Every +EV bettor eventually asks: should I bet a 0.8% edge? What about 1.5%? There is no single right number, but there is a right way to think about it. Your minimum edge should cover the error in your own fair line, and that error depends heavily on the market.
Why not bet every edge above 0%?
In a perfect world, any positive edge is worth taking. In the real world, your fair line is an estimate. If it is off by 1.5% on average, then many of your 1% "edges" are actually negative. A threshold is a safety margin against your own measurement error.
Think of it this way:
Raise the threshold and you bet less often, but a larger share of your bets are genuinely +EV.
Suggested starting thresholds
| Market | Fair-line reliability | Suggested minimum edge |
|---|---|---|
| NFL/NBA main spreads & totals | Very high (deep sharp markets) | 1.0–1.5% |
| Main moneylines (major leagues) | High | 1.5–2% |
| Alternate spreads & totals | Medium (derived lines) | 2.5–3% |
| Popular player props | Medium | 3–4% |
| Niche props, small leagues | Low | 5%+ |
| Live betting | Low (fast-moving prices) | 4%+ and fresh quotes only |
These are starting points, not laws. Tighten them if your closing line value is weak, loosen them slowly if it is strong.
Factors that should raise your threshold
- Few sharp sources. A fair line from one source is noisier than one built from four.
- Old quotes. If the sharpest price is more than a few minutes old, demand more.
- Long odds. Devig methods disagree most on longshots, so a +600 edge is less certain than a −110 edge. (How to devig odds.)
- Big moves. If the sharp line just jumped, the slow book may follow before you can bet.
- Low limits. If you can only bet $10, a marginal edge is not worth the account attention it draws.
Factors that let you lower it
- Several fresh sharp sources in agreement
- A steam move toward your side (the sharp line is moving your way and your book has not followed). (Steam moves explained.)
- A long history of positive CLV in that market type
Edge vs confidence
Edge tells you how much a bet is worth if the fair line is right. Confidence tells you how likely the fair line is to be right. A 2% edge with high confidence is usually better than a 9% edge with low confidence.
OddsHype shows both on every edge, and lets you set alert filters for each, for example "edge ≥ 2% and confidence ≥ 70", so you only hear about bets worth acting on.
The cost of a higher threshold
Raising your minimum from 1% to 3% might cut your bet volume by more than half. Fewer bets means slower progress and slower proof. The sweet spot is the lowest threshold at which your CLV stays clearly positive. Track it by edge bucket:
| Edge bucket | Bets | Avg CLV |
|---|---|---|
| 1–2% | 420 | +0.4% |
| 2–4% | 310 | +1.6% |
| 4%+ | 95 | +2.9% |
In this example the 1–2% bucket barely beats the close, so raising the floor to 2% would trade a little volume for a lot of reliability.
Frequently asked questions
What EV percentage should I bet?
Start around 1–2% on main lines from major leagues and 3–4% on player props, then adjust based on your closing line value in each market.
Is a 1% edge worth betting?
On deep, well-priced markets with fresh sharp lines, yes. On props, niche leagues or live betting, a 1% edge is usually smaller than the error in the fair line.
Why do some edges disappear when I try to bet?
Because the book corrected its price, or because the edge was based on a stale sharp line. Fast alerts and freshness checks reduce this.