Total-Return Index 97.9 -2.1% since inception Performance ›
Today's recommendations
0 to acquire · 1 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 |
|---|---|---|---|---|---|---|---|---|---|
| TSM | Taiwan Semiconductor Manufacturing C | 404.25 | SELL ALL | — | 406.37⚠ | n/a | n/a | 347.18 | Closing the last of a position we have been stepping out of since mid-July. TSMC's business is not the problem - Q2 revenue grew 36% with net income up 77%, and the stock bounced 7% last week on renewed AI-infrastructure confidence - but our position is still below its protective stop for a seventh session, still under both its 20- and 50-day trend lines, and the model's 12-week projection of sits 14% below today's price. The signal engine itself now reads the name as directionless. We take the bounce as a better exit price, not as a reason to re-open a trade whose case has not recovered. Realised loss about -7.7%. OPS: 6th arming, executor stalled 7 sessions (UT-11 re-fire, do not re-litigate on a bounce). L2 legs: neg-12wk holds (-14.12%); stop-break decayed to -0.52% (0.10 ATR, not decisive). Written withdrawal trigger TESTED AND NOT FIRED - needs 2 consecutive settled closes ABOVE 406.37 with a rising 12wk; closes were 403.31 and 404.25, both below. Trigger carried forward unchanged. G2 suppresses the valuation amplify (DCF vs, ~9x broken). PROFIT_ARMED at T12_Progress 116.44 with Gain_ATR -1.67 is the UT-9 arm-on-loss mislabel, inert - this is a loss-cut, never a profit-take. Proceeds, realised ~-. TSMC July monthly revenue 08-10 falls 5 sessions after the exit - not a sale into a catalyst. Model band 373.46-426.82. · news |
| DELL | Dell Technologies Inc | 405.37 | HOLD | — | 358.88 | n/a | n/a | 511.06 | Holding all. Last session's proposed half-exit was withdrawn and today's fresh read confirms that call rather than inheriting it: the shares trade 13% above our protective stop, the model's 12-week projection of sits 26% above today's price, and none of the deterioration signals we said would trigger a trim has appeared. The business news supports patience - a raised dividend paid this week, $3bn of new notes and a $6bn revolver secured, and new Vera Rubin-based AI servers shipping to CoreWeave. We are down about 7.7% on the position and see no signal-based reason to realise it. OPS: not in the sell pool - 0, blank Sell_Rank, no Exit_Warning code, Gain_ATR -0.93 so NOT profit-armed. All 3 written completion triggers unfired (close +12.95% above 358.88; 12wk +26.07% above close; no NO-TRADE streak, label is AVOID). Stop HELD at 358.88, NOT raised - the engine's Stop_Price 405.37 is the degenerate entry=stop=close AVOID-row artifact and ATR 36.62 is 9.0% of price, so tightening invites a whipsaw. UT-10: no ADD, no averaging down (both prior pullback ADDs graded stop_hit). Held value; risk to stop. · news |
| AMZN | Amazon.com Inc | 271.58 | NO ACTION | 258.62 | 223.92 | 279.43 (+8.0%) | 328.00 (+26.8%) | 256.85 (-0.7%) | Standing aside on the strongest signal this book has produced in weeks, and it is worth being clear why. Amazon's quarter was outstanding - revenue 6bn and operating income 5bn both well ahead, with AWS growing 36.7%, its fastest in eighteen quarters, on a $496bn backlog - and the stock rose 15.3% on it. Two things stop us buying today. First, our fundamental quality screen still fails the name, and that screen is computed from financials filed before this quarter, so it may simply be a step behind reality. Second, and independently, buying a stock the day after a 15% earnings gap is the exact chase our own record says loses money; the model's own entry is 5% below today's close. We would rather wait for the pullback and for the fundamentals to be re-measured than pay up on the day. OPS: adj 70 CLEARS the 65 floor - first candidate in 8 sessions to do so - but Quality_Pass FALSE (F-Score 6, Quality_Rank 21/26, FV method EPV, DCF blank) is a HARD block on new entries. Independently blocked by L6 (post-earnings gap-chase) and the no-chase rule (ED -0.95 <= -0.25 -> queue-only). RE-ARM TRIGGER: re-run fairvalue once Q2 financials land in the fundamentals feed; if Quality_Pass flips TRUE and price pulls back to ED >= -0.25 of the then-live Entry_Price, AMZN becomes this book's leading acquire at adj ~70. Sizing had it been taken: @ 258.62 (0.6695 x x 0.20 = x0.75); entry range 255.21-262.03. NOTE the engine defect: 12wk 256.85 is BELOW both the close (271.58, -5.4%) and its own entry, on a conv-80 Rank#1 row. No DCF anchor exists to annotate against. · news |
| SNOW | Snowflake Inc. | 293.28 | NO ACTION | 301.37 | 256.05 | 328.56 (+9.0%) | 392.00 (+30.1%) | 417.92 (+38.7%) | Passing on Snowflake for an eighth straight session. The momentum is real - the stock made a 52-week high this week and the trade would only trigger on further strength - but the company still fails our fundamental quality screen on profitability, and on the valuation work its earnings power is negative on a per-share basis, ranking 23rd of 26 names in the book. Paying above a record high for a business that does not yet cover its own cost of capital is not a trade we want, and the signal is not strong enough to argue otherwise. OPS: adj (val_adj -8 at the recomputed anchor MoS -1.61), three below the 65 floor, AND independently hard-gated on Quality_Pass FALSE (F-Score 5, EPV -/sh, rank 23/26) - 8th consecutive session. L1/UT-12 discounts MomentumContinuation_Up (own-book cell 0/2 on 1wk, report-only). Entry 301.37 is a buy-stop ABOVE the 293.28 close, ED +0.54 so NOT extended - the block is quality and floor, not chase. Sizing had it been taken: @ 301.37 (x0.75). 12wk 417.92 EXCEEDS DCF anchor - projection assumes growth beyond priced-in. |
| CRM | SalesForce Inc | 184.02 | NO ACTION | 184.02 | 177.03 | 189.66 (+3.1%) | 202.83 (+10.2%) | 191.72 (+4.2%) | Salesforce is the only buy candidate today that passes our fundamental quality screen, and it is reasonably valued, but we are still not buying it. The trading signal appeared in a single session - the name went from no signal at all to a buy overnight - on the least reliable of our setup patterns, and it arrives straight into a major broker downgrade: Morgan Stanley cut the stock to neutral last week and lowered its target by 35% to questioning when Agentforce revenue actually shows up. The shares closed at right on that new target. A one-day signal into a fresh downgrade is not where we want to spend capital. OPS: adj (val_adj -10 at recomputed anchor MoS -2.09), six below the 65 floor. Quality_Pass TRUE (F-Score 7, rank 15/26, FAIR_VALUE, Bargain T) - clears the gate that killed AMZN/SNOW/NET. Conviction trail 0/0/0/69 = zero stability. L1: FailedBreakdown_Up_20D is the weakest chase family (own-book cell n=1, stop-first 1.0). Entry 184.02 = close, ED -0.0, clean non-extended entry, range 181.67-186.37. Sizing had it been taken: @ 184.02 (x0.75). 12wk 191.72 well under the DCF - no annotation. · news |
| NET | Cloudflare, Inc. | 278.98 | NO ACTION | 278.98 | 271.04 | 288.51 (+3.4%) | 310.73 (+11.4%) | 357.54 (+28.2%) | Cloudflare triggered a buy signal but ranks last of all 26 valued names in the book on fundamental quality, and its per-share earning power is effectively zero, which makes any margin-of-safety measure meaningless. The setup is a weak pattern on our record, and the conviction score does not clear our bar even before the quality screen blocks it. No action. OPS: adj (val_adj -10), below the 65 floor, AND Quality_Pass FALSE with the worst rank in the book (F-Score 3, 26/26). MoS_FV -940.26 is a broken denominator (EPV /sh) - see systemic finding 7; it pins val_adj at the floor for a reason unrelated to valuation. size_mod 0.5 follows mechanically from MoS <= -3. Ledgered per the UT-12(d)(iii) measurement-gap backlog so the quality gate's cost on high-Buy_Rank names can be priced. |
Outcome & track record
How accurate the calls have been.
9 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 →
0.134 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.
24 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.
Portfolio & signals
Every name we scanned, and where it stands.
| Ticker | Name | Close | Signal The chart pattern the engine detected on this name. Hover a signal for its one-line definition; click it for the full setup/trigger rules, the trade box it produces, and the newsletter issue that teaches the underlying concept. | Action | Conv | 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. |
|---|---|---|---|---|---|---|---|---|
| DELLH | Dell Technologies Inc | 405.37 | FailedBreakout_Down_20D | AVOID | 57 | n/a | n/a | 511.06 (+26.1%) |
| TSMH | Taiwan Semiconductor Manufacturing C | 404.25 | NoTrade | NO TRADE | 0 | n/a | n/a | 347.18 (-14.1%) |
| AMZN | Amazon.com Inc | 271.58 | BreakoutUp_20D | STRONG BUY | 80 | 279.43 (+2.9%) | 328.00 (+20.8%) | 256.85 (-5.4%) |
| AAOI | Applied Optoelectronics, Inc. | 94.32 | FailedBreakout_Down_20D | AVOID | 76 | n/a | n/a | 53.76 (-43.0%) |
| KLAC | KLA Corporation | 182.82 | FailedBreakout_Down_20D | AVOID | 76 | n/a | n/a | 124.75 (-31.8%) |
| TSLA | Tesla Inc | 311.21 | FailedBreakout_Down_20D | AVOID | 76 | n/a | n/a | 177.39 (-43.0%) |
| PLTR | Palantir Technologies Inc. | 123.06 | FailedBreakout_Down_20D | AVOID | 76 | n/a | n/a | 97.40 (-20.9%) |
| GOOGL | Alphabet, Inc.- Class A | 356.13 | FailedBreakout_Down_20D | AVOID | 73 | n/a | n/a | 301.01 (-15.5%) |
| SNOW | Snowflake Inc. | 293.28 | MomentumContinuation_Up | BUY | 70 | 328.56 (+12.0%) | 392.00 (+33.7%) | 417.92 (+42.5%) |
| CRM | SalesForce Inc | 184.02 | FailedBreakdown_Up_20D | BUY | 69 | 189.66 (+3.1%) | 202.83 (+10.2%) | 191.72 (+4.2%) |
| NET | Cloudflare, Inc. | 278.98 | FailedBreakdown_Up_20D | BUY | 68 | 288.50 (+3.4%) | 310.73 (+11.4%) | 357.54 (+28.2%) |
| META | Meta Platforms | 556.71 | RallySell_BearTrend | AVOID | 68 | 547.04 (-1.7%) | 524.49 (-5.8%) | 455.65 (-18.2%) |
| ARM | Arm Holdings plc | 239.69 | FailedBreakout_Down_20D | AVOID | 68 | n/a | n/a | 136.62 (-43.0%) |
| CRWV | CoreWeave, Inc. | 71.77 | FailedBreakout_Down_20D | AVOID | 68 | n/a | n/a | 40.91 (-43.0%) |
| IREN | IREN Limited | 36.80 | FailedBreakout_Down_20D | AVOID | 68 | n/a | n/a | 20.98 (-43.0%) |
| MSFT | Microsoft Corp | 464.72 | ReversalDown_Overbought | AVOID | 66 | 445.96 (-4.0%) | 402.19 (-13.5%) | 513.69 (+10.5%) |
| AVGO | Broadcom Inc | 389.28 | FailedBreakout_Down_20D | AVOID | 64 | n/a | n/a | 354.67 (-8.9%) |
| AAPL | Apple | 308.91 | FailedBreakdown_Up_20D | WATCH | 60 | 316.75 (+2.5%) | 335.04 (+8.5%) | 355.44 (+15.1%) |
| QCOM | Qualcomm Incorporated | 147.61 | ReversalUp_Oversold | WATCH | 58 | 154.80 (+4.9%) | 171.59 (+16.2%) | 84.14 (-43.0%) |
| NFLX | Netflix, Inc | 71.71 | FailedBreakout_Down_20D | AVOID | 45 | n/a | n/a | 49.61 (-30.8%) |
| COHR | Coherent Corp. | 262.89 | NoTrade | NO TRADE | 0 | n/a | n/a | 149.85 (-43.0%) |
| MU | Micron Technology, Inc. | 823.03 | NoTrade | NO TRADE | 0 | n/a | n/a | 730.66 (-11.2%) |
| SNDK | Sandisk Corporation | 1214.83 | NoTrade | NO TRADE | 0 | n/a | n/a | 692.45 (-43.0%) |
| NVDA | NVIDIA Corporation | 200.75 | NoTrade | NO TRADE | 0 | n/a | n/a | 177.29 (-11.7%) |
| NBIS | Nebius Group, N.V. | 190.41 | NoTrade | NO TRADE | 0 | n/a | n/a | 126.61 (-33.5%) |
| AMD | Advanced Micro Devices, Inc | 476.15 | NoTrade | NO TRADE | 0 | n/a | n/a | 481.46 (+1.1%) |
| ASML | ASML Holding NV | 1629.00 | NoTrade | NO TRADE | 0 | n/a | n/a | 1453.57 (-10.8%) |
| INTC | Intel Corporation | 90.20 | NoTrade | NO TRADE | 0 | n/a | n/a | 53.24 (-41.0%) |
| SAP | SAP AE | 183.62 | NoTrade | NO TRADE | 0 | n/a | n/a | 198.24 (+8.0%) |
| LITE | Lumentum Holdings | 713.94 | NoTrade | NO TRADE | 0 | n/a | n/a | 409.16 (-42.7%) |
Archive
Earlier end-of-day records.
- 1 August 2026 0 acquire / 1 dispose EOD 2026-08-01
- 31 July 2026 0 acquire / 1 dispose EOD 2026-07-31
- 29 July 2026 0 acquire / 2 dispose EOD 2026-07-29
- 28 July 2026 1 acquire / 1 dispose EOD 2026-07-28
- 25 July 2026 1 acquire / 1 dispose EOD 2026-07-25
- 24 July 2026 0 acquire / 0 dispose EOD 2026-07-24
- 23 July 2026 0 acquire / 0 dispose EOD 2026-07-23
- 22 July 2026 0 acquire / 1 dispose EOD 2026-07-22
- 21 July 2026 0 acquire / 1 dispose EOD 2026-07-21
- 18 July 2026 1 acquire / 1 dispose EOD 2026-07-18
- 17 July 2026 0 acquire / 1 dispose EOD 2026-07-17
- 16 July 2026 0 acquire / 0 dispose EOD 2026-07-16
- 15 July 2026 1 acquire / 0 dispose EOD 2026-07-15
- 14 July 2026 0 acquire / 0 dispose EOD 2026-07-14
- 11 July 2026 0 acquire / 0 dispose EOD 2026-07-11
- 10 July 2026 1 acquire / 0 dispose EOD 2026-07-10
- 9 July 2026 1 acquire / 0 dispose EOD 2026-07-09
- 8 July 2026 1 acquire / 0 dispose EOD 2026-07-08
- 7 July 2026 2 acquire / 0 dispose EOD 2026-07-07
- 4 July 2026 1 acquire / 0 dispose EOD 2026-07-04
- 3 July 2026 0 acquire / 4 dispose DISPOSE: QCOM 90@176.25. Cross-book mirrors of us verdicts. ACQUIRE: none (shared spent via us).
- 1 July 2026 0 acquire / 0 dispose NET-NEW NO ACTION. Mirror us 07-01 on shared shares (execute once). Reconcile QCOM->0 (phantom), NVDA->800. No standalone ustech capital.
- 30 June 2026 2 acquire / 1 dispose ⚠ SAME BOOK AS us (execute ONCE; us 06-30 governs); MIRROR the us adds KLAC+MU (cv93 each); NVDA/GOOGL exits stand down; QCOM reconcile→0. NO net-new ustech capital (shared spent by us).
- 26 June 2026 2 acquire / 2 dispose ⚠ SAME BOOK AS us (execute ONCE); MIRROR the us 06-26 rotation: NVDA trim 25% / GOOGL exit -> MU add; DELL = the combined-book new (us log). No net-new ustech capital.