For investors, not traders
Everything worth watching,
read by one mind.
The price and the momentum behind it. The news. What the company earns and what it owes. Who runs it, what they promised, and whether they delivered. All weighed together, then one answer: worth buying, not yet, or stay away, for the next three to six months.
1 · What each source found
2 · How much each one counted
The number is how strongly that one has predicted the next six months, measured across the whole watchlist. Bigger means it earned more say.
3 · The answer
Moderate. Five signals agree, one does not.
Where it disagrees. The downgrade disagrees with everything else here. That is the finding, not noise, so it is shown rather than averaged away.
What would break it. A close below the trend line would end this read. So would a second downgrade.
The horizon
Not for trading.
A one day call is barely better than a coin flip, so we took those out.
The deliverable
Four things, every time.
The same shape every time, so two companies compare.
A direction
No price target. Nobody can produce one honestly this far out.
A horizon
Fixed first, so it cannot be re-dated into being right.
The reasoning
Every factor that counted, including those against.
What would change it
Named before the fact: a level to watch, or the day it falls behind the market. If one goes, you hear that day.
What goes in
Four kinds of evidence, one call.
What the price did
Trend, momentum, drawdown, and how it moved against the index.
The news, labelled
Marked good or bad, with a reason. Most are thrown out: they name the company without being about it.
The business underneath
What it earns, what it owes, and what the price pays for that.
The calendar
Results dated in advance. You hear before, not after.
Weights measured, not chosen.
Each one ranked against what the price actually did next.
- 50/200 gap +0.228
- Price vs 200d +0.227
- Vs the index, 6m +0.210
- Momentum 6m +0.207
- Range position +0.204
- Drawdown +0.188
- Momentum 3m +0.165
- Momentum 12m +0.114
- Volatility -0.009
The honest test
Being right is not the same as being useful.
Say a tool gets it right 68 times out of 100. Good? It depends entirely on what the market did without it.
would have beaten it.
So we never show you a hit rate on its own. Every call is scored against what the market did by itself, and only the difference counts.
Why this one
Built to be checked, not believed.
Confidence in a method is not confidence in an outcome.
It keeps score in public
Every call graded against the index. The wrong ones stay on the record, and you can read it before you trust it.
It commits first
Horizon and breaking level fixed up front. A story written after the fact fits anything.
It is allowed to say no
When nothing qualifies it says so, and holds the cash.
It does not sell certainty
A six month read is a probability, not a promise. We grade the edge rather than quote it at you.
Start with one you already have an opinion on.
Point it at something you understand.
Sign in with Telegram. No password, no form.
Name one company. Two years of history loads on the spot.
Disagree with it. The reasoning is on screen, factor by factor. If it tells you nothing new, you lost two minutes.
Invite only until the scorecard is worth reading.