What can go wrong, how Soomario manages risk, and how to think about your own exposure.
No algorithmic trading system wins every trade. No backtest perfectly predicts the future. No amount of sophistication eliminates the possibility of loss. If anyone tells you otherwise, they're lying to you.
What a well-designed system does is manage risk — controlling the size of losses, limiting exposure to any single position, and ensuring that the inevitable losing trades don't wipe out the gains from winning trades. That's what separates sustainable strategies from gambling.
A drawdown is the peak-to-trough decline in your account before a new high is reached. Every strategy experiences drawdowns — the question is how large and how long. Soomario's strategies have backtested max drawdowns ranging from 0.9% (Zones, early data) to 28.1% (AVAX v3 over 3 years). A 25% drawdown means a $10,000 account temporarily drops to $7,500 before recovering.
Most Soomario vault strategies use leverage (2x–8x). Leverage multiplies both gains and losses. A 5% move against you at 5x leverage is a 25% loss on your capital. This is why stop-losses and position sizing are critical — and why the Accumulator ($7/mo, no leverage) is the lowest-risk entry point.
Strategies that work in one market environment may underperform in another. A momentum strategy thrives in trending markets but suffers in choppy, range-bound conditions. Soomario mitigates this by running multiple strategy types (momentum, reversal, DCA, grid farming) — but no single product is immune to regime changes.
Backtested results are calculated on historical data. Live trading faces challenges that backtests don't: slippage (your execution price differs from the signal price), liquidity gaps (especially in crypto during extreme volatility), exchange downtime, and network latency. Soomario publishes both backtested and live results so you can compare.
Every leveraged position has a hard stop-loss. For Max Pain strategies, these are per-coin calibrated. For Contrarian, it's a 6% hard stop. For Farms, breakout detection closes grid positions when price escapes the zone. No position is ever left unprotected.
No single trade risks the entire account. Strategies allocate a fixed percentage of capital per trade (typically 2–5%), so even a string of losses doesn't destroy the account. Farms additionally isolates capital per grid — one grid failing can't consume more than 20% of total capital.
Max Pain runs 20 strategies across 14 coins. Elite runs 3 bots on 3 different assets. Zones monitors 21 coins. Farms scans 50. This diversification means a bad trade on one coin is balanced by performance across others. Backtested inter-strategy correlation for Max Pain is only 0.05 — the strategies behave independently.
Every product page publishes detailed risk metrics: max drawdown, Sharpe ratio (risk-adjusted return), Sortino ratio (downside risk), profit factor (gross gains / gross losses), and win rate. These aren't hidden — they're front and center so you can make an informed decision.
Rule 1: Never invest more than you can afford to lose entirely. Algo trading is not a savings account.
Start small. The Accumulator at $7/mo with no leverage is the lowest-risk way to experience signal-driven trading. If you move to vault strategies, start with a small deposit and increase only after you've seen how the strategy performs over weeks or months.
Understand your product's risk level. Accumulator (low risk, no leverage) is fundamentally different from Elite (medium-high risk, leveraged perps). Make sure the product matches your risk tolerance.
Check the drawdown numbers. If a strategy has a backtested max drawdown of 25%, assume live drawdowns could be 30–40% in the worst case. Can you emotionally and financially handle watching your account drop by that much before recovering? If not, choose a lower-risk product.
Diversify across strategies. Don't put 100% into one vault. Spreading capital across different strategy types (DCA, momentum, zone farming) reduces your exposure to any single approach underperforming.