| """AlphaForge Metrics Guide - Deep explanations of every metric and how to trade with them. |
| |
| This module teaches you WHAT each number means, WHY it matters, and HOW to act on it. |
| Think of this as your quant trading playbook. |
| """ |
|
|
| METRICS_GUIDE = { |
| "sharpe_ratio": { |
| "name": "Sharpe Ratio", |
| "formula": "(Portfolio Return - Risk-Free Rate) / Portfolio Volatility", |
| "what_it_means": """ |
| The Sharpe Ratio tells you how much EXTRA return you get per unit of risk taken. |
| It's the #1 metric every hedge fund looks at first. |
| |
| - Sharpe = 0.5 → You're getting some reward for risk, but barely worth it |
| - Sharpe = 1.0 → Good. You're getting $1 of return for every $1 of risk |
| - Sharpe = 1.5 → Very good. Professional-grade |
| - Sharpe = 2.0+ → Excellent. Top-quartile hedge fund territory |
| - Sharpe = 3.0+ → Legendary. Renaissance Technologies territory |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If Sharpe < 0.5: STOP. Your strategy is broken. Reduce position sizes by 50% or stop trading. |
| - If Sharpe 0.5-1.0: CAUTION. You're making money but taking too much risk. Tighten stops, reduce leverage. |
| - If Sharpe 1.0-1.5: SOLID. This is your baseline. Keep doing what you're doing. |
| - If Sharpe 1.5-2.0: GREAT. Consider INCREASING position sizes by 20-30%. |
| - If Sharpe > 2.0: EXCEPTIONAL. This is when you go ALL IN. Max out your allocation. |
| |
| TWO SIGMA BENCHMARK: Their flagship fund runs ~1.8 Sharpe |
| CITADEL BENCHMARK: Their equity fund runs ~2.1 Sharpe |
| YOUR GOAL: Consistently above 1.2 Sharpe |
| """, |
| "real_example": """ |
| Example: Your strategy returned 15% with 12% volatility. Risk-free rate is 4%. |
| Sharpe = (15% - 4%) / 12% = 0.92 |
| |
| Verdict: Decent, but not amazing. You're getting $0.92 per $1 of risk. |
| To improve: Either increase return (better signals) or reduce volatility (hedging). |
| """, |
| "tier": "CRITICAL", |
| "weight": 0.25 |
| }, |
|
|
| "sortino_ratio": { |
| "name": "Sortino Ratio", |
| "formula": "(Portfolio Return - Risk-Free Rate) / Downside Volatility", |
| "what_it_means": """ |
| Sortino is like Sharpe, but it ONLY counts bad volatility (losses). |
| Sharpe punishes you for ALL volatility — even the good kind (big upward jumps). |
| Sortino is smarter: it only cares about downside risk. |
| |
| - Sortino < 1.0 → Your losses are too frequent or too large |
| - Sortino 1.0-1.5 → Acceptable. You handle downside okay |
| - Sortino 1.5-2.5 → Strong. You make money while controlling losses |
| - Sortino 2.5+ → Elite. You're a loss-cutting machine |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If Sortino < Sharpe: GOOD. Means your upside volatility is helping you. |
| - If Sortino ≈ Sharpe: NEUTRAL. Your upside and downside are balanced. |
| - If Sortino > Sharpe by 0.5+: EXCELLENT. You have asymmetric upside — the holy grail. |
| - If Sortino drops below 0.8: RAISE STOPS IMMEDIATELY. Your downside is out of control. |
| |
| PRO TIP: Always compare Sortino to Sharpe. The GAP between them tells you |
| if your strategy has positive skew (good) or negative skew (dangerous). |
| """, |
| "real_example": """ |
| Example: Return = 12%, Downside volatility = 5%, Total volatility = 15% |
| Sharpe = (12-4)/15 = 0.53 |
| Sortino = (12-4)/5 = 1.6 |
| |
| Verdict: Sharpe looks mediocre, but Sortino is strong! This means most of your |
| volatility is UPWARD (good jumps). Strategy has positive skew. KEEP TRADING. |
| """, |
| "tier": "CRITICAL", |
| "weight": 0.20 |
| }, |
|
|
| "information_coefficient": { |
| "name": "Information Coefficient (IC)", |
| "formula": "Rank correlation between predicted returns and actual returns", |
| "what_it_means": """ |
| IC measures: "How good are my predictions ACTUALLY?" |
| It's the correlation between what your model PREDICTED and what REALLY happened. |
| |
| - IC = 0.00 → Your model is random. Coin-flip level. |
| - IC = 0.02 → Barely useful. Weak but tradable at scale. |
| - IC = 0.05 → Decent. This is where most quant strategies live. |
| - IC = 0.10 → Strong. You're genuinely predicting better than chance. |
| - IC = 0.15+ → Exceptional. Top-tier signal quality. |
| |
| KEY INSIGHT: IC compounds. If you have 0.05 IC daily, that's MASSIVE over a year. |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If IC < 0.02: YOUR SIGNAL IS DEAD. Stop trading it. Retrain the model. |
| - If IC 0.02-0.05: WORKABLE. Trade smaller sizes. This is your bread-and-butter range. |
| - If IC 0.05-0.08: STRONG. Increase allocation to this signal. Scale up positions. |
| - If IC 0.08-0.12: EXCELLENT. This is your golden goose. Maximize exposure. |
| - If IC turns NEGATIVE: SIGNAL DECAY. Your model is now WRONG. FLIP THE SIGNAL or stop. |
| |
| IC IR (IC / std(IC)): |
| - IC IR < 0.5: Unstable signal. Too noisy. |
| - IC IR 0.5-1.0: Decent stability. |
| - IC IR > 1.0: Rock solid. You can build a career on this. |
| """, |
| "real_example": """ |
| Example: You predicted AAPL would return +2%, MSFT +1%, TSLA -1%. |
| Actual returns: AAPL +3%, MSFT 0%, TSLA -2%. |
| |
| Your ranking was: AAPL > MSFT > TSLA |
| Actual ranking was: AAPL > MSFT > TSLA |
| IC = 1.0 (perfect!) |
| |
| But if actual was: TSLA +5%, AAPL +1%, MSFT -1% |
| Then your ranking was wrong. IC would be negative. |
| |
| This is why IC is rank-based — it cares about ORDER, not exact values. |
| """, |
| "tier": "CRITICAL", |
| "weight": 0.20 |
| }, |
|
|
| "max_drawdown": { |
| "name": "Maximum Drawdown", |
| "formula": "(Peak Value - Trough Value) / Peak Value", |
| "what_it_means": """ |
| Max Drawdown is: "What's the worst losing streak I've ever had?" |
| It's the biggest drop from your highest point to your lowest point. |
| |
| - Max DD < 5% → You're a conservative saint |
| - Max DD 5-10% → Conservative. Sleep-well-at-night level |
| - Max DD 10-20% → Moderate. Normal for active strategies |
| - Max DD 20-30% → Aggressive. You'll have sleepless nights |
| - Max DD 30-50% → Dangerous. Most people panic-sell here |
| - Max DD > 50% → You will blow up. Guaranteed. |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If Max DD exceeds your personal pain threshold: REDUCE RISK by 50%. |
| - If Max DD > 25%: Add hedges (buy puts, short index, reduce beta). |
| - If Max DD < 10% for 6 months: You can INCREASE leverage by 25%. |
| - If drawdown lasts > 3 months: Something is structurally broken. Review strategy. |
| |
| THE PSYCHOLOGY RULE: |
| Most traders can handle 10% drawdown. |
| At 20%, they start questioning the strategy. |
| At 30%, they panic and sell the bottom. |
| At 40%, they quit trading forever. |
| |
| Set your max pain level BEFORE you trade. AlphaForge suggests 15% as default. |
| """, |
| "real_example": """ |
| Example: You started with $1M. Grew to $1.3M (peak). Then dropped to $1.1M (trough). |
| Max DD = ($1.3M - $1.1M) / $1.3M = 15.4% |
| |
| This means at your worst moment, you were down $200K from your best moment. |
| Can you handle seeing $200K evaporate? If not, trade smaller. |
| |
| RECOVERY MATH: After a 15.4% drawdown, you need +18.2% to get back to peak. |
| After a 50% drawdown, you need +100% just to break even. This is why drawdown KILLS. |
| """, |
| "tier": "CRITICAL", |
| "weight": 0.15 |
| }, |
|
|
| "calmar_ratio": { |
| "name": "Calmar Ratio", |
| "formula": "Annualized Return / |Max Drawdown|", |
| "what_it_means": """ |
| Calmar answers: "How much return do I get relative to my worst nightmare?" |
| It compares your yearly gain to your maximum drawdown. |
| |
| - Calmar < 1.0 → Your drawdown is BIGGER than your annual return. Dangerous. |
| - Calmar 1.0-2.0 → Decent. You're getting paid for the risk. |
| - Calmar 2.0-3.0 → Strong. Good risk/reward balance. |
| - Calmar 3.0+ → Excellent. Your returns dwarf your worst losses. |
| |
| Calmar is MORE important than Sharpe for long-term survival. |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If Calmar < 1.0: EMERGENCY. Your strategy will eventually blow up. Cut size by 75%. |
| - If Calmar 1.0-1.5: CAUTION. You're surviving but not thriving. Add diversification. |
| - If Calmar 1.5-2.5: SOLID. This is sustainable long-term. |
| - If Calmar > 3.0: EXCEPTIONAL. You've found a gem. Scale up carefully. |
| |
| THE SURVIVAL RULE: No strategy with Calmar < 1.0 survives 5 years. Period. |
| """, |
| "real_example": """ |
| Example: Annual return = 20%, Max drawdown = 15% |
| Calmar = 20 / 15 = 1.33 |
| |
| Verdict: Okay but not great. You're making 20% but had a 15% nightmare. |
| |
| Better: Annual return = 15%, Max drawdown = 5% |
| Calmar = 15 / 5 = 3.0 |
| |
| Verdict: MUCH better! Lower return but tiny drawdown. You'll compound wealth |
| steadily without panic-selling. THIS is the path to GOAT status. |
| """, |
| "tier": "HIGH", |
| "weight": 0.10 |
| }, |
|
|
| "win_rate": { |
| "name": "Win Rate", |
| "formula": "Number of winning days / Total trading days", |
| "what_it_means": """ |
| Win Rate is: "How often am I right?" |
| But here's the SECRET: Win rate ALONE is meaningless. |
| |
| A 30% win rate can be profitable (if wins are 3x bigger than losses). |
| A 70% win rate can be unprofitable (if losses are 3x bigger than wins). |
| |
| What matters is: Win Rate × Average Win / (1 - Win Rate) × Average Loss |
| This is your EXPECTANCY. If it's > 0, you're profitable. |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If Win Rate < 40% but profitable: You have positive skew. Your wins are HUGE. |
| → Keep going but prepare for long losing streaks mentally. |
| - If Win Rate > 60% but barely profitable: You have negative skew. |
| → DANGER. One big loss wipes out 10 small wins. Tighten stops. |
| - If Win Rate 45-55% with good profits: BALANCED. The sweet spot. |
| - If Win Rate drops 10% from historical average: SIGNAL DECAY. Retrain. |
| |
| THE MENTAL GAME: |
| A 50% win rate means you'll have 5 losses in a row ~3% of the time. |
| You'll have 10 losses in a row ~0.1% of the time (but it WILL happen). |
| Prepare mentally. This is why position sizing > prediction accuracy. |
| """, |
| "real_example": """ |
| Example A: 40% win rate. Average win = $300, average loss = $100 |
| Expectancy = (0.4 × $300) - (0.6 × $100) = $120 - $60 = +$60 per trade |
| Verdict: PROFITABLE despite low win rate! |
| |
| Example B: 65% win rate. Average win = $100, average loss = $250 |
| Expectancy = (0.65 × $100) - (0.35 × $250) = $65 - $87.50 = -$22.50 per trade |
| Verdict: LOSING despite high win rate! This is a classic trap. |
| """, |
| "tier": "MEDIUM", |
| "weight": 0.05 |
| }, |
|
|
| "profit_factor": { |
| "name": "Profit Factor", |
| "formula": "Gross Profits / Gross Losses", |
| "what_it_means": """ |
| Profit Factor is: "For every $1 I lose, how much do I make?" |
| |
| - PF < 1.0 → You're losing money. Period. |
| - PF = 1.0 → You're breaking even before costs. Losing money after costs. |
| - PF 1.0-1.5 → Marginal. Profitable but fragile. |
| - PF 1.5-2.0 → Good. Solid edge. |
| - PF 2.0-3.0 → Strong. Professional grade. |
| - PF 3.0+ → Exceptional. Don't tell anyone your secret. |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If PF < 1.0: STOP IMMEDIATELY. You have no edge. |
| - If PF 1.0-1.3: BARELY ALIVE. One bad streak kills you. Reduce size 50%. |
| - If PF 1.3-1.8: HEALTHY. This is sustainable. |
| - If PF > 2.0: ROBUST. You can survive multiple bad streaks. |
| |
| THE COMPOUNDING RULE: |
| PF of 1.5 with 50% win rate = steady wealth builder |
| PF of 1.2 with 60% win rate = fragile, will eventually break |
| |
| Higher PF > Higher Win Rate (for survival). |
| """, |
| "real_example": """ |
| Example: Over 100 trades, your winners totaled $50,000. Your losers totaled $30,000. |
| Profit Factor = $50,000 / $30,000 = 1.67 |
| |
| Verdict: For every $1 you lost, you made $1.67. Solid. |
| After transaction costs (say $5,000), net PF = $45K/$35K = 1.29 |
| |
| Transaction costs dropped your PF from 1.67 to 1.29! This is why |
| keeping costs low is CRITICAL. Use low-cost brokers, trade less. |
| """, |
| "tier": "HIGH", |
| "weight": 0.05 |
| }, |
|
|
| "alpha": { |
| "name": "Alpha (Jensen's Alpha)", |
| "formula": "Portfolio Return - (Risk-Free Rate + Beta × (Market Return - Risk-Free Rate))", |
| "what_it_means": """ |
| Alpha is: "How much EXTRA return did I generate ABOVE what the market explains?" |
| If Beta = 1 and market goes up 10%, you "should" make 10%. |
| If you made 15%, your Alpha = 5%. That's YOUR skill. |
| |
| - Alpha < 0% → You're underperforming a simple index fund. Just buy SPY. |
| - Alpha 0-2% → Matching the market. No edge. |
| - Alpha 2-5% → Good. You're adding value. |
| - Alpha 5-10% → Strong. Significant edge. |
| - Alpha 10%+ → Elite. You're in the top 1% of managers. |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If Alpha < 0% for 6+ months: JUST BUY THE INDEX. You're wasting time and money. |
| - If Alpha 0-3%: You're a "closet indexer." Either commit to active or go passive. |
| - If Alpha 3-7%: SOLID EDGE. Keep refining but don't over-optimize. |
| - If Alpha > 7%: EXCEPTIONAL. Document your process. This edge won't last forever. |
| |
| THE HUMBLING TRUTH: |
| 80% of active managers have NEGATIVE alpha after fees. |
| S&P 500 has returned ~10%/year for decades. |
| If you can't beat that, just buy SPY and enjoy life. |
| """, |
| "real_example": """ |
| Example: You returned 18%. Market (SPY) returned 12%. Your Beta = 1.2. Risk-free = 4%. |
| Expected return = 4% + 1.2 × (12% - 4%) = 4% + 9.6% = 13.6% |
| Alpha = 18% - 13.6% = 4.4% |
| |
| Verdict: You beat what the market explains by 4.4%. That's YOUR skill. |
| But if you paid 2% in fees, net alpha = 2.4%. Still good, but fees HALVED it. |
| |
| This is why hedge funds charge 2% + 20% — they claim to generate alpha. |
| Most don't. You can beat them with discipline. |
| """, |
| "tier": "HIGH", |
| "weight": 0.05 |
| }, |
|
|
| "beta": { |
| "name": "Beta", |
| "formula": "Covariance(Portfolio, Market) / Variance(Market)", |
| "what_it_means": """ |
| Beta measures: "How much do I move when the market moves?" |
| |
| - Beta = 0.0 → You're market-neutral. Market crashes, you don't care. |
| - Beta = 0.5 → You're conservative. Market up 10%, you up 5%. |
| - Beta = 1.0 → You move exactly with the market (like SPY). |
| - Beta = 1.5 → You're aggressive. Market up 10%, you up 15%. Market down 10%, you down 15%. |
| - Beta = -0.5 → You move opposite to market. Market crashes, you gain. |
| |
| Beta is NOT good or bad. It's a CHOICE based on your goals. |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If you want MARKET EXPOSURE (bullish): Target Beta 0.8-1.2 |
| - If you want LOW RISK (wealth preservation): Target Beta 0.3-0.6 |
| - If you want MARKET NEUTRAL (hedge fund style): Target Beta 0.0 ± 0.1 |
| - If you want INVERSE (betting on crash): Target Beta -0.5 to -1.0 |
| |
| DYNAMIC BETA MANAGEMENT: |
| - In bull markets (regime = 'bull'): Increase beta to 1.2-1.5 to capture upside |
| - In bear markets (regime = 'bear'): Reduce beta to 0.3-0.5 to protect capital |
| - In high volatility: Reduce beta to 0.0-0.3. Survive first, profit second. |
| |
| THE BETA TRAP: |
| Most amateur traders have Beta > 1.5 without knowing it. |
| They think they're skilled, but they're just LEVERAGED to the market. |
| When the market crashes, they get destroyed. |
| """, |
| "real_example": """ |
| Example: Market (SPY) drops 20% in a month. |
| - If your Beta = 0.3: You drop ~6%. Annoying but survivable. |
| - If your Beta = 1.5: You drop ~30%. Devastating. |
| - If your Beta = -0.2: You GAIN ~4%. You're the smart one. |
| |
| This is why Beta management is MORE important than stock picking. |
| Being in the right "market exposure" at the right time beats being right about Apple. |
| """, |
| "tier": "HIGH", |
| "weight": 0.05 |
| }, |
|
|
| "information_ratio": { |
| "name": "Information Ratio", |
| "formula": "(Portfolio Return - Benchmark Return) / Tracking Error", |
| "what_it_means": """ |
| Information Ratio measures: "How much excess return do I get per unit of ACTIVE risk?" |
| It's like Sharpe, but compares you to a BENCHMARK instead of risk-free rate. |
| |
| - IR < 0.0 → You're losing to the benchmark. Just buy the index. |
| - IR 0.0-0.3 → Barely adding value. Probably not worth the effort. |
| - IR 0.3-0.5 → Decent. You're adding some value. |
| - IR 0.5-1.0 → Good. Meaningful active management skill. |
| - IR 1.0+ → Excellent. Top-tier active manager. |
| |
| THE GOLDEN RULE: IR > 0.5 means active management is WORTH the fees/time. |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If IR < 0: STOP ACTIVE TRADING. Buy SPY and save yourself the stress. |
| - If IR 0-0.3: You're "expensive index fund." Reduce active bets, increase passive. |
| - If IR 0.3-0.7: SOLID ACTIVE MANAGEMENT. Keep your current strategy. |
| - If IR > 0.7: EXCEPTIONAL. You have genuine skill. Consider managing money for others. |
| |
| TRACKING ERROR: |
| Low tracking error + positive IR = "closet indexer with slight edge" |
| High tracking error + positive IR = "concentrated bets paying off" |
| High tracking error + negative IR = "you're gambling, not investing" |
| """, |
| "real_example": """ |
| Example: You returned 14%. SPY returned 10%. Tracking error = 8%. |
| IR = (14% - 10%) / 8% = 0.5 |
| |
| Verdict: You're making 4% extra but with 8% of "different" risk. |
| IR = 0.5 means it's worth it, but barely. |
| |
| Better scenario: You returned 16%. SPY returned 10%. Tracking error = 5%. |
| IR = (16% - 10%) / 5% = 1.2 |
| |
| Verdict: MUCH better! More excess return with LESS tracking error. |
| This is the definition of skill. IR > 1.0 is top 5% of all managers. |
| """, |
| "tier": "MEDIUM", |
| "weight": 0.05 |
| }, |
|
|
| "turnover": { |
| "name": "Portfolio Turnover", |
| "formula": "Sum of absolute weight changes / 2", |
| "what_it_means": """ |
| Turnover measures: "How much of my portfolio did I trade?" |
| 100% turnover = You sold everything and bought new stuff. |
| |
| - Turnover < 20%/year → Very low. Buy-and-hold style. |
| - Turnover 20-50%/year → Low. Long-term focused. |
| - Turnover 50-100%/year → Moderate. Typical for active strategies. |
| - Turnover 100-200%/year → High. Trading frequently. |
| - Turnover > 300%/year → Very high. Day-trading territory. |
| |
| HIGH TURNOVER = HIGH COSTS. Every trade costs money. |
| """, |
| "how_it_influences": """ |
| ACTIONABLE RULES: |
| - If turnover > 150% but Sharpe < 1.0: You're trading too much. Reduce rebalancing frequency. |
| - If turnover < 30% but Sharpe > 1.5: Perfect. Low cost, high return. |
| - If transaction costs > 20% of gross profits: Your strategy is paying your broker, not you. |
| |
| THE COST RULE: |
| At 3bps (0.03%) per trade + 150% annual turnover: |
| Total cost = 150% × 0.03% × 2 (buy + sell) = 0.09% per year |
| |
| At 10bps + 300% turnover: |
| Total cost = 300% × 0.10% × 2 = 0.60% per year |
| |
| This doesn't sound like much, but on $1M, that's $6,000/year in friction. |
| Over 20 years at 10% return, that's $40,000+ in lost compounding. |
| """, |
| "real_example": """ |
| Example: Your strategy rebalances weekly. Each rebalance has 40% turnover. |
| Annual turnover = 40% × 52 weeks = 2,080% |
| |
| At 3bps cost: Annual friction = 2,080% × 0.03% × 2 = 1.25% |
| |
| If your gross alpha is 3%, your NET alpha is 3% - 1.25% = 1.75% |
| You lost 42% of your edge to transaction costs! |
| |
| SOLUTION: Rebalance monthly instead of weekly. |
| Annual turnover = 40% × 12 = 480% |
| Annual friction = 480% × 0.03% × 2 = 0.29% |
| Net alpha = 3% - 0.29% = 2.71% |
| |
| You DOUBLED your net alpha just by trading less! This is why patience pays. |
| """, |
| "tier": "MEDIUM", |
| "weight": 0.05 |
| } |
| } |
|
|
|
|
| def get_metric_explanation(metric_key: str) -> dict: |
| """Get full explanation for a metric""" |
| return METRICS_GUIDE.get(metric_key, {}) |
|
|
|
|
| def print_all_metrics(): |
| """Print the complete metrics guide""" |
| print("=" * 80) |
| print("ALPHAFORGE METRICS GUIDE - YOUR PATH TO GOAT STATUS") |
| print("=" * 80) |
| print() |
| |
| for key, info in METRICS_GUIDE.items(): |
| print(f"\n{'='*80}") |
| print(f"📊 {info['name'].upper()} [{info['tier']} | Weight: {info['weight']*100:.0f}%]") |
| print(f"{'='*80}") |
| print(f"\n📝 FORMULA:\n{info['formula']}") |
| print(f"\n💡 WHAT IT MEANS:\n{info['what_it_means']}") |
| print(f"\n🎯 HOW IT INFLUENCES YOUR TRADING:\n{info['how_it_influences']}") |
| print(f"\n📖 REAL EXAMPLE:\n{info['real_example']}") |
| print() |
|
|
|
|
| def get_goat_score(metrics: dict) -> dict: |
| """ |
| Calculate a composite GOAT score based on all metrics. |
| This tells you how close you are to elite quant status. |
| """ |
| score = 0 |
| max_score = 0 |
| breakdown = {} |
| |
| weights = { |
| 'sharpe_ratio': 0.25, |
| 'sortino_ratio': 0.20, |
| 'information_coefficient': 0.20, |
| 'max_drawdown': 0.15, |
| 'calmar_ratio': 0.10, |
| 'win_rate': 0.05, |
| 'profit_factor': 0.05, |
| 'alpha': 0.05, |
| 'beta': 0.05, |
| 'information_ratio': 0.05, |
| 'turnover': 0.05 |
| } |
| |
| |
| sharpe = metrics.get('sharpe_ratio', 0) |
| sharpe_score = min(sharpe / 2.0, 1.0) * 100 |
| score += sharpe_score * weights['sharpe_ratio'] |
| max_score += 100 * weights['sharpe_ratio'] |
| breakdown['sharpe'] = {'value': sharpe, 'score': sharpe_score, 'max': 100} |
| |
| |
| sortino = metrics.get('sortino_ratio', 0) |
| sortino_score = min(sortino / 2.5, 1.0) * 100 |
| score += sortino_score * weights['sortino_ratio'] |
| max_score += 100 * weights['sortino_ratio'] |
| breakdown['sortino'] = {'value': sortino, 'score': sortino_score, 'max': 100} |
| |
| |
| ic = metrics.get('ic_mean', metrics.get('mean_ic', 0)) |
| ic_score = min(ic / 0.10, 1.0) * 100 |
| score += ic_score * weights['information_coefficient'] |
| max_score += 100 * weights['information_coefficient'] |
| breakdown['ic'] = {'value': ic, 'score': ic_score, 'max': 100} |
| |
| |
| mdd = abs(metrics.get('max_drawdown', 0)) |
| mdd_score = max(0, (1 - mdd / 0.30)) * 100 |
| score += mdd_score * weights['max_drawdown'] |
| max_score += 100 * weights['max_drawdown'] |
| breakdown['max_drawdown'] = {'value': mdd, 'score': mdd_score, 'max': 100} |
| |
| |
| calmar = metrics.get('calmar_ratio', 0) |
| calmar_score = min(calmar / 3.0, 1.0) * 100 |
| score += calmar_score * weights['calmar_ratio'] |
| max_score += 100 * weights['calmar_ratio'] |
| breakdown['calmar'] = {'value': calmar, 'score': calmar_score, 'max': 100} |
| |
| total_score = (score / max_score * 100) if max_score > 0 else 0 |
| |
| |
| if total_score >= 85: |
| tier = "LEGENDARY GOAT" |
| emoji = "🐐" |
| elif total_score >= 70: |
| tier = "ELITE QUANT" |
| emoji = "⭐" |
| elif total_score >= 55: |
| tier = "SOLID PRO" |
| emoji = "💪" |
| elif total_score >= 40: |
| tier = "DEVELOPING" |
| emoji = "📈" |
| else: |
| tier = "NEEDS WORK" |
| emoji = "🔧" |
| |
| return { |
| 'total_score': total_score, |
| 'tier': tier, |
| 'emoji': emoji, |
| 'breakdown': breakdown |
| } |
|
|
|
|
| if __name__ == '__main__': |
| print_all_metrics() |
|
|
