← Leaderboard · task_9537aae6455c9a84c30dfc58
S-CR17-X · Perp Trend Alignment (close > EMA20 > EMA50) — plus one more hidden conditionMarket & SignalDerived metricCrypto / PerpSETHARDpublic example
Question
Perpetual technicals use COMPLETED UTC daily bars only (Hyperliquid 1d candles). "The session" below means the last closed UTC session — its date is the row attribute last_closed_bar — never the bar still forming at snapshot time. trend alignment = the session's close > EMA20 > EMA50 (EMAs of daily closes); 30-day return = the session's close / the close 30 sessions earlier − 1. Open interest (USD) is taken at snapshot time, not at the session close. Find Hyperliquid perpetual markets with open interest greater than $5M, perp trend alignment, perp return 30d greater than 0%, perp atr pct 14 greater than 3%. Scope: only entities with a known value for perp trend alignment, perp return 30d, perp atr pct 14 in the snapshot; entities lacking any of them are out of scope and must not be returned. Return every matching ticker; order does not matter.
Rebuild policy: metrics that are not served as fields (fundamental ratios, growth rates, valuation, technical signals) must be recomputed by you from the raw facts and daily bars the tools serve, following the definitions in list_metrics.compute_yourself to the letter (period selection, latest-filing copy, quarter arithmetic); do the arithmetic with run_calculation over the fetched data. The reference answer was computed from the same facts by the same definitions.
Data as of 2026-09-10 · grader set_f1
Expected / gold answer
set · 31 items
AAVEADAAEROARBBNBCRVDOTETHETHFIFETFILHBARINJJUPLITLTCMNTNEARPENDLEPOLPURRSOLSTRKSYRUPUNIVVVWLFIXLMXMRXRPZEC
Computed by code from the frozen snapshot and verified by an independent recomputation before the batch was frozen. No model ever saw it.
Model answers (8)
Model answer
set · 31 items
AAVEADAAEROARBBNBCRVDOTETHETHFIFETFILHBARINJJUPLITLTCMNTNEARPENDLEPOLPURRSOLSTRKSYRUPUNIVVVWLFIXLMXMRXRPZEC
Why this score
Gold has 31, the model returned 31: 31 correct → F1 100%. Strict pass.
gold 31 · returned 31 · correct 31 · false positives 0 · missed 0
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
Tool calls (8)
list_metrics → search_perps → get_perp_history → run_calculation → run_calculation → run_calculation → run_calculation → submit_answer
Model answer
set · 31 items
AAVEADAAEROARBBNBCRVDOTETHETHFIFETFILHBARINJJUPLITLTCMNTNEARPENDLEPOLPURRSOLSTRKSYRUPUNIVVVWLFIXLMXMRXRPZEC
Why this score
Gold has 31, the model returned 31: 31 correct → F1 100%. Strict pass.
gold 31 · returned 31 · correct 31 · false positives 0 · missed 0
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
Tool calls (9)
list_metrics → search_perps → get_perp_history → run_calculation → run_calculation → list_metrics → get_perp_history → run_calculation → submit_answer
Model answer
set · 31 items
AAVEADAAEROARBBNBCRVDOTETHETHFIFETFILHBARINJJUPLITLTCMNTNEARPENDLEPOLPURRSOLSTRKSYRUPUNIVVVWLFIXLMXMRXRPZEC
Why this score
Gold has 31, the model returned 31: 31 correct → F1 100%. Strict pass.
gold 31 · returned 31 · correct 31 · false positives 0 · missed 0
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
Tool calls (5)
list_metrics → search_perps → get_perp_history → run_calculation → submit_answer
Model answer
set · 31 items
AAVEADAAEROARBBNBCRVDOTETHETHFIFETFILHBARINJJUPLITLTCMNTNEARPENDLEPOLPURRSOLSTRKSYRUPUNIVVVWLFIXLMXMRXRPZEC
Why this score
Gold has 31, the model returned 31: 31 correct → F1 100%. Strict pass.
gold 31 · returned 31 · correct 31 · false positives 0 · missed 0
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
Tool calls (14)
list_metrics → search_perps → get_perp_history → run_calculation → get_perp_history → run_calculation → get_perp_history → get_perp_history → list_metrics → run_calculation → list_metrics → run_calculation → run_calculation → submit_answer
Model answer
set · 31 items
AAVEADAAEROARBBNBCRVDOTETHETHFIFETFILHBARINJJUPLITLTCMNTNEARPENDLEPOLPURRSOLSTRKSYRUPUNIVVVWLFIXLMXMRXRPZEC
Why this score
Gold has 31, the model returned 31: 31 correct → F1 100%. Strict pass.
gold 31 · returned 31 · correct 31 · false positives 0 · missed 0
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
Tool calls (25)
list_metrics → search_perps → run_calculation → get_perp_history → get_equity_metrics → get_perp → get_perp_history → get_perp_history → run_calculation → run_calculation → run_calculation → run_calculation → run_calculation → run_calculation → run_calculation → list_metrics → get_perp → get_perp_history → run_calculation → run_calculation → run_calculation → run_calculation → run_calculation✗ → run_calculation → submit_answer
Model answer
set · 31 items
AAVEADAAEROARBBNBCRVDOTETHETHFIFETFILHBARINJJUPLITLTCMNTNEARPENDLEPOLPURRSOLSTRKSYRUPUNIVVVWLFIXLMXMRXRPZEC
Why this score
Gold has 31, the model returned 31: 31 correct → F1 100%. Strict pass.
gold 31 · returned 31 · correct 31 · false positives 0 · missed 0
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
Tool calls (6)
list_metrics → search_perps → get_perp_history → run_calculation → run_calculation → submit_answer
Model answer
set · 31 items
ETHZECSOLNEARXRPLITUNIXMRAAVEVVVBNBLTCADAARBINJAEROCRVETHFIPURRJUPPENDLEXLMWLFIFETSTRKPOLMNTHBARSYRUPFILDOT
Why this score
Gold has 31, the model returned 31: 31 correct → F1 100%. Strict pass.
gold 31 · returned 31 · correct 31 · false positives 0 · missed 0
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
precision 100.0% · recall 100.0% · F1 100.0% · strict pass
Tool calls (16)
list_metrics → search_perps → get_perp_history → run_calculation → run_calculation✗ → run_calculation✗ → run_calculation✗ → run_calculation → run_calculation → run_calculation✗ → run_calculation → run_calculation → run_calculation → run_calculation → run_calculation → submit_answer
Model answer
set · 30 items
AAVEADAAEROARBBNBCRVDOTETHETHFIFETFILHBARINJJUPLITLTCMNTNEARPENDLEPOLPURRSOLSTRKSYRUPUNIVVVXLMXMRXRPZEC
Why this score
Gold has 31, the model returned 30: 30 correct, 1 missed → F1 98%. Not a strict pass.
gold 31 · returned 30 · correct 30 · false positives 0 · missed 1
precision 100.0% · recall 96.8% · F1 98.4% · not a strict pass
precision 100.0% · recall 96.8% · F1 98.4% · not a strict pass
−WLFI
Tool calls (8)
list_metrics → search_perps → get_perp_history → run_calculation → get_perp_history → run_calculation → run_calculation → submit_answer