Chapter 2
Types of algo trading strategies
six strategies. each one is betting on something different. know what before you run it.
Every algo trading strategy belongs to one of two camps — rule-based or model-driven — and within those camps sit the main strategy types: momentum, mean reversion, market neutral, statistical and quant approaches, ML-based models, and HFT. Each is betting on a different behavior of markets.
Rule-based vs. model-driven
What is the difference between rule-based and model-driven strategies?
The distinction matters when you're evaluating a strategy. Rule-based systems are transparent — you can read the logic, understand when it fires, and reason about when it might fail. Model-driven systems are less legible; their edge comes from statistical patterns the model identified in historical data. Both can work. Both can fail. Knowing which type you're looking at tells you what questions to ask.
Rule-based
Fires when a specific condition is met — price crosses a threshold, volume spikes, a moving average flips. The logic is explicit and readable.
✓ Transparent — you can read the logic
✓ Easier to reason about failure modes
✓ Backtests are more auditable
Model-driven
Makes decisions using a statistical model trained on historical patterns. The edge is learned, not written — which makes it harder to interpret.
✓ Can find non-obvious patterns
✓ Adapts to complex market structure
⚠ Harder to audit — performance can degrade silently
What this means when evaluating a strategy
When you look at a strategy on Wick, the strategy card tells you which type it is — rule-based or model-driven. For rule-based, ask whether the logic makes sense for current conditions. For model-driven, the track record matters more than the logic, because the logic isn't visible. How it has behaved is the evidence you have.
The main strategy types
What are the main types of algo trading strategies?
Momentum
Bets that assets moving strongly in one direction will continue. Buys recent winners, sells recent losers.
examples: trend following · breakout · cross-sectional momentum · dual momentum
Mean reversion
Bets that prices deviate from fair value and snap back. Buys underperformers, sells overperformers.
examples: RSI reversion · Bollinger band strategies · overnight gap fade
Market neutral
Holds offsetting long and short positions to eliminate directional market exposure. Profits from relative performance.
examples: long/short equity · sector-neutral · dollar-neutral
Statistical / quant
Exploits statistical relationships between instruments — pairs trading, arbitrage, factor models.
examples: pairs trading · cointegration · multi-factor models · index arbitrage
ML-based
Uses machine learning models trained on historical data to identify patterns. Harder to interpret than rule-based.
examples: gradient boosting · LSTM · reinforcement learning · ensemble models
HFT Institutional only
Exploits microstructure on millisecond timeframes — order flow, bid-ask spreads, latency arbitrage. The edge depends on infrastructure: servers physically colocated next to exchange matching engines and direct market access most retail traders can't get.
examples: market making · latency arbitrage · order flow trading
Event-driven
Trades around predictable corporate or market events — earnings releases, index rebalances, mergers, dividends. The edge is in being positioned correctly before the event resolves.
examples: post-earnings drift · index rebalance trades · merger arbitrage · dividend capture
Carry
Earns a yield from holding an exposure — selling volatility, capturing dividends, harvesting term-structure premia in futures. Steady income in calm conditions; tail risk when conditions shift.
examples: short volatility · futures roll yield · FX carry · option income strategies
How strategy types perform across market conditions
How do different strategy types perform across market conditions?
every strategy is a bet. the type tells you what it's betting on.
Understanding which market conditions favor each type of algo trading strategy is as important as understanding what each type does. No algo trading strategy works in every environment. This is what makes diversification across strategy types meaningful — they tend to underperform in different conditions.
strategy type vs market condition
MomentumPerforms well in sustained trending markets. Struggles in choppy, range-bound conditions where price reverses frequently before trends establish.
Mean reversionPerforms well in range-bound, sideways markets. Struggles badly in strong trends — buying weakness in a genuine downtrend means catching falling knives.
Market neutralDesigned to be condition-agnostic. In practice, correlation breakdowns during market stress events can cause both legs of the trade to move against you simultaneously.
Statistical / quantSensitive to regime changes — relationships that held statistically may break when the underlying market structure changes. Pairs trading suffers when previously correlated instruments decouple.
ML-basedOften strong in the conditions similar to its training data. Can fail without warning when market conditions shift to something the model hasn't seen.
HFTRelatively condition-agnostic as it exploits microstructure rather than directional moves. Vulnerable to low-volume periods and sudden liquidity withdrawal.
Event-drivenPerforms when scheduled events resolve as expected — earnings, index rebalances, M&A. Vulnerable to deal breaks, surprise announcements, and crowded trades that compress the post-event move.
CarryPerforms in calm, low-volatility regimes where premia are collected steadily. Vulnerable to sudden volatility spikes — the income earned over months can be wiped out in days when conditions shift.
color key
● Momentum
● Mean reversion
● Market neutral
● Statistical / quant
● ML-based
● HFT
● Event-driven
● Carry
What this means for you
When you evaluate a strategy on a leaderboard, look at when it drew down, not just how much. A momentum strategy that lost during a choppy, directionless market is behaving exactly as expected. The same drawdown during a trending market would be more concerning.
This is why a live track record covering multiple market conditions — trending periods, choppy periods, a drawdown — is more meaningful than one from a single favorable stretch. The timing of a drawdown — what market conditions surrounded it — is as informative as its size.
Common questions
Types of algo trading strategies — FAQs
What is momentum trading?
Momentum trading buys assets that have been rising and sells (or shorts) assets that have been falling — betting that recent trends will continue. One explanation: news takes time to be fully priced in, and institutional investors gradually build positions in winners. Moving average crossovers and RSI-based strategies are common implementations.
What is trend following in trading?
Trend following is a momentum-based strategy that bets assets moving strongly in one direction will continue. It buys rising assets and sells falling ones, often using indicators like moving averages or breakouts to identify the trend. It performs well in sustained trending markets and struggles when trends reverse sharply or markets chop.
What is pairs trading?
Pairs trading is a statistical arbitrage strategy that exploits the mean-reverting relationship between two historically correlated assets. When the spread between them widens beyond a threshold, the strategy buys the underperformer and shorts the outperformer, betting the relationship will normalize.
What is a breakout strategy in algo trading?
A breakout strategy identifies when an asset's price moves beyond a defined support or resistance level and trades in the direction of the break. It is typically classified as a momentum strategy — it bets that the breakout signals the start of a new trend rather than a reversion to the prior range.
How do I know which strategy type is right for me?
The right strategy type depends on your risk tolerance and the market conditions you expect. Momentum and trend-following strategies perform well in trending markets but draw down sharply in reversals. Mean reversion strategies perform well in range-bound markets but break down in trends. Most experienced algo traders run strategies across multiple types to reduce dependence on any single market regime.
What is mean reversion in trading?
Mean reversion is the idea that prices tend to return to a long-run average after moving away from it. A mean reversion strategy buys when prices fall unusually far below their average and sells when they rise unusually far above it. Bollinger band strategies and RSI-based systems are common mean reversion approaches.
Which algo trading strategy is best for beginners?
There's no single best strategy — the right fit depends on your risk tolerance, account size, and how you'd respond to different types of drawdowns. Trend-following and momentum strategies tend to be more intuitive to evaluate because their logic is transparent. The most important factor isn't the strategy type — it's finding one with a credible live track record and a max drawdown you could genuinely sit through.
What is market neutral trading?
A market neutral strategy holds both long and short positions simultaneously, structured so the portfolio has minimal exposure to broad market direction. The aim is to profit from the spread between related instruments — not from whether the market goes up or down. "Neutral" is a goal, not a guarantee: correlation breakdowns and tail events can still cause significant losses.
key takeaway
every strategy is trading on something about how markets behave. Knowing what each type is trading on — and which conditions favor it — means you can build a portfolio of strategies that don't all lose at the same time. That's not just diversification. That's an edge in itself.