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 20 Aug 2026, 16:00:00 GMT-4

US — End-of-Day Verdict

EOD 2026-08-21 (us)

Recommendations

1 to acquire · 3 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
KLACKLA Corporation183.99SELL 30%183.99184.00185.38 (+0.8%)182.12 (-1.0%)180.05 (-2.1%)We are cutting 30% of KLA on risk discipline. The protective stop under this position has been walked down from to - a 32% retreat over eighteen sessions - and in all that time the position has never once been reduced. A risk line that keeps moving away while the holding never shrinks is a risk failure, so we take the mandated 30% off and keep. The business case is intact: the CEO says momentum is accelerating into 2027, advanced-packaging revenue is guided up more than 70%, and the street-high target implies 60% upside with no sell ratings - which is exactly why this is a trim and not an exit. Shares are still 5% below on the twelve-week view, so we bank a small gain and reduce the risk. OPS: ADR-0020 Trail Erosion session 18, first fire on the 272.00 watermark, VETO-IMMUNE per US-36 (US-17/US-18/34/US-22/US-30 do not apply). Corroborating not originating: Target_12Week 180.05 is 2.1% below close (L2); 3rd settled close under the pre-cut 193.38. ADR-0012 adds no leg - armed via T12_Progress 102.19 so US-33 blocks tick (c); cv rose (no b); AVOID not NO TRADE (no d); no 2nd Exit_Warning (no a). Valuation tier-bump SUPPRESSED by G2: T12 180.05 / DCF 61.80 = 2.91x. Watermark re-bases to 184.00 ONLY on an observed decrease. · news
GOOGLAlphabet, Inc.- Class A344.82SELL 30%344.82332.73346.45 (+0.5%)342.66 (-0.6%)362.90 (+5.2%)Banking 30% of a position that is up 180% from cost. Alphabet has gone quiet for eight straight sessions - the model has had no active view on it since mid-August - and it still trades below its short-term averages after the AI-leadership shake-up and a capex guide lifted to -205bn with more flagged for 2027. Against that, Berkshire Hathaway added roughly $17bn in the second quarter, making Alphabet its third-largest holding, and the Marvell TPU partnership was just extended. The twelve-week projection is still 5% higher, so this is a profit-take, not a change of mind: we take a third off the table and keep. OPS: ADR-0012 armed via the Gain_ATR limb (33.04, US-26 verified - Unreal_PnL +180.34%), so US-33 does not apply. Fire tick (d) - 8 consecutive NO TRADE prints. Tick (c) does NOT fire: 362.90 > 1.02 x 344.82 = 351.72. Single soft tick -> profit_take_1 = 30% (US-19, first fire, no escalation). US-37 escalation - carded 08-19 and 08-20 unexecuted, this is session 3 and needs an explicit Disposition. MoS -0.68 is above the -3 amplify threshold, no tier bump. Trail held 332.73. · news
OKTAOkta Inc.135.14SELL 30%135.14140.55135.58 (+0.3%)142.52 (+5.5%)151.57 (+12.2%)Okta has now closed below its protective stop for two sessions running and the model has flipped to its strongest sell reading, ranking it the number-one candidate to reduce. The stock is trading under all its short-term averages - a rally being sold into a downtrend - and second-quarter results land on 26 August with a plausible 10% move either way. We take 30% off ahead of that and keep. We are deliberately not cutting deeper: Citizens upgraded the stock this month to Market Outperform with a target and Wells Fargo to Overweight at the Permiso acquisition and the expanded Google Cloud partnership are real product progress, and the model still projects the shares 12% higher over twelve weeks. This is a risk cut on a broken line, not a verdict on the business. OPS: 08-20 pre-committed trigger FIRED on the first limb - 2nd consecutive settled close below 140.55 (134.16 then 135.14, -3.85%). US-18/US-34 satisfied via the INVALIDATION limb, not a lone label flip. Valuation tier-bump DECLINED despite MoS -5.01 and G2 NOT suppressing (T12 151.57 / DCF 107.59 = 1.41x, the only reliable DCF anchor tonight): governing cell RallySell_BearTrend|sell is -6 (n=10, hit_1w.222), forward is alive so L2's negative-12wk limb is unmet, and the news tape is two upgrades. US-19 - execute the first fire at the committed tier. Catalyst-aware hold permits a trim on a decisive stop break. Trail held 140.55. · news
PLTRPalantir Technologies Inc.179.94BUY — QUEUED @157.65157.65149.34179.58 (+13.9%)187.01 (+18.6%)203.25 (+28.9%)Palantir is the strongest business in tonight's buy pool and the only candidate that passes both our conviction bar and our fundamental quality screen - it ranks 6th of 52 names on quality with a near-perfect financial-strength score and 115% revenue growth. Second-quarter revenue grew 93%, US commercial 149%, government 90%, and management raised full-year guidance by roughly. But the shares have run 33% in a month and sit right under resistance near on about 80 times next year's earnings. So we place a limit order at - the model's own entry level, near the volume-weighted average price - rather than buying into the resistance shelf. If it comes back to us we buy with a stop at; if it does not, we have lost nothing. Position size is deliberately halved because the valuation is stretched. OPS: adj = 78 - 10 (val, MoS -12.93 full haircut) + 2 (PullbackBuy_VWAP|buy, the sole positive actionable buy cell) = 70, clear of the 65 floor. Quality gate PASSES - first time in 5 sessions (US-20). NO CHASE: Entry_Distance_ATR -3.11, close is 14.1% above the model entry; limit is the anchor's printed Entry_Price (provenance rule), between VWAP 156.51 and EMA20 161.59. Fill odds are poor - no_chase_queue n=24 at +1.51% [cost], US-21 says these limits structurally do not fill in this regime; queued anyway as the model-faithful action. Single STRONG BUY print (NO TRADE x2 -> WATCH cv59 -> STRONG BUY cv78), low stability confidence. Size: 0.8549 x 200k x 20% = x 0.5 = ->. Est_* are computed off the 179.94 close and overstate a filled-limit outcome. Target_Price 174.28 sits below tonight's close, an artefact of a target struck off a pullback entry. Est_12W 203.25 is 4.36x the 46.62 DCF anchor - widest in the book. expires_on 2026-08-26 (3 sessions). · news
NVDANVIDIA Corporation214.72NO ACTION217.07215.23215.89230.36Nvidia has slipped below its protective stop for a second session, but we are holding into second-quarter results on 26 August rather than trimming three days before the print. The signal itself is not deteriorating - the model's conviction actually rose, the label stayed at hold, and the twelve-week projection is still 7% higher. Bank of America expects revenue of -95bn against $91bn guided and sees guidance going to -108bn, calling it a possible multi-quarter upgrade cycle; 36 analysts rate it a buy and it trades on 25 times forward earnings. The risks are named - up to $5bn of China revenue hostage to policy, and of insider selling in three months. If it closes below a third time, or below today's low, we sell immediately, earnings or not. OPS: qualified for DISPOSE and DISPLACED at slot on dollar rank (vs KLAC GOOGL OKTA) - the skill caps the slate at 3. The 08-20 pre-commitment IS satisfied (-1.08%, 2nd consecutive close), but ADR-0012 is ARMED AND NOT FIRED: armed via T12_Progress 93.21; cv ROSE (no b); Est_12W +7.3% alive (no c); HOLD twice not NO TRADE (no d); no 2nd Exit_Warning (no a). L2 unmet - 12wk is positive. Earnings 08-26, 3 sessions out, +/-6.75% band, horizon stance hold-through; also the marking event for MU/KLAC/NBIS/NET. RE-COMMITTED TRIGGER (US-35, tightened): 3rd consecutive settled close below 217.07 OR any settled close below 214.50 fires SELL 30% immediately, catalyst notwithstanding. Model Stop_Price 204.49 is a loosening, REFUSED under US-29. Avg_Cost still blank - US-26 cannot be hand-verified. · news
LLYEli Lilly & Co1255.40HOLD1168.461268.801281.211384.93Eli Lilly rose 0.9% to and is back within 3% of its 52-week high. The position is far enough in profit to be flagged for potential profit-taking, but nothing has deteriorated - the model's conviction moved by a single point, the label is still hold, and the twelve-week projection is 10% higher. So instead of selling we raise the protective stop beneath it from to, locking in more of the gain while leaving the position to run. OPS: PROFIT_ARMED via T12_Progress 90.65; armed but NOT fired - cv is a 1-pt move (<20, no b), HOLD not NO TRADE (no d), Est_12W +10.3% alive (no c), US-33 blocks (c) on the T12 limb regardless. Model Stop_Price 1168.46 is ABOVE the standing 1159.93 - US-29 permits raises, tightening taken. Avg_Cost still blank - fill it.
MUMicron Technology, Inc.966.78HOLD929.07978.641026.301222.90Micron gave back 0.8% to after opening near and the model has no active view on it for a second session. That is not a reason to sell. It remains the highest-quality name we own on the fundamental screen - ranked 1st of 52 - sits 12.5% above cost with no exit warning, and carries the strongest forward projection in the book at 26% over twelve weeks. We hold with the stop unchanged at. OPS: US-17 - a bare NO TRADE label is never a dispose basis (and US-38 removed the quantified veto that used to argue the opposite direction). Gain_ATR 2.02. Model prints no stop; the 929.07 trail set on 08-20's low stands.
NBISNebius Group, N.V.219.13HOLD213.00220.32241.23253.98Nebius slipped 0.5% to, comfortably above the level we set after the convertible-offering shock. The model has had no active view for six sessions, there is no exit warning, the position is 3.6% in profit and the twelve-week projection is 16% higher. We hold and add nothing - the standing instruction to keep the core but make no additions until the Meta relationship is clarified still governs, including through next week's Nvidia results. OPS: US-17 - bare NO TRADE is not a dispose basis. ADR-0020 erosion -9.63%, below the 15% threshold. Standing overlay override OUTRANKS Event Horizon's add-after-print stance on the 08-26 NVDA read-through; NBIS excluded from any post-print add. Overlay is 48 days stale (US-31) - it restrains here, which is the safe direction, but needs re-confirming.
SPCXSpace Exploration Tech CorpHOLD129.00n/an/an/aSpaceX has no tradable quote in our data, so we assess it on news alone and make no signal-driven call. The second block of insider shares - roughly 319 million from early employees and investors - was released on schedule today, and the stock has drifted back to its listing price on the week. That is absorption, not breakdown: the first release on 6 August more than doubled the public float and the stock rose 6% that day, and Morgan Stanley still reads fundamentals as largely unchanged. We hold on a manual stop at now only about 4% below the quoted level, so this is a live line. The next pressure points are a 1.3 billion-share release around third-quarter results in early November and the 180-day expiry in December. OPS: permanent no_close - muted on signals, no Action/conviction/entry/target synthesized, news-assessed per the edge-case rule. Classification: watch. Auto-unmutes on a non-null Close. · news

Outcome & track record

Accuracy at the time of this record.

154 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 →
-0.148 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.
52 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.