decision log — ten-agent consensus rating changes
Decision Log
Ten independent agents score each covered name on price history and combine into a single BUY / SELL / HOLD rating. Each rating change is recorded with the prior rating, the new rating, the conviction and the reason drawn from the agent votes. The log is persisted and is not back-filled.
| Change | Reason (agent votes) | |||
|---|---|---|---|---|
| 10-12 21:46Z | AMD | BUY→HOLD | 3% | The directional votes no longer agree (2 buy / 2 sell / 6 hold) — the desk steps back to hold. |
| 10-09 22:16Z | UNH | HOLD→BUY | 8% | Now 3 buy / 1 sell / 6 hold — MACD: MACD histogram positive and expanding — bullish momentum building. |
| 10-09 17:59Z | BAC | HOLD→SELL | 11% | Now 2 buy / 3 sell / 5 hold — Relative Strength: Lagging the market by 12.8% over ~3 months — a laggard. |
| 10-09 16:53Z | HD | HOLD→SELL | 13% | Now 2 buy / 3 sell / 5 hold — Relative Strength: Lagging the market by 17.4% over ~3 months — a laggard. |
| 10-09 16:53Z | JNJ | SELL→HOLD | 1% | The directional votes no longer agree (1 buy / 1 sell / 8 hold) — the desk steps back to hold. |
| 10-09 16:53Z | NFLX | HOLD→SELL | 17% | Now 2 buy / 4 sell / 4 hold — Trend: Price below a falling 50 under 200-day average — established downtrend. |
| 10-09 16:53Z | UNH | SELL→HOLD | 3% | The directional votes no longer agree (2 buy / 1 sell / 7 hold) — the desk steps back to hold. |
| 10-09 16:53Z | AMZN | HOLD→BUY | 36% | Now 5 buy / 0 sell / 5 hold — Trend: Price above a rising 50 over 200-day average — established uptrend. |
| 10-09 08:11Z | HD | SELL→HOLD | 1% | The directional votes no longer agree (3 buy / 2 sell / 5 hold) — the desk steps back to hold. |
| 10-09 08:11Z | BAC | SELL→HOLD | 4% | The directional votes no longer agree (2 buy / 2 sell / 6 hold) — the desk steps back to hold. |
| 10-09 08:11Z | NFLX | SELL→HOLD | 3% | The directional votes no longer agree (2 buy / 2 sell / 6 hold) — the desk steps back to hold. |
| 10-09 08:11Z | WMT | HOLD→BUY | 19% | Now 4 buy / 1 sell / 5 hold — Momentum: Trading 2.9% above its 20-day average — positive momentum. |
| 10-09 08:11Z | UNH | HOLD→SELL | 16% | Now 1 buy / 3 sell / 6 hold — Relative Strength: Lagging the market by 15.2% over ~3 months — a laggard. |
| 10-09 08:11Z | AMZN | BUY→HOLD | 6% | The directional votes no longer agree (1 buy / 0 sell / 9 hold) — the desk steps back to hold. |
| 10-08 17:18Z | DIS | HOLD→BUY | 21% | Now 3 buy / 0 sell / 7 hold — Trend: Price above a rising 50 over 200-day average — established uptrend. |
| 10-08 17:18Z | BAC | HOLD→SELL | 12% | Now 2 buy / 3 sell / 5 hold — Momentum: Trading -5.7% below its 20-day average — negative momentum. |
| 10-08 17:18Z | WMT | SELL→HOLD | 5% | The directional votes no longer agree (2 buy / 2 sell / 6 hold) — the desk steps back to hold. |
| 10-06 09:49Z | DIS | SELL→HOLD | 0% | The directional votes no longer agree (1 buy / 1 sell / 8 hold) — the desk steps back to hold. |
| 10-06 09:49Z | V | HOLD→BUY | 15% | Now 2 buy / 0 sell / 8 hold — Trend: Price above a rising 50 over 200-day average — established uptrend. |
| 10-06 09:49Z | UNH | SELL→HOLD | 3% | The directional votes no longer agree (2 buy / 1 sell / 7 hold) — the desk steps back to hold. |
| 10-06 09:49Z | GOOGL | HOLD→BUY | 19% | Now 4 buy / 1 sell / 5 hold — Trend: Price above a rising 50 over 200-day average — established uptrend. |
| 10-05 15:26Z | V | SELL→HOLD | 6% | The directional votes no longer agree (1 buy / 0 sell / 9 hold) — the desk steps back to hold. |
| 10-05 15:26Z | BAC | SELL→HOLD | 6% | The directional votes no longer agree (2 buy / 2 sell / 6 hold) — the desk steps back to hold. |
| 10-05 15:26Z | TSLA | HOLD→BUY | 10% | Now 3 buy / 1 sell / 6 hold — Momentum: Trading 3.3% above its 20-day average — positive momentum. |
| 10-05 15:26Z | AMZN | HOLD→BUY | 13% | Now 2 buy / 0 sell / 8 hold — MACD: MACD histogram positive and expanding — bullish momentum building. |
| 10-03 04:07Z | TSLA | SELL→HOLD | 2% | The directional votes no longer agree (2 buy / 1 sell / 7 hold) — the desk steps back to hold. |
| 10-03 04:07Z | GOOGL | SELL→HOLD | 3% | The directional votes no longer agree (1 buy / 1 sell / 8 hold) — the desk steps back to hold. |
| 10-03 04:07Z | AMZN | SELL→HOLD | 6% | The directional votes no longer agree (1 buy / 0 sell / 9 hold) — the desk steps back to hold. |
| 10-02 11:41Z | DIS | HOLD→SELL | 15% | Now 1 buy / 3 sell / 6 hold — Momentum: Trading -3.7% below its 20-day average — negative momentum. |
| 10-02 11:41Z | V | BUY→SELL | 21% | Now 1 buy / 4 sell / 5 hold — MACD: MACD histogram negative and expanding — bearish momentum building. |
| 10-02 11:41Z | JNJ | BUY→SELL | 22% | Now 1 buy / 4 sell / 5 hold — Momentum: Trading -3.8% below its 20-day average — negative momentum. |
| 10-02 11:41Z | GOOGL | HOLD→SELL | 16% | Now 1 buy / 3 sell / 6 hold — Relative Strength: Lagging the market by 8.6% over ~3 months — a laggard. |
| 09-29 08:40Z | TSLA | HOLD→SELL | 18% | Now 1 buy / 3 sell / 6 hold — Relative Strength: Lagging the market by 13.3% over ~3 months — a laggard. |
| 09-27 07:32Z | TSLA | BUY→HOLD | 3% | The directional votes no longer agree (2 buy / 1 sell / 7 hold) — the desk steps back to hold. |
| 09-25 12:27Z | DIS | BUY→HOLD | 6% | The directional votes no longer agree (1 buy / 0 sell / 9 hold) — the desk steps back to hold. |
| 09-25 06:42Z | DIS | SELL→BUY | 12% | Now 2 buy / 0 sell / 8 hold — Relative Strength: Outperforming the market by 3.2% over ~3 months — leadership. |
| 09-25 06:42Z | V | HOLD→BUY | 16% | Now 3 buy / 1 sell / 6 hold — Trend: Price above a rising 50 over 200-day average — established uptrend. |
| 09-25 06:42Z | WMT | BUY→SELL | 13% | Now 1 buy / 2 sell / 7 hold — Relative Strength: Lagging the market by 11.6% over ~3 months — a laggard. |
| 09-25 06:42Z | GOOGL | SELL→HOLD | 1% | The directional votes no longer agree (1 buy / 1 sell / 8 hold) — the desk steps back to hold. |
| 09-24 11:29Z | V | BUY→HOLD | 6% | The directional votes no longer agree (2 buy / 3 sell / 5 hold) — the desk steps back to hold. |
How the agent weighting adapts
learning — weights graded from realized outcomesAgent voting weights are set from each agent's graded hit rate on past calls, bounded to 0.50×–1.50×. Each hit rate is shown with its 95% confidence interval and the number of distinct graded calls; excess edge is the average return vs the market benchmark over the same window. A figure with no graded outcomes is shown as “—”. A figure not yet earned from graded outcomes shows as “—”.
trend1.00×hit 43% · 95% CI 31–57%51 gradededge -0.73% vs markettrend: hit 43% of excess-return calls on 51 distinct bar-dates (95% CI 31–57%) — the interval still spans the coin-flip line, equal weight held.
momentum1.00×hit 48% · 95% CI 35–61%51 gradededge -0.37% vs marketmomentum: hit 48% of excess-return calls on 51 distinct bar-dates (95% CI 35–61%) — the interval still spans the coin-flip line, equal weight held.
meanRevert1.00×hit 50% · 95% CI 36–63%51 gradededge -0.67% vs marketmeanRevert: hit 50% of excess-return calls on 51 distinct bar-dates (95% CI 36–63%) — the interval still spans the coin-flip line, equal weight held.
macd1.00×hit 46% · 95% CI 33–60%51 gradededge -0.53% vs marketmacd: hit 46% of excess-return calls on 51 distinct bar-dates (95% CI 33–60%) — the interval still spans the coin-flip line, equal weight held.
breakout1.00×hit 47% · 95% CI 34–60%51 gradededge -0.94% vs marketbreakout: hit 47% of excess-return calls on 51 distinct bar-dates (95% CI 34–60%) — the interval still spans the coin-flip line, equal weight held.
volume1.00×hit 48% · 95% CI 35–61%51 gradededge -0.88% vs marketvolume: hit 48% of excess-return calls on 51 distinct bar-dates (95% CI 35–61%) — the interval still spans the coin-flip line, equal weight held.
volatility1.00×hit —— gradededge — vs marketvolatility: regime commentary only — excluded from reweighting by design; weight pinned at 1.00x.
rsiTrend1.00×hit 45% · 95% CI 33–59%51 gradededge -0.70% vs marketrsiTrend: hit 45% of excess-return calls on 51 distinct bar-dates (95% CI 33–59%) — the interval still spans the coin-flip line, equal weight held.
levels1.00×hit 47% · 95% CI 34–60%51 gradededge +0.16% vs marketlevels: hit 47% of excess-return calls on 51 distinct bar-dates (95% CI 34–60%) — the interval still spans the coin-flip line, equal weight held.
relStrength1.00×hit 48% · 95% CI 35–61%51 gradededge -0.58% vs marketrelStrength: hit 48% of excess-return calls on 51 distinct bar-dates (95% CI 35–61%) — the interval still spans the coin-flip line, equal weight held.
window 180 days · 10160 graded calls · effective n 51
Effective n is the equivalent number of independent observations. Calls are graded over overlapping multi-day windows and cluster on correlated names, so effective n is below the graded-call count; the confidence intervals above are computed on effective n, not on the raw call count.
Method: bounded, disclosed online reweighting — every change is graded against realized outcomes and replayable · Equal weight (every agent at 1.0x) is the permanent unlearned baseline every disclosed weight is read against.
Rows are added only when a consensus rating changes.