This website is for educational purposes only and does not constitute investment advice. Always do your own research and assess your own risk tolerance before making investment decisions.

US End-of-Day Verdict 3 Aug 2026, 16:00:00 GMT-4

US Tech — End-of-Day Verdict

EOD 2026-08-04

Recommendations

1 to acquire · 1 to dispose

TickerNameCloseActionAction 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
CRMSalesForce Inc185.95BUY185.95181.22191.62 (+3.0%)204.86 (+10.2%)198.33 (+6.7%)Salesforce is bought as a beaten-down turnaround, not a momentum chase: the shares fell ~30% this year to a June low and are now up ~12% over both the past week and the past month as the selling exhausts, while the business underneath keeps compounding - Agentforce and Data 360 combined ARR reached 4bn, up more than 200% year on year, and management guides FY27 revenue to -46.2bn against a $63bn FY30 target. The valuation screen calls it fairly valued with passing quality, Wall Street's consensus target sits ~30% higher at, and the entry is at the settled close rather than above it, with risk capped 2.5% below at a level that says quickly whether the bottom is real. OPERATIONAL: BUY conv 76, Buy_Rank #1, ED 0.00, Order_Type Market, 2 consecutive BUY prints strengthening. adj = 76 + val_adj -10 + bias 0 = 66, one above the 65 floor. Quality_Pass TRUE (F-Score 7, rank 15/26, Bargain T, DCF 379.39) - first name in 9 sessions to clear BOTH floor and gate. Sizing 0.5071 x 987,863.41 x 0.20 x 0.75 size-mod = 75,142 -> @185.95 = 75,123.80 (7.6% of book). Entry range 183.59-188.31 (+-0.25 ATR). f=0.30 (L5, calib 0.134 on 7 graded buys). DISCOUNTS ACCEPTED: (i) L1 - FailedBreakdown_Up_20D is this book's most-discounted family, own cell n=1 stop-first 1.0 (TSM 07-07, -6.95% at 1wk), report-only so no numeric bias, respected via 0.75x size not a veto; (ii) stop 181.22 is only 0.50 ATR / 2.54% - inside one day's noise - offset by 4.0:1 nominal R:R and trivial dollar risk (0.19% of book), expectancy +4.72/sh even at a 60% stop-first rate; (iii) close is BELOW EMA200 198.59, so this is a bounce inside a downtrend, and the 12wk 198.33 lands on that 200-day line i.e. mean-reversion to trend, not through it. 12wk < 4wk = systemic finding 2. Countervailing: Morgan Stanley cut to Equal Weight 07-21 with PT 185 - the 185.95 close sits ON it, which is why 0.75x and not full size. L6 clean: thirteen names printed +4.5% to +19.5% today and none is bought; CRM rose +1.05%, a laggard not a gap. · news
TSMTaiwan Semiconductor Manufacturing C406.11SELL ALL406.37n/an/a353.23This closes out the last third of a position that has gone against us, and it is a decision about our signal rather than about TSMC's business - which is excellent. The company just reclaimed a trillion valuation on second-quarter revenue up ~34% to 2bn, record profits, third-quarter guidance of +37% at 65-67% gross margin, and it trades near 20x forward earnings. But our model has had no directional read on the shares for three straight sessions, and its own twelve-week projection still sits 13% below today's price while the holding is 7.3% underwater. If TSMC is worth owning at this multiple it deserves a fresh position with a live signal and a real stop, not the tail of a losing one. OPERATIONAL: 7th arming, unfilled 6x on the executor stall (8th session) - UT-11 re-fire, do not re-litigate on a bounce. BASIS RE-STATED: the L2 stop-break leg is DEAD - close 406.11 vs stop 406.37 = -0.06% / 0.014 ATR, 3rd non-decisive session, and the 12wk has RISEN 3 sessions (341.79->347.18->353.23). The exit now rests on (i) sustained NO TRADE on a held position, its own CLAUDE.md section 9 origination class - conv 0, 'Consensus conflict + low conviction', Pillar_Consensus Indeterminate, Exit_Strategy none, no stop/entry/4wk target, 3 consecutive sessions and 4 of the last 5; and (ii) 12wk 353.23 = -13.02% below the close. Below EMA20 411.11 and EMA50 414.52, above EMA200 364.68, RSI 46.9. NOT a profit-take: PROFIT_ARMED at T12_Progress 114.97 with Gain_ATR -1.77 is the UT-9 arm-on-loss mislabel, inert, ~19th session. G2 suppresses the valuation amplify (DCF 45.15 vs 406.11, ~9.0x broken), 15th session. Proceeds ~41,829.33, realised ~-3,286.73 (-7.29% vs 438.02). 1wk/4wk n/a (blank Target_Price); model band 374.60-425.74. TSMC July monthly revenue on 08-10 is 4 sessions AFTER the exit - not a sale into a catalyst. WITHDRAWAL TRIGGER CARRIED FORWARD UNCHANGED: two consecutive settled closes ABOVE 406.37 with a rising 12wk cancels the exit and re-bases the residual as a fresh position - it missed by today, so the 08-05 run MUST test it before an 8th arming. · news
DELLDell Technologies Inc429.02HOLD390.11n/a504.50553.14Dell is held and its protective stop is lifted after the best session of this drawdown - the shares rose 5.8% and the position is now barely underwater, having been down 11% a week ago. The market is paying up for AI server demand that keeps outrunning supply: AI server revenue grew 757% year on year to 1bn last quarter against a $60bn annual target, the order backlog is at a record, and management has raised full-year revenue and profit guidance. Dell is up 232% this year and outpacing Hewlett Packard Enterprise. With the forward view still ~29% above today's price, the right move is to protect the recovery by raising the risk line, not to sell into it. OPERATIONAL: STOP RAISED 358.88 -> 390.11 (+8.7%), the engine's own non-degenerate Stop_Price at 1.03 ATR below the close (the 07-31/08-01 rows had entry=stop=close and were unusable). Cuts residual risk ~24,318 -> ~13,502. Close 429.02 (+5.83%), -7.75% -> -2.36%, Gain_ATR -0.93 -> -0.28. HOLD conv 46, 0, blank Sell_Rank, NO Exit_Warning; NOT profit-armed (Gain_ATR negative) so no ADR-0012 tick applies. Above EMA20 411.78, EMA50 382.20, far above EMA200 254.96. NO ADD (UT-10: both prior pullback ADDs graded stop_hit; and buying a name up 5.83% on the day is the L6 chase). ADR-0020 TRAIL EROSION tested: Peak Stop 423.45 vs live stop 358.88 = 15.25% erosion with NO size reduction since that peak (the 50% trim was armed 07-29/30 and withdrawn 07-31) - this WOULD have fired a mandatory 30% trim under the us book's 15%/30% config, but ADR-0020 ships default-off and ustech has not enabled it, so REPORT-ONLY, no trim originated; moot at the anchor anyway since the new 390.11 stop is only 7.87% below the peak. Not trimming is also what G1/G3 and the optional_trim counterfactual ('cost', median -16.67% at 4wk, n=17) argue for. · news
QCOMQualcomm Incorporated151.57NO ACTION151.57128.77157.01 (+3.6%)169.69 (+12.0%)86.39 (-43.0%)Qualcomm screens as the cheapest quality name in the book and prints a buy label off a deeply oversold reading, and it is still declined - because the model's own twelve-week projection for it sits 43% BELOW today's price. Buying a stock whose forward view is a 43% loss cannot be justified by a bounce signal. The fundamentals side with the projection rather than the label: fiscal third-quarter revenue fell 4% year on year to 9bn with earnings a shade light, and while automotive hit a record 6bn (+61%), management flagged that Qualcomm's share of the next iPhone will fall well below the 20% it held, with Apple-related revenue declining sharply. The shares are down ~11% this year while the semiconductor index is up ~68% - that is a structural de-rating, not an oversold swing. OPERATIONAL: BUY conv 70, Buy_Rank #2, ED -0.00 clean entry at the close, RSI 32.8. Best valuation in the book - FAIR_VALUE, Bargain T, Quality_Pass TRUE, MoS -0.74 -> val_adj only -4 -> adj 66, ONE ABOVE the 65 floor, and refused by hand anyway. Corroborating: Pillar_Consensus Consolidation_Bearish (Trend 42 / Momentum 37 / Volatility 22), close below EMA20 169.69, EMA50 181.02 AND EMA200 172.32, VWAP distance -10.66%, stop 15.0% / 2.81 ATR away (the engine cannot find a tight invalidation). Ledgered direction:none / optional:true so the counterfactual grader prices what standing down cost. Feeds systemic finding 2 - a bullish label with a 12wk below the entry needs a Forward_View_Conflict flag in core/predictions.py. · news
SNOWSnowflake Inc.307.53NO ACTION304.82262.85329.00 (+7.9%)388.76 (+27.5%)438.23 (+43.8%)Snowflake has the strongest uptrend in the book and is passed over for the ninth session running, on valuation and balance-sheet quality rather than on the chart. The shares are priced well beyond what the current cash flows support - the earnings-power valuation is negative - and the quality screen fails outright, which blocks a brand-new position even when the trend is excellent. After a 4.9% jump the stock is also overbought, so buying here would be paying up into strength rather than entering on a pullback. OPERATIONAL: BUY conv 65, Buy_Rank #3, buy-stop 304.82 BELOW the 307.53 close (ED -0.16), stop 262.85, strongest trend pillar in the book (Trend 94). MoS recomputed at anchor -1.74 -> val_adj -9 -> adj 56, NINE below the 65 floor; independently hard-gated by Quality_Pass FALSE (F-Score 5, EPV -/sh, Quality_Rank 23/26). RSI 74.5 also fails the no-chase bar. Had it been taken: 0.75x -> @304.82 74,376; 12wk 438.23 (+43.8%) exceeds the DCF anchor of 112.27. Ledgered direction:none / optional:true for the counterfactual grader - this name is now the single largest cumulative cost of the quality gate in this book.

Outcome & track record

Accuracy at the time of this record.

11 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.
53 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.

This website is for educational purposes only and does not constitute investment advice. Always do your own research and assess your own risk tolerance before making investment decisions.