Every optionable name on the watchlist is read on its closed 5-minute candles, a few seconds after each candle closes, for one setup:
BREAKOUT → RETEST → VWAP HOLD → CONTINUATION
On the left is the board of proposed trades and, for the name you pick, the full read: technicals, the confirmation checklist for each side, support and resistance, the execution table, the option analysis, what the record can say about its grade, and a plain CALL, PUT or WAIT. On the right is the chart those readings came from — the 5-minute candles with the 8 EMA, VWAP, volume and RSI, and the premarket high and low (PMH / PML) and opening-range high and low (ORH / ORL) drawn on it.
Open Daily Claude →This is a members-only tab. The guide below is free to read; opening the live board requires a signed-in paid account.
Every one of those is decided on a closed candle. The forming candle is drawn on the chart but never read, because "a confirmed 5-minute close" is the rule the whole framework leans on.
The grade is a count of confirmations that were met — not an impression of how good a setup looks. Each one is listed on the tab with the number behind it.
A+ every supporting confirmation that can be measured passes, at least four of them · A three or more pass, at most one fails · B two pass · C fewer. Entering within half an R of a key level, or with RSI past 78 (or below 22), caps the grade at C — those are the framework's own "don't buy into resistance" and "don't chase" rules. Only A+, A and B are ever offered as a CALL or PUT. A C is tracked for the record and shown as WAIT, before it triggers and after.
SPX has no volume, so it has no VWAP of its own; SPY's side of its VWAP stands in, and the tab says so wherever it is used. The volume checks read n/a for SPX, which means an SPX setup can reach A but not A+.
The stop goes a tenth of an ATR beyond the retest's extreme. The three targets sit on the next levels in the trade's direction when one is at a sensible distance, and at 1R, 2R and 3R when not — the tab says which for each. Targets are set from the levels that existed when the trade armed, so a plan never points at a high made after it. Risk/reward is the distance to each target measured against the stop.
Read off the live chain for the nearest expiry — never estimated. When the chain cannot be read, the tab says so and shows nothing in its place.
Low, Medium or High, from the confirmations that are missing: ordinary breakout volume, a level that has already failed once today, a key level within 1R, the index against it, RSI out of its band, or a continuation candle that did not expand volume. Each reason is listed.
Every trade the engine triggers is written down at that moment — entry, stop, targets and grade — and never edited again; only its outcome moves, and only forward. It is settled against the candles that follow: first touch wins, a candle that touches both the target and the stop counts as the stop, and anything still open at 15:55 is marked out at that candle's close.
The tab shows a grade's hit rate only once at least 20 of its trades have resolved, always with the count and the standard error beside it. Until then it says how many it has. The C trades the tab never offers are tracked too, so that whether an A really does better than a C can be checked rather than assumed. Trades first read after their session had closed — a restart over a weekend, say — are kept in the record but left out of the numbers, because nobody saw them live.
Before launch, the engine was replayed over eight sessions on eight names. It triggered 89 trades: A+ reached its first target on 1 of 1, A on 6 of 18, B on 14 of 37, and C on 16 of 33. The C trades the tab never offers did slightly better on average than the ones it does. At that size those numbers are all within noise of each other, so there is not yet any evidence that the grades predict outcomes. The live record is how that will be settled.
Once an hour a scheduled Claude routine reads the board and writes a short commentary — the market in a few sentences, and a note on the setups worth attention. It appears above the board and on each name it mentions. It is commentary on the readings: it cannot change a grade, a level or a decision.
It is not a recommendation to trade. It is a rulebook applied the same way to every name, every five minutes, with its workings shown — where the levels are, what has happened at them, what still has to happen, and what would prove the setup wrong. 0DTE options decay fast and gap through stops; the record on this page is of the underlying's price reaching its levels, not of an option's profit and loss.
Related: the Trades tab for the rule-based AI desks that actually settle and score their trades, the Daily Levels tab for the per-ticker daily plan, and the 0DTE HeatMap for the dealer gamma picture. See them all on the Help hub.
Educational content only — nothing on this page is financial advice or a recommendation to buy or sell any security. Setups are produced by fixed rules applied to market data, with written commentary from an AI model; neither is a forecast, and historical results do not guarantee future performance. Markets involve risk. © SPX Lens · About · Help & FAQ