Can You Trust These Returns?
A Plain-Language Validation of Our Grid Trading Methodology
The Question We Wanted to Answer
Our Efficient Frontier page shows impressive returns for B- rated crypto portfolios using our Dynamic Grid strategy. The obvious question is: are these returns repeatable, or did we just get lucky with the time period?
To find out, we ran the same methodology across multiple overlapping time periods — like rewinding the clock and asking "would this have worked if we started 3 months earlier? 6 months earlier?" Each time, the grid bot sees the future data for the first time, just like a real user would.
How We Tested It
We used a blind forward test — the gold standard for validating trading strategies. The grid bot never sees future data when making decisions.
Think of it like this: ┌─────────── 30 months of price data ──────────────┐ │ │ │ Test 1: [Learn from Year A] → [Trade Year B] │ │ Test 2: [Learn from Year A'] → [Trade Year B'] │ │ Test 3: [Learn from Year A''] → [Trade Year B'']│ │ │ │ Each test is blind — the bot never peeks ahead. │ └────────────────────────────────────────────────────┘
For each test period, we:
Used the first year of data to set up the grid (price range, levels, spacing)
Rated each crypto based on that year's performance
Selected B- and above rated assets for the portfolio
Built optimized portfolios using three different methods
Let the grid bot trade the second year with zero adjustments
What We Found
Portfolio Returns Across All Test Periods
Dynamic Grid Portfolio Returns (MVO Sharpe allocation)
Every test period produced positive returns. The range was 14% to 91%, showing that while results vary with market conditions, the methodology consistently generated returns.
Does the Dynamic Grid Strategy Add Value?
Dynamic Grid vs Basic Grid (average across all periods)
Yes. The Dynamic Grid strategy added +40.4 percentage points on average. Its ability to trail prices upward and recycle grid levels means it captures more opportunities in rising markets.
Does Our Rating System Actually Pick Better Assets?
We split the rated assets into "top half" (higher-rated) and "bottom half" (lower-rated), then checked which group actually performed better:
The rating system correctly identified better-performing assets in 100% of test periods, with higher-rated assets returning an average of 31 percentage points more than lower-rated ones.
Which Portfolio Method Worked Best?
We tested three ways to divide your money among the selected cryptos:
Calmar CDaR produced the highest average return (61%), but all methods generated positive returns. The difference between methods is less important than the underlying strategy and asset selection.
What Should You Expect Going Forward?
Based on our testing, here is a reasonable range of expectations for a B- rated portfolio:
Important: These numbers come from a limited number of test periods during a specific market environment. They are not guarantees. Real-world results will depend on overall crypto market direction, how well prices stay within grid trading ranges, exchange fees and execution quality, and which assets qualify in your specific period.
Honest Limitations
We believe in transparency. Here is what this study cannot tell you:
We only have ~30 months of data.
This gives us 3 test periods — enough to see a pattern, but not enough for statistical certainty. More data over time will strengthen (or challenge) these findings.
The test periods overlap.
Each period shares some data with the next, so they are not fully independent. Think of it as 3 related experiments, not 3 separate ones.
This was a bull market.
Most of our test period (2023-2026) saw rising crypto prices. Grid trading performs best when prices oscillate within a range. A prolonged bear market would likely produce different results.
Simulations are not reality.
Real trading involves slippage, exchange outages, emotional decisions, and costs that simulations may underestimate.
Past performance ≠ future results.
This is the most important caveat in all of finance, and it applies fully here.
The Bottom Line
Our methodology produced positive returns in every test period we examined. The returns varied — some periods were much better than others — but the core approach of rating assets, selecting the best ones, and using grid trading consistently generated value.
We are not claiming guaranteed returns. What we are claiming is that our methodology has been tested against real historical data using a blind forward-test design, and the results support its continued use as a sound approach to crypto portfolio management.
Related Research
Rating System Validation
100% accuracy for B-rated and above across 60 cryptocurrencies
Rating Effectiveness Study
0.886 correlation with actual returns across 19 windows
Grid Profitability by Exchange
480 backtests: Binance.US 85% vs Coinbase 45%
HRP vs MVO Construction
Comparing portfolio optimization methods for crypto
See It For Yourself
Start with Discovery to find B- rated and above cryptocurrencies, then build your own optimized portfolio using the Efficient Frontier.