Recommendations
1 to acquire · 0 to dispose
| Ticker | Name | Close | Action | Action Price Where to act — a buy-limit on a pullback or a buy-stop above resistance, not the current price. Sanity-check vs Close: on a held name, an action price stranded above the close means the queued level was set when the stock was higher — stale, awaiting reprice. | Stop The risk-defined exit, set below the entry for a long (~1.2–1.8×ATR). A held long whose Stop sits at/above the Close is already below its stop — a stale box awaiting reprice. | 1wk Model estimate of the likely price at this horizon — a forward projection, not a guarantee, and separate from the trade's Action Price or Stop. Where the ML estimator is enabled these are per-ticker calibrated forecasts (~80% coverage band) that are refreshed periodically and can drift between refreshes; otherwise they're interpolated from the signal's swing target. Confirm against the settled close. | 4wk Model estimate of the likely price at this horizon — a forward projection, not a guarantee, and separate from the trade's Action Price or Stop. Where the ML estimator is enabled these are per-ticker calibrated forecasts (~80% coverage band) that are refreshed periodically and can drift between refreshes; otherwise they're interpolated from the signal's swing target. Confirm against the settled close. | 12wk Model estimate of the likely price at this horizon — a forward projection, not a guarantee, and separate from the trade's Action Price or Stop. Where the ML estimator is enabled these are per-ticker calibrated forecasts (~80% coverage band) that are refreshed periodically and can drift between refreshes; otherwise they're interpolated from the signal's swing target. Confirm against the settled close. | Rationale |
|---|---|---|---|---|---|---|---|---|---|
| XLY | Consumer Discretionary SPDR | 118.30 | BUY | 118.30 | 117.50 | 118.99 (+0.6%) | 120.40 (+1.8%) | 130.28 (+10.1%) | Consumer discretionary is the one fund on this board the model expects higher at every horizon - one week, four weeks and twelve weeks all sit above the close (+0.58%, +1.78%, +10.13%) - and it is the only name clearing the conviction floor for which that is true. It offers the best four-week reward-to-risk on the board at 2.63x, it closed in the bottom third of its daily range rather than at the highs, and it rose +0.24% on a session when the Nasdaq-100 fell 1.00% and the chip funds fell 2.4-2.7%, so the sector is absorbing the AI de-risking rather than being driven by it. The sector is one of only two still negative for 2026 on oil-price drag, and today's Iran sanctions package pushed oil lower, which relieves that drag. OPERATIONAL: entry is at-market at the anchor close; L8 - executes at the 2026-08-25 open, reprice if the open gaps more than ~1.5%. Sizing 49.24% x x 20% =, size_mod 1.0x (valuation blind, all-ETF book), at =. Stop = 0.50 ATR = total risk (0.066% of). SE-1/L1 haircut of -2 applied for the FailedBreakdown_Up_20D family (actionable n=12, hit_1w 0.417); still clears the floor by 6. SE-2 does NOT fire on any leg, so this is not forced into tactical-swing framing. RISK FLAGGED: XLY is ~40% Amazon+Tesla, so it is not insulated from the 08-26 NVIDIA print; the 0.50-ATR stop caps that event loss at, which is the reason the position can be carried through it. CAVEAT: the 0.50-ATR stop sits inside one average daily range (08-24 range was) and this family's stop_first rate is 0.667, so the headline 2.63x is partly a small-denominator artefact - see systemic finding 2. BRANCH: on the declared CASH row; cuts to if the 08-07 XBI card was TAKEN. · news |
| — | Consumer Staples SPDR | — | NO ACTION | — | — | n/a | n/a | n/a | Consumer staples was the best fund on the board and the top-gaining sector on the tape, up 1.70% as investors rotated defensively ahead of NVIDIA's Wednesday print - the cleanest news-to-signal alignment in the book today, and the highest conviction on the board at 85. It is declined anyway on price: the model sees only +1.05% over four weeks against a stop, a 0.92x payoff that sits under the 1.2x bar this book applies, and the entry is at-market at after the fund closed at the 99th percentile of its own daily range. Paying the high of the strongest up-day on the board for less than one unit of four-week reward is the chase this book's own playbook says to skip. OPERATIONAL: SE-1/L1 haircut -2 applied (FailedBreakdown_Up_20D actionable n=12). Second consecutive decline - the 08-21 card declined it at at 0.85x, and it has since risen +1.70% with the ratio improving only to 0.92x, so the running cost of this objection is now +1.70% and compounding. The objection is SATISFIABLE (needs a higher Est_4W or a pullback), which is what separates it from the GLD streak. METHOD FLAG: under the Target_Price convention XLP clears at 2.53x and would have been a second acquire - see systemic finding 1; this run uses Est_4W_Point for consistency with 08-21 and flags the disagreement rather than silently picking the permissive method. RE-ARM: Est_4W above the close by more than 1.2x the stop distance, on an entry not set at the top of the day's range. |
| — | SPDR Gold Shares ETF | — | NO ACTION | — | — | n/a | n/a | n/a | Gold rose 0.43% to and the 10-year yield fell to 4.70% on a session when the Nasdaq dropped 0.76%, making GLD the third-best fund on the board with conviction recovered from - the news confirms the direction and always has. It is declined for a fourth consecutive session because the trade construction fails every leg of its own written re-arm for the first time: the four-week estimate is, which is 2.47% BELOW the close, so the model expects to lose money over the horizon being bought; the order prints Buy Stop @ sitting 0.39% BELOW the close, a market order wearing a stop label rather than the genuine below-close limit the re-arm requires; and the stop at is 3.11 ATR / -5.73% from entry, the identical pin GLD has carried on 08-08, 08-19 and 08-20. The objection has never been the direction. OPERATIONAL: no scorecard bias applied - BreakoutUp_20D|buy is n=2, below the n>=5 floor, report-only in both directions. HONEST COST: first declined 08-08 at, GLD now trades - the streak has forgone +7.08%, up from +4.21% two sessions ago and the largest single running cost in this book's decline ledger. STRUCTURAL CHANGE WORTH CARRYING: EMA20 has now crossed ABOVE EMA200, repairing a ladder that had been inverted for weeks; GLD remains 16.3% below its 52-week high. Jackson Hole 08-27 to 08-29 with a Chair Warsh keynote is the event most able to reprice the rates story the whole gold case rests on. RE-ARM (fourth writing, restated in satisfiable form): a non-paying-up entry (limit at or below the close, OR at/below EMA20), stop inside 2.0 ATR, and Est_4W above the close. Size had it been eligible: x =. |
| — | Financial Select Sector SPDR | — | NO ACTION | — | — | n/a | n/a | n/a | Financials carry the strongest structural trend in the book - Trend score 92, EMA ladder fully stacked at 57.41 > 56.14 > 53.76, up 1.29% on the day and closing within 0.33% of a 52-week high - and the order is well built, a self-limiting buy-stop at that only fills on further strength. It is declined on the forward view: the twelve-week estimate is, which is 0.46% BELOW the entry and just 0.14% above the close, so the model sees no twelve-week upside at all (SE-2 fires), and the four-week payoff is 0.34x because the chandelier stop sits 2.99 ATR / away. Paying 2.04 of risk for 0.70 of four-week reward on a fund the model expects flat in three months is not a trade regardless of trend quality. OPERATIONAL: no scorecard bias applied - MomentumContinuation_Up|buy is n=2, below the n>=5 floor, report-only. Note the stop-construction contrast in systemic finding 2: XLF and GLD get ~3-ATR stops while the FailedBreakdown rows get 0.50-0.79 ATR stops, a 6x spread with nothing in between, so these payoff ratios are being driven by exit construction rather than setup quality. Jackson Hole 08-27 to 08-29 is a direct rates catalyst for this sector. RE-ARM: Est_12W back above the entry, or a stop inside 2.0 ATR that lifts the four-week payoff above 1.2x. |
| — | First Trust Cloud Computing ETF | — | NO ACTION | — | — | n/a | n/a | n/a | Cloud computing prints the best-constructed order on the board and is still declined, because conviction is 58 against a hard floor of 65. The construction is everything this book has been asking for: PullbackBuy_VWAP - the family L1 favours, actionable at n=8 with a clean +0 bias - a genuine Buy Limit at sitting 2.47% below the close rather than a market order in disguise, a sensible 1.16-ATR stop, a 2.28x four-week payoff worth 2.64 ATR of reward, and all three horizons above both close and entry (+2.81%, +5.55%, +9.94% vs entry). It fell 1.40% and closed in the bottom 9% of its daily range, which is exactly when a pullback limit should be set. OPERATIONAL: the floor is the engine's own BUY - ADD threshold and this run does not invent a new scale, so the 7-point shortfall stands. Conviction is building though - 0, 0, 54, 58 across the last four snapshots. The recurring divergence between this book's best order construction and its conviction score is logged as systemic finding 6. RE-ARM: conviction >= 65 with the PullbackBuy_VWAP limit still intact below the close. |
| XBI | SPDR S&P Biotech ETF | 164.17 | WATCH | — | 160.10 | 164.49 | 169.21 | 174.21 | Biotech fell 0.94% to and, more to the point, printed a low - its closest approach yet to the recorded stop, just / 0.28 ATR clear. The monitor still confirms no breach and the exit case has not matured: L2 requires a decisive close below stop AND a negative twelve-week view, and neither leg fires, with Est_1W, Est_4W and Est_12W all above the close (+6.12% on the twelve-week). Held on the manual stop. OPERATIONAL: signal fell back from WATCH 60 to NO TRADE conviction 0 on the verbatim reason 'Consensus conflict + low conviction'. Chandelier check: close - 2.0 x ATR =, BELOW the recorded, so the stop is HELD at and never widened. CONTINGENT ON THE UNRESOLVED 08-07 BRANCH - JSON says (flat, the assumption taken on this unattended run); the instructed branch says @, on which the position closed at for an unrecorded +, down today. The 08-21 cost-basis discrepancy is unresolved: the ledger's 08-07 card records BUY 217 @157.37 while memory has consistently used. |
| — | iShares Biotech ETF | — | NO ACTION | — | — | n/a | n/a | n/a | Biotech collapsed from WATCH 63 to NO TRADE conviction 0 on the verbatim reason 'Consensus conflict + low conviction', falling 0.48% with RSI still elevated at 68.9 - momentum pillar 24 against trend 78, which is precisely the conflict the engine is reporting. No entry, no stop, no card. The stale limit from 08-19 remains correctly unrepublished; IBB now trades 7.3% above it and that order will never fill. |
| — | Global X AI & Tech ETF | — | NO ACTION | — | — | n/a | n/a | n/a | The AI fund fell 1.78% and went from WATCH 62 to NO TRADE conviction 0 on consensus conflict, with momentum pillar at 24. It sits below its EMA20 for the first time this week. The twelve-week estimate is still +8.7% above the close, but with conviction at zero two sessions before the NVIDIA print there is nothing to act on. |
| — | SPDR Dow Jones Industrial Avg | — | NO ACTION | — | — | n/a | n/a | n/a | The Dow proxy was the only major index green, up 0.27% against the published +0.26% Dow - a 1 basis point tie-out that validates the whole snapshot. Conviction recovered, but that is 1 point under the raw floor and 3 under after the SE-1 haircut, and is 0 so the engine is not sizing it either. Est_4W is below the close (SE-2). Watch, no card. |
| — | Invesco QQQ (Nasdaq-100) | — | NO ACTION | — | — | n/a | n/a | n/a | The Nasdaq-100 proxy fell 1.00% against the -0.76% Composite, the gap explained by the index's heavier chip weight on a day chips led the decline. Conviction slipped, well below the floor, and the close broke under both EMA20 and EMA50. RSI 46.4. Forward view remains constructive (Est_12W +7.1%) but there is no actionable print. |
| — | SPDR S&P 500 ETF | — | NO ACTION | — | — | n/a | n/a | n/a | The S&P proxy fell 0.29% against the published -0.28% - another 1 basis point tie-out. Conviction shed 15 points in one session, the largest single-session conviction loss on the board, and the close slipped fractionally under EMA20. Well below the floor. |
| — | iShares Russell 2000 ETF | — | NO ACTION | — | — | n/a | n/a | n/a | Small caps fell 0.66% and lost 15 conviction points, closing below EMA20. Est_4W is fractionally below the close (SE-2). Below the floor, no card. |
| — | Roundhill Magnificent Seven ETF | — | NO ACTION | — | — | n/a | n/a | n/a | The mega-cap seven basket was remarkably resilient, down only 0.12% on a day the chip funds fell 2.4-2.7% - it sits exactly on its EMA20 (vs a close). Conviction, below the floor. The 08-26 NVIDIA print is the direct catalyst for this fund and there is no reason to pre-position into it at conviction 45. |
| — | Defiance Quantum ETF | — | NO ACTION | — | — | n/a | n/a | n/a | Quantum fell 2.35%, the fourth-worst fund on the board, caught in the same semiconductor de-risking. Conviction flat at 49, below the floor, and the close is 2.5% under EMA20. RSI 45.0. No card. |
| — | Tech Select Sector SPDR | — | NO ACTION | — | — | n/a | n/a | n/a | Technology fell 1.78% and closed below both EMA20 and EMA50, with RSI down to 44.6. Conviction - flat and well below the floor. The twelve-week estimate is +7.6% above the close, but the sector is the direct casualty of the pre-NVIDIA de-risking and the trend pillar has fallen to 50. |
| — | iShares Semiconductor ETF | — | NO ACTION | — | — | n/a | n/a | n/a | Semiconductors fell 2.67%, the second-worst fund on the board, as Intel, AMD and TSMC dropped 3-5% and NVIDIA fell 2.9% for a seventh straight red session ahead of Wednesday's print. The signal went from AVOID 64 to NO TRADE conviction 0. The twelve-week estimate of is 5.4% BELOW the close - the most bearish forward view in the book. Long-only, so this read produces no card, but as a gauge signal it is the clearest directional call on the board and the exact counterpart of the defensive rotation. |
| — | VanEck Semiconductor ETF | — | NO ACTION | — | — | n/a | n/a | n/a | The other semiconductor fund fell 2.43% and also collapsed from AVOID 71 to NO TRADE conviction 0. Est_12W is 3.6% below the close. Closed well under EMA20 and EMA50. Same read as SOXX: the gauge's most confident directional call of the week is one a long-only book structurally cannot express. |
| — | Procure Space ETF (SpaceTech) | — | NO ACTION | — | — | n/a | n/a | n/a | Space fell 2.72%, the worst non-DXYZ fund on the board, and remains the only bearish-signal name with meaningful conviction (AVOID 48, up from 31). It is the weakest structure in the book - trend pillar 36, Consolidation_Bearish, price below EMA20, EMA50 and EMA200, RSI 39.3, and 34.5% below its 52-week high. Long-only, no action. |
| — | Destiny Tech100 Inc | — | NO ACTION | — | — | n/a | n/a | n/a | The book's only priceable name fell 5.90%, by far the worst on the board, and stayed at NO TRADE conviction 0 on 'Consensus conflict + low conviction'. It is also the one row carrying real valuation: OVERVALUED_VS_RI, MoS_FV -0.78, Quality_Pass False, which would apply val_adj -10 if conviction were anywhere near the floor - it is 65 points under it, so the quality gate never binds. The valuation read is 74 days stale (FetchedAt 2026-06-11 at a price) and is advisory only. RSI 60.2. |
| — | Energy Select Sector SPDR | — | NO ACTION | — | — | n/a | n/a | n/a | Energy fell 0.83% and printed NO TRADE conviction 0 for a third consecutive session. Notable that it fell on the day Treasury announced expanded Iran sanctions - oil moved slightly LOWER rather than higher as traders assessed the package, which is the opposite of the reflexive read. RSI 67.3 keeps it among the more extended funds in the book on a signal that has been zero for three days. No card. |
| — | Industrial Select Sector SPDR | — | NO ACTION | — | — | n/a | n/a | n/a | Industrials fell 0.69% and stayed at NO TRADE conviction 0 on 'Consensus conflict + low conviction'. Est_4W is below the close (SE-2). Price is under both EMA20 and EMA50 with RSI 40.1, the second-weakest in the book. No order, no card. |
Outcome & track record
Accuracy at the time of this record.
32 Graded decisions Past recommendations old enough to score against what actually happened (target hit / stopped out / closed). This is the evidence base — the bigger the number, the more history behind the calls. Click to inspect every one. See the decisions →
-0.068 1-week calibration How much of the eventual 4-week move typically shows up in the first week (0 = almost none, 1 = most of it). Use it to read the 1-week estimates: a low number means prices drift toward the target slowly, so don't over-read a soft week-1.
50 Published records How many dated end-of-day verdicts exist for this book — the length of the public track record you can browse in the Archive below.