## How Strategy Choice Affects Variance
Two bettors with identical expected ROI can experience very different variance profiles based on their market choices. Understanding this allows deliberate variance optimisation.
## The Variance Spectrum
**Low-variance betting:**
- Even-money markets (1X2 on strong favourites, AH at −0.25)
- High bet frequency (50+ bets/month)
- Small edge per bet, many bets
Characteristic: smooth equity curve, tight confidence interval that narrows quickly, lower maximum drawdown.
**High-variance betting:**
- Long-odds markets (5.00+, outright winners)
- Low bet frequency (5–10 bets/month)
- Large edge per bet, few bets
Characteristic: volatile equity curve, wide confidence interval that narrows slowly, higher maximum drawdown.
## Calculating Your Portfolio Variance
Portfolio variance = Σ wᵢ² σᵢ² (for independent bets)
Where wᵢ = stake as fraction of bankroll, σᵢ = standard deviation per bet type.
If you mix 70% even-money bets (σ = 1) and 30% 4.00 bets (σ = 2):
Portfolio σ² = (0.70)² × 1² + (0.30)² × 2² = 0.49 + 0.36 = 0.85
Portfolio σ = 0.92 (lower than pure 4.00 betting, slightly lower than pure even-money)
## The Variance-Return Tradeoff
Higher-odds bets with the same EV produce higher expected value per bet (because Kelly stakes are lower but the return when winning is larger) — but also produce much wider variance bands.
For bankroll management purposes, lower-odds markets are generally preferable for the same EV% because:
- Confidence interval narrows faster (validates edge sooner)
- Maximum drawdown is smaller (more psychologically sustainable)
- Ruin risk is lower (smaller stakes as % of bankroll per Kelly)
Create a free account to track your progress and save bookmarks.
