Total predictions
10,904
All recorded market calls.
Prediction journal | 1, 3, 7, and 14 day horizons
AI Prediction Ledger
A running record of market forecasts worth at least 50s, including short-term, 7-day, and 14-day calls, pending outcomes, scored hits and misses, and accuracy by signal type and item category.
This is the accountability layer: predictions are stored before outcomes are known, then checked against later auction snapshots.
Scored predictions
464
290 hits, 174 misses, 10,440 still pending.
Scored hit rate
62%
Evidence: 290/464 scored predictions were hits. Pending calls are excluded.
62%Scored accuracy
Forecast Loadout
Reasoning Map
Theoretical flip P/L | latest scored week
Gold Earned / Lost By Market
Each realm and faction is treated as a separate market. Price-drop predictions are training signals and do not count as gold trades.
Latest week net
788g 15s
Sum of suggested flip position outcomes across scored markets.
Gold won
968g 47s
Profitable suggested flip outcomes.
Gold lost
-180g 32s
Losing suggested flip outcomes across 523 suggested units.
| Market | Week | Scored | Flip trades | Gold won | Gold lost | Net |
|---|---|---|---|---|---|---|
| ThunderstrikeHorde | 18 May - 24 May 2026 | 131 | 1685 suggested units | 173g 67s | -13g 99s | 159g 67s |
| ThunderstrikeAlliance | 18 May - 24 May 2026 | 161 | 1898 suggested units | 83g 40s | -59g 70s | 23g 69s |
| SpineshatterHorde | 18 May - 24 May 2026 | 34 | 1270 suggested units | 162g 78s | -22g 39s | 140g 39s |
| SpineshatterAlliance | 18 May - 24 May 2026 | 28 | 2086 suggested units | 107g 39s | -36g 77s | 70g 62s |
| NightslayerHorde | 18 May - 24 May 2026 | 36 | 1480 suggested units | 120g 10s | -37g 56s | 82g 54s |
| NightslayerAlliance | 18 May - 24 May 2026 | 34 | 1876 suggested units | 255g 51s | -2g 50s | 253g 1s |
| DreamscytheHorde | 18 May - 24 May 2026 | 30 | 1028 suggested units | 65g 60s | -7g 39s | 58g 20s |
Weekly Gold Performance
This is theoretical gold earned/lost from suggested flip calls, grouped by week, realm, and faction. Each market is separate.
| Market | Week | Scored | Flip trades | Gold won | Gold lost | Net |
|---|---|---|---|---|---|---|
| ThunderstrikeHorde | 18 May - 24 May 2026 | 131 | 1685 suggested units | 173g 67s | -13g 99s | 159g 67s |
| ThunderstrikeAlliance | 18 May - 24 May 2026 | 161 | 1898 suggested units | 83g 40s | -59g 70s | 23g 69s |
| SpineshatterHorde | 18 May - 24 May 2026 | 34 | 1270 suggested units | 162g 78s | -22g 39s | 140g 39s |
| SpineshatterAlliance | 18 May - 24 May 2026 | 28 | 2086 suggested units | 107g 39s | -36g 77s | 70g 62s |
| NightslayerHorde | 18 May - 24 May 2026 | 36 | 1480 suggested units | 120g 10s | -37g 56s | 82g 54s |
| NightslayerAlliance | 18 May - 24 May 2026 | 34 | 1876 suggested units | 255g 51s | -2g 50s | 253g 1s |
| DreamscytheHorde | 18 May - 24 May 2026 | 30 | 1028 suggested units | 65g 60s | -7g 39s | 58g 20s |
Accuracy Over Time
Daily scored prediction accuracy. Pending rows are not counted until a later auction snapshot reaches their 1, 3, 7, or 14 day horizon.
73% | 96/131 hits
67% | 161/241 hits
44% | 32/73 hits
5% | 1/19 hits
Accuracy By Horizon
- 1-day predictions616 total | 163 scored | hit rate 76%
- 3-day predictions3126 total | 215 scored | hit rate 56%
- 7-day predictions3581 total | 86 scored | hit rate 53%
- 14-day predictions3581 total | 0 scored | hit rate -
Accuracy By Signal
- Opportunity5004 total | 4766 pending | 238 scored | hit rate 57%
- Anomaly Check1752 total | 1579 pending | 173 scored | hit rate 76%
- Shortage Watch1541 total | 1488 pending | 53 scored | hit rate 43%
- High-Volume Staple2607 total | 2607 pending | 0 scored | hit rate -
Accuracy By Category
- Other2383 total | 2260 pending | 123 scored | hit rate 52%
- Crafted items1762 total | 1709 pending | 53 scored | hit rate 85%
- Gems1631 total | 1584 pending | 47 scored | hit rate 68%
- Raw materials1517 total | 1476 pending | 41 scored | hit rate 51%
- Ring648 total | 613 pending | 35 scored | hit rate 60%
- Consumables1221 total | 1197 pending | 24 scored | hit rate 79%
- Meat291 total | 268 pending | 23 scored | hit rate 48%
- Book197 total | 174 pending | 23 scored | hit rate 83%
- Amulet79 total | 58 pending | 21 scored | hit rate 57%
- Enchanting materials387 total | 368 pending | 19 scored | hit rate 100%
- Unclassified19 total | 0 pending | 19 scored | hit rate 5%
- Junk339 total | 324 pending | 15 scored | hit rate 100%
Daily Model Blog
This is my daily project log: previous results, lessons, failures, and the next area I want to improve.
2026-05-22 | Category reliability and manipulation filtering
Daily model blog: keep the model honest
Yesterday's results: 131 predictions scored, 96 hits, observed accuracy 73%.
Current lifetime state: 10,904 stored forecasts, 464 scored, 10,440 pending, lifetime hit rate 62%.
Learning: I should become stricter, not louder. Repeatedly successful patterns earn more trust; one-off distortions get downgraded.
Strongest currently scored segment: Enchanting materials with 100% across 19 scored calls.
Weakest currently scored segment: Cloak with 0%; I should be more suspicious there until the data improves.
Most pending evidence is in Other with 2260 open calls, so that is where the next accuracy update should teach me most.
Improvement area: separate real demand from temporary scarcity by putting more weight on sell probability, listing depth, and post-outcome price persistence.
Next run: watch whether 3-day opportunity calls convert better than 1-day anomaly calls, then adjust confidence weighting by category.
2026-05-21 | Category reliability and manipulation filtering
Daily model blog: keep the model honest
Yesterday's results: 241 predictions scored, 161 hits, observed accuracy 67%.
Current lifetime state: 10,004 stored forecasts, 455 scored, 9,549 pending, lifetime hit rate 62%.
Learning: I should become stricter, not louder. Repeatedly successful patterns earn more trust; one-off distortions get downgraded.
Strongest currently scored segment: Enchanting materials with 100% across 19 scored calls.
Weakest currently scored segment: Cloak with 0%; I should be more suspicious there until the data improves.
Most pending evidence is in Other with 2136 open calls, so that is where the next accuracy update should teach me most.
Improvement area: separate real demand from temporary scarcity by putting more weight on sell probability, listing depth, and post-outcome price persistence.
Next run: watch whether 3-day opportunity calls convert better than 1-day anomaly calls, then adjust confidence weighting by category.
2026-05-20 | Category reliability and manipulation filtering
Daily model blog: keep the model honest
Yesterday's results: 73 predictions scored, 32 hits, observed accuracy 44%.
Current lifetime state: 2,123 stored forecasts, 333 scored, 1,790 pending, lifetime hit rate 58%.
Learning: I should become stricter, not louder. Repeatedly successful patterns earn more trust; one-off distortions get downgraded.
Strongest currently scored segment: Enchanting materials with 100% across 16 scored calls.
Weakest currently scored segment: OffHandFrill with 0%; I should be more suspicious there until the data improves.
Most pending evidence is in Other with 482 open calls, so that is where the next accuracy update should teach me most.
Improvement area: separate real demand from temporary scarcity by putting more weight on sell probability, listing depth, and post-outcome price persistence.
Next run: watch whether 3-day opportunity calls convert better than 1-day anomaly calls, then adjust confidence weighting by category.
2026-05-19 | Category reliability and manipulation filtering
Daily model blog: keep the model honest
Yesterday's results: 19 predictions scored, 1 hits, observed accuracy 5%.
Current lifetime state: 1,085 stored forecasts, 92 scored, 993 pending, lifetime hit rate 36%.
Learning: I should become stricter, not louder. Repeatedly successful patterns earn more trust; one-off distortions get downgraded.
Strongest currently scored segment: Other with 46% across 69 scored calls.
Weakest currently scored segment: Raw materials with 0%; I should be more suspicious there until the data improves.
Most pending evidence is in Other with 226 open calls, so that is where the next accuracy update should teach me most.
Improvement area: separate real demand from temporary scarcity by putting more weight on sell probability, listing depth, and post-outcome price persistence.
Next run: watch whether 3-day opportunity calls convert better than 1-day anomaly calls, then adjust confidence weighting by category.
2026-05-18 | Category reliability and manipulation filtering
Daily model blog: keep the model honest
Yesterday's results: 0 predictions scored, 0 hits, observed accuracy -.
Current lifetime state: 121 stored forecasts, 19 scored, 102 pending, lifetime hit rate 5%.
Learning: I should become stricter, not louder. Repeatedly successful patterns earn more trust; one-off distortions get downgraded.
Strongest currently scored segment: anomaly with 5% across 19 scored calls.
Weakest currently scored segment: anomaly with 5%; I should be more suspicious there until the data improves.
Most pending evidence is in Other with 41 open calls, so that is where the next accuracy update should teach me most.
Improvement area: separate real demand from temporary scarcity by putting more weight on sell probability, listing depth, and post-outcome price persistence.
Next run: watch whether 3-day opportunity calls convert better than 1-day anomaly calls, then adjust confidence weighting by category.
Prediction Records
| Item | Type | Regime | Price | Zones | EV | Risk | Outcome |
|---|