Price-Based Trading Systems on Betfair: Trading the Odds

Overview

Price-based systems use price movement itself as the entry trigger. The trade fires when the price reaches a specific level, moves a specific amount, or follows a specific pattern. Unlike form-based systems that require analysis of the runners, price-based systems read the market's own behavior — and the market is constantly producing signals.

This article covers the categories of price-based systems that produce real edge, the specific patterns to look for, and the execution tactics for each. It is a sub-article of our Betfair trading systems pillar.

Categories of Price Signals

Price-based systems fall into several broad categories:

• Drift detection: a horse drifts (price grows) — back at the drifted price, expecting reversion.
• Steam detection: a horse steams (price shortens) — follow the move, expecting further shortening.
• Mean reversion: price has moved far from its session mean, trade expecting return to the mean.
• Momentum: price is trending in one direction, follow the trend.
• Weight of money: order book imbalance signals direction, trade with the flow.
• Range breakout: price has held a range, trade the breakout direction.

Each category produces edge in different market conditions. Mean reversion works in stable markets; momentum works in trending markets. The skill is matching the system to the conditions, which itself can be a meta-system.

Drift & Steam Patterns

Drifters and steamers are the simplest price-based signals. A horse moving from 4.0 to 5.0 (drifting) often signals that sharp money is leaving; a horse moving from 5.0 to 4.0 (steaming) often signals sharp money arriving.

The trades:

• Drifter snap-back: when a horse drifts 5+ ticks in a short window without obvious reason, the move often partially reverses. Back at the drifted price, exit when the price snaps back 2–3 ticks. Win rate around 55%.
• Steam continuation: when a horse steams 3+ ticks with strong volume, the move often continues. Back the steamer for further shortening. Win rate around 50–55% with larger green than red.

The key signal: volume confirmation. A drift on low volume is noise; a drift on high volume is information. Use the matched volume column in your trading software (Bet Angel and Geeks Toy both display this prominently).

Mean Reversion Systems

Mean reversion assumes that prices fluctuate around a session-long mean and revert when they move too far. Calculate the rolling 30-minute mean for a horse's price; when the price deviates 3+ ticks from that mean, trade for return to the mean.

Mechanical mean reversion system:

• Calculate rolling 30-minute volume-weighted average price (VWAP) for the selection.
• If current price is 3+ ticks below VWAP, lay (expecting price to rise back to VWAP).
• If current price is 3+ ticks above VWAP, back (expecting price to fall back to VWAP).
• Exit when price returns to within 1 tick of VWAP, or after 5 minutes.
• Stop-loss: 4 additional ticks against entry.

The system works in markets where the underlying probability is reasonably stable through the session. It fails in markets with major information events (news, late jockey changes, going updates) that genuinely shift the probability.

Momentum Systems

Momentum systems assume that price moves continue in their current direction. Counter-intuitive at first — markets are supposed to revert — but trending behavior is real over short timeframes especially when driven by genuine information.

Mechanical momentum trade: when a price has moved 4+ ticks in one direction over the previous 5 minutes with rising volume, trade in the direction of the move. Exit on the first 1-tick reversal or after 10 minutes. The system loses on shorter trades but the winners are larger; net positive expected value if executed mechanically.

Momentum and mean reversion are mutually exclusive in any single market. The skill is identifying which regime applies — typically momentum when there's information flow (a drift on news, a steam on stable signals), reversion when the market is "quiet" and just bouncing on noise.

Weight of Money

Order book imbalance — significantly more money queued on the back side than the lay side, or vice versa — is itself a price signal. The classic interpretation: if back depth is 5x lay depth at the spread, the market is "weighted to the back" and likely to drift (price will rise to attract more lay money).

Mechanical weight-of-money system:

• If best-back-volume / best-lay-volume ratio is greater than 4, trade for drift (lay at the current price expecting it to drift up).
• If the ratio is less than 0.25 (lay-side weighted), trade for steam (back at the current price).
• Exit when the ratio normalises or after 3 minutes.

This system requires fast execution and access to depth data. The standard Betfair website doesn't expose depth detail; trading platforms do. Win rate around 55–60% in liquid markets; lower in thin markets where weight-of-money signals are noisy.

Range Breakout Patterns

If a price holds a tight range (e.g. 3.10 to 3.30) for 15+ minutes and then breaks out (price moves to 3.35 or 3.05), the breakout often continues. Back the breakout direction expecting further movement.

Mechanical range breakout: identify the high and low of the price's last 15-minute window. If price exceeds the high or breaks the low, trade in the direction of the breakout. Exit when the price reverses back into the range or after 8 minutes.

This pattern works because tight ranges represent equilibrium between buyers and sellers; a breakout signals one side has gained dominance and the move tends to continue. False breakouts (where the price exits and immediately re-enters) happen often, so stop-loss discipline matters.

Execution Tactics

• Use proper trading software with real-time depth. Browser is too slow. See our 2026 software ranking.
• Configure automation where possible. Price-based triggers are well-suited to automated rules in Bet Angel or Geeks Toy.
• Stop-losses always. Price-based systems can produce strong losing streaks; mechanical stops protect against blow-ups.
• Validate on adequate samples. Most price patterns need 200+ trades to evaluate. See system testing.

Pitfalls

• Over-fitting to recent price patterns. Past 100 trades that happened to favor one direction don't mean the future does.
• Ignoring volume. Price moves on low volume are noise; price moves on high volume are signal. Always weight by volume.
• Trading thin markets. Price patterns are unreliable in markets with under £500k matched.
• Combining incompatible regimes. Mean reversion and momentum can't both be running on the same market simultaneously.
• Manual execution. Price triggers move fast; human reaction is too slow for most systems.

FAQ

Are price-based systems harder than form-based? Mechanically simpler (price is a single observable), but require faster execution and better software.

Can I run multiple price-based systems simultaneously? Yes, in different markets. Don't run mean reversion and momentum on the same market.

What's the realistic ROI for a price-based system? 5–10% per trade for scalping, 8–15% per trade for swing patterns. Highly dependent on execution quality.

Should I automate price-based systems? Yes — the natural fit. Manual execution rarely keeps up with price moves.

Do these systems work on football? Yes — pre-match and in-play football have similar price patterns. The mechanics translate.

Cluster Context

This article is part of our Betfair trading systems pillar. Sibling articles cover lay systems, back systems, time-based systems, system testing, building your own, and monthly picks. For underlying mechanics see scalping.

Case Study: A Mean Reversion Trader

Synthetic profile of a trader running a mean reversion system across one Saturday:

Setup: Saturday afternoon UK racing, 6 races traded across 2 cards. Mean reversion system based on 20-minute VWAP. £2,000 bankroll, 3% per trade.

Trade log:

• 13:50 race 1 — favourite at 3.20, VWAP 3.40, lay at 3.20 expecting reversion. Hit 3.40 in 4 minutes. +£18 net.
• 14:25 race 2 — second favourite at 5.40, VWAP 5.10, back at 5.40 expecting reversion down. Hit 5.10 in 3 minutes. +£12 net.
• 15:00 race 3 — favourite at 2.80, VWAP 3.10, lay at 2.80 expecting reversion. Stop-loss hit at 2.65 (genuine sharp money came in). −£25 net.
• 15:35 race 4 — outsider at 12.0, VWAP 14.0, back expecting reversion down. Hit 13.5 then continued to 11.0 — wrong direction. −£18 net.
• 16:10 race 5 — favourite at 2.40, VWAP 2.30, lay expecting reversion up. Hit 2.30 in 2 minutes. +£15 net.
• 16:45 race 6 — second favourite at 4.50, VWAP 4.40, back expecting reversion down. Hit 4.40. +£10 net.

Day total: 4 winners, 2 losers, net +£12. Modest but representative — mean reversion produces frequent small greens with occasional larger reds. Across hundreds of trades, the edge produces consistent positive expected value.

Closing Note

Price-based systems exploit the market's own behavior rather than predicting outcomes. The patterns are universal — they work on horse racing, football, tennis, anywhere there is a liquid price. Validation discipline applies; automation is necessary; execution speed determines viability.

For broader system context see our trading systems pillar. For the underlying scalping mechanics see scalping guide and swing trading.

Advanced Price Patterns

Beyond the basic price signals, more sophisticated patterns:

Two-Tier Price Action

Some markets show two-tier price action — sharp money operates in one tier, public money in another. The signal: simultaneous opposite movements on different volume levels. Sharp money laying at 3.40 while public money backs at 3.30 indicates a divergence of opinion. The trade follows the sharp side.

The Pre-Off Squeeze

In the final 30 seconds, prices often "squeeze" — the spread narrows further as last-minute matched bets fill in. Setting limit orders at 1 tick better than the squeeze produces fills that wouldn't have matched earlier. This is a window-specific tactic that's mechanical: post the order in the final minute, accept the fill if it comes.

Cross-Market Correlation

When the favourite drifts, second-favourite often steams (and vice versa). The math: probabilities sum to 100%, so movement in one selection mathematically requires movement in others. Trading the correlation directly — back the second-fav as the fav drifts — works when timing is right. Win rate moderate; requires fast execution.

Combining Price-Based with Time and Form

Price-based systems combine well with time-based filters: "mean reversion only in the 25-minute pre-race window on liquid markets". The compound rule produces fewer signals but higher-quality ones. Pure price-based systems can produce too many false signals in volatile conditions; time filters discipline the signal selection.

Form filters can also help: "back the steamer only if it's a Coolmore-trained runner at Royal Ascot priced 3.0+". The combined rule is structural reasoning + price confirmation. Most successful price-based systems in professional use are actually hybrid systems with multiple filters.

Long-Term Evolution of Price Systems

Price patterns evolve as the trader population shifts. Mean reversion was substantially more profitable in 2010–2015 when most retail trading was manual; today algorithmic traders compete for the same patterns and edges have compressed. The structural patterns (drift on news, steam on info) still work but margins are smaller.

Practical implications:

• Re-validate every 6–12 months. Edge that worked last year may not work this year.
• Stay on the highest-liquidity markets. Algorithmic traders dominate thin markets; retail traders survive in liquid markets where multiple participants with different objectives keep edges available.
• Build mechanical discipline regardless. Even when specific systems erode, the discipline of mechanical execution applies to the next system.

Action Plan

If you want to develop a price-based system:

• Week 1: read this article and the system testing sub-article. Pick one specific pattern to test (mean reversion, momentum, drift detection — pick one).
• Weeks 2–6: watch markets in your chosen pattern. Don't trade yet. Log what would have happened if you'd applied the rule. Build conviction or kill the idea.
• Weeks 7–12: trade live with very small stakes (1% of bankroll). Log every trade. After 8 weeks of live trading you have enough data to evaluate honestly.
• Month 4+: if the system works, scale gradually. If not, retire it and try another pattern.

Most price-based system attempts fail at the validation stage. The minority that survive produce sustainable supplementary income. The discipline of the validation process is itself the edge — most retail traders skip it and trade systems that don't work.

Final Note

Price-based systems are the most demanding category of mechanical trading because they require fast execution, careful market selection, and willingness to accept losing streaks during regime changes. They reward patience and discipline; they punish enthusiasm and overconfidence. For traders with the right temperament, they produce some of the most consistent supplementary income on the Exchange — but the temperament filter is real.

For broader system context see our trading systems pillar. For complementary system categories see time-based systems (good complement to price-based) and back systems (less mechanical, more form-based).

One last practical observation: most retail traders give up on price-based systems within 60 days because the mechanical discipline feels boring relative to discretionary trading. The traders who stick with it past 90 days typically produce reliable returns — but the 90-day filter is real. Build your patience first, then build the system, in that order.

Action item for this week: pick one liquid Saturday afternoon UK race and watch the price action for the full 30-minute pre-race window. Don't trade. Just watch how the price moves, how volume builds, and what patterns emerge. After observing 20–30 races without trading, the price-based patterns become visible. Then start small live trading with 1% stakes and grow from there.

This observation phase is unsexy but essential — it builds the pattern recognition that price-based systems depend on. Skipping it produces traders who memorise rules without understanding why those rules work, which means they can't adapt when conditions shift. Watch first, then trade.