. When the lines fan wide, the rules are disagreeing — that’s the whole experiment in one mark.
RETURN
Plain price return since entry — what a brokerage screen shows. Where the stock is versus what it cost.
RS
The same move measured against SPY. +5% means it beat the index by five points; a stock up 8% while SPY rose 8% shows 0. The gap between RETURN and RS is beta — the part of the move that was just the market. Every rule in Koraki fires on RS, never on RETURN.
TO STOP
The column that earns a daily look. Points of relative move left before the fixed stop fires. Teal is room, amber is close, red is about to go. Rows are sorted by this — whatever is nearest the edge sits at the top.
LOTS
How many of the four planned buys have filled. Adds trigger at +7%, +14%, +21% RS.
ALPHA $
Dollars ahead of — or behind — the same money sitting in SPY over the same days. Not profit and loss. A position can be up in dollars and negative here.
BOOKS OPEN
Six dots, one per book, in lineup order. Filled means still holding, hollow means exited. Six of six is a fresh pick; one of six means five books have quit.
CLICK A ROW
Expands all six books for that pick — lots, capital deployed, each book’s own RS, its stop level, and why it exited.
Nothing here asks you to act. The books run on their own rules; the scoreboard below stays empty until positions complete, and it will say nothing meaningful for months. The one thing that keeps this alive is a new line in picks.csv when a name comes in — and log_trade.py on the days you actually buy something.
Row values are the fixed book. Click a row for all six. Dashed line in each glyph is hold.
DID THE PICKS BEAT THE INDEX?
CLOSED POSITIONS
PAIRED DIFFERENCE VS HOLD
The method
THE METHOD

Four stages, one measurement

Every pick enters all six 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.

01

The Pick

A name enters the watchlist. Ticker and date go in. The entry fills at the next open, and the SPY level is stamped alongside it.

02

Six Books

The same pick opens a position in every book at once. Four apply trend rules. One just holds, full size, for twelve months. The sixth records what actually happened.

03

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.

04

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

Six ways to hold the same stock

ENTRY TODAY
One pick. Six books. Hollow marks are exits — the moment a book stopped believing.
FIXEDStop at −7% relative to entry, and it never moves. The live book.
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.
ACTUALNot simulated. Recorded — real fills, real sizes, reasons written at the time.
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 this page. Not looking is the cheapest kill signal there is.

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, the clustered 90% interval on fixed minus hold sitting entirely within ±$150 per pick per $10k and containing zero. An edge smaller than that gets ignored in practice anyway, which is what makes it the right threshold. 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.

READ THE DESIGN NOTES →