1 · The board right now
loading…
This is live. The feed re-caches every 30 minutes and a name stays on the board for as long as the flow keeps pointing the same way, so what you see below is the desk's current ranking — not a snapshot from whenever this page was published.
2 · How a name earns its score
the weights, straight from the scanner
The score is a weighted blend, not a vibe. These are the actual weights the feed is running — read them as "what this desk thinks matters for a multi-day hold":
| Factor | Weight | What it's asking |
|---|
Note the shape of it: persistence is the single heaviest factor, and IV rank is inverted — a name that has already repriced its options scores worse, not better. That's the difference between a swing board and a momentum screen.
3 · By holding horizon
bucketed on the picked contract's DTE
Every name below shows the contract the scanner actually chose — 30–60 delta, no lotto strikes — and the plan derived from its premium. The buckets come from that contract's days-to-expiry, which the scanner picks per name based on where the tradeable liquidity is. A bucket can be empty on any given day; that's information too, and it means the flow isn't reaching that far out right now.
4 · The plan, and why it's the same shape every time
stop & target
The stop is a fixed percentage of premium and the target scales with the score — a higher-conviction name gets a wider target, not a tighter stop. That asymmetry is deliberate: the thing you control on a swing is how much you're willing to be wrong by, and that shouldn't move around based on how much you like the setup.
Read the R:R, not the target. A 0.6× target on a 30% stop is 2:1. A 1.2× target on the same stop is 4:1. The board prints both so you can size on the ratio rather than on the headline number — and so a low-conviction name can't quietly become a high-conviction position.
5 · What this board deliberately doesn't do
the honest part
There is no yearly or LEAPS bucket, and there won't be. The scanner caps at 183 DTE — about six months — so the longest contract it will ever hand you is roughly two quarters out. That isn't an oversight. The signal this board ranks on is multi-day flow persistence, and flow that has pointed one way for four of the last five sessions tells you something about the next few weeks. It tells you almost nothing about next August. A twelve-month position is a thesis, not a flow read, and pretending otherwise would be the most expensive kind of false precision.
It is an end-of-day product. The feed caches for 30 minutes on purpose — hourly churn would thrash the "since alert" ledger, which is the opposite of what a swing signal is for. If a name's rank moves because of one loud hour, that's noise the cache is there to absorb.
A blank contract means no recommendation. When nothing in the 30–60 delta band inside the DTE window qualifies, the scanner returns nothing rather than reaching for a deep out-of-the-money strike. No rec beats a junk rec — so a name can rank well and still show no contract, and that's the system working.
Earnings inside the window is flagged, not filtered. If a name reports before the contract expires you'll see an ER chip on the card. Holding a swing through a print is a decision, not an accident — the board makes sure it's the former.
← AlgoxFlow