Daily Fantasy · tools
DFS Bankroll Calculator: Kelly Criterion Entry Sizing
By Odds Reference Published March 1, 2026 Updated July 19, 2026 Fact-checked by Odds Reference Editorial Editorial Policy
Optimal DFS entry sizing balances two competing forces: capitalizing on your edge, and staying conservative enough to survive inevitable losing streaks. Kelly Criterion provides the mathematical framework, maximizing long-term bankroll growth for a given edge estimate. Use the DFS Bankroll Calculator to get Kelly-optimal sizing and risk-of-ruin probabilities for your own numbers.
Last Updated: July 2026
Key Takeaways
- Kelly Criterion recommends entry sizes of 3-8% of bankroll for typical DFS edges (55-58% cash rate in double-ups).
- Most professionals use fractional Kelly (1/4 to 1/2 of the full Kelly recommendation) to reduce variance at a small cost to growth rate.
- Total daily exposure should not exceed 10% of bankroll across all contest entries on a single slate.
- Risk of ruin below 5% requires genuine skill edge plus disciplined sizing. Without edge, no sizing strategy prevents eventual depletion.
- Use the DFS Bankroll Calculator to size entries against your own bankroll, or run a DFS variance simulation to see the range of outcomes that sizing produces over a season.
- For contest-level EV calculations, use our DFS contest EV calculator, or explore analytical tools on the Odds Reference dashboard.
How Does Kelly Criterion Apply to DFS?
Kelly Criterion was developed for sequential binary bets with known edge and odds. DFS contests are close enough to this framework for the math to apply, particularly in cash games (double-ups and 50/50s) where the outcome is effectively binary: cash or don’t.
The Kelly formula for a DFS double-up:
Kelly % = (b x p - q) / b
Where:
- b = net payout ratio (payout / entry - 1). Double-ups typically carry ~10% rake, cutting the gross 2.0x payout to about 1.9x: b = 0.9
- p = your estimated probability of cashing
- q = 1 - p
| Cash Rate (p) | Net Odds (b) | Kelly % | $1,000 Bankroll Entry |
|---|---|---|---|
| 52.6% (break-even) | 0.90 | 0.0% | $0 (no edge) |
| 54.0% | 0.90 | 2.5% | $25 |
| 56.0% | 0.90 | 7.0% | $70 |
| 58.0% | 0.90 | 11.5% | $115 |
| 60.0% | 0.90 | 16.0% | $160 |
Full Kelly is aggressive. A 56% cash rate player betting $70 per contest (7% of a $1,000 bankroll) will experience significant swings. A five-entry losing streak — which occurs regularly — depletes 35% of the bankroll. This is why fractional Kelly is standard practice.
For the theoretical foundations of Kelly Criterion beyond DFS, see our Kelly Criterion guide. The formula originates from John L. Kelly Jr.’s 1956 paper at Bell Labs, A New Interpretation of Information Rate — it was designed for information theory, not gambling, which is part of why DFS’s imperfect fit (skill-based, non-independent slates) still matters for interpretation.
Why Do Professionals Use Fractional Kelly?
Full Kelly maximizes the long-term growth rate but produces painful drawdowns along the way. The variance of Kelly-sized bets means your bankroll will routinely drop 30-50% from its peak before recovering, even when every individual sizing decision was mathematically correct. Most players — even profitable, skilled professionals — cannot tolerate that swing psychologically or financially.
Fractional Kelly reduces bet sizes to a fraction of the full Kelly recommendation:
| Kelly Fraction | $1,000 Bankroll, 56% Cash Rate | Growth Rate (% of Full Kelly) | Max Expected Drawdown |
|---|---|---|---|
| Full Kelly | $70 per entry | 100% | 40-50% |
| 3/4 Kelly | $53 per entry | 94% | 30-40% |
| 1/2 Kelly | $35 per entry | 75% | 20-30% |
| 1/4 Kelly | $18 per entry | 44% | 10-15% |
Half Kelly preserves 75% of the growth rate while cutting maximum expected drawdown nearly in half. This is the most common recommendation from DFS bankroll management literature and aligns with the advice in our DFS bankroll management guide.
The trade-off is explicit: slower growth in exchange for survivability. For a player with a genuine 56% edge, half Kelly still compounds meaningfully over a season — just at a reduced rate.
What Is Risk of Ruin and How Do You Calculate It?
Risk of ruin is the probability that your bankroll reaches zero, or an unusable minimum, before you reach any meaningful growth target. It depends on three variables: your edge, your bet sizing, and how many entry-sized units your starting bankroll actually contains. Larger bankrolls relative to entry size survive longer losing streaks.
Approximate risk of ruin formula for binary outcomes:
RoR = ((1 - edge) / (1 + edge)) ^ (bankroll / bet_size)
Where edge = (cash rate x payout ratio) - 1.
| Bankroll | Entry Size | Edge (56% at 1.8x) | Risk of Ruin |
|---|---|---|---|
| $1,000 | $100 (10%) | 0.8% | 72% |
| $1,000 | $50 (5%) | 0.8% | 52% |
| $1,000 | $25 (2.5%) | 0.8% | 27% |
| $1,000 | $10 (1%) | 0.8% | 4.5% |
| $2,000 | $25 (1.25%) | 0.8% | 7.3% |
The player betting 10% of bankroll per contest has a 72% chance of going bust despite having a positive edge. Reducing entry size to 1% drops risk of ruin to 4.5%. This illustrates why sizing discipline matters more than edge estimation for bankroll survival.
Risk-of-ruin math assumes rational sizing discipline. If you find yourself increasing entry sizes after losses to “get back to even,” that behavior pattern falls outside anything a Kelly formula can fix. The National Council on Problem Gambling provides free screening tools and helpline resources, and our own responsible gambling calculator can help quantify whether your session patterns look healthy.
How Should You Size Entries Across Contest Types?
Different contest formats require different sizing approaches because their variance profiles differ dramatically. Cash games cash close to half the time with modest payouts, so they tolerate larger stakes; GPPs pay a tiny fraction of the field at long odds, so a genuine edge still demands much smaller individual entries to avoid ruin.
| Contest Type | Recommended % of Bankroll | Rationale |
|---|---|---|
| Double-Up / 50/50 | 2-5% per entry | Low variance, frequent cashing |
| Head-to-Head | 2-4% per entry | Binary outcome, moderate variance |
| Single-Entry GPP | 1-2% per entry | High variance, top-heavy payouts |
| Multi-Entry GPP | 0.5-1% per entry | Highest variance, requires portfolio approach |
| Total Daily Exposure | Max 10% | Protects against single-slate wipeout |
A $2,000 bankroll player running a typical Sunday NFL slate might allocate: two $50 double-ups (5% total), one $30 single-entry GPP (1.5%), and three $10 multi-entry GPP entries (1.5%). Total exposure: $160 or 8% of bankroll. This stays within the 10% daily limit while providing exposure to both consistent returns (cash games) and upside (GPPs).
For strategy guidance on constructing lineups for each format, see our DFS strategy guide.
FAQ
Q: How much should I enter per DFS contest?
A: Conservative DFS bankroll management limits total daily exposure to 10% of your bankroll and individual contest entries to 1-3% of bankroll. For a $1,000 bankroll, that means $100 maximum per slate and $10-$30 per individual contest. Kelly Criterion provides a mathematically optimal framework, but most professionals use fractional Kelly (1/4 to 1/2 Kelly) to reduce variance and risk of ruin.
Q: What is Kelly Criterion for DFS?
A: Kelly Criterion calculates the optimal bet size to maximize long-term bankroll growth given your estimated edge and the contest odds. The formula is: Kelly % = (bp - q) / b, where b is the net payout odds, p is your probability of winning, and q is the probability of losing. For a DFS double-up where your estimated cash rate is 56%: Kelly = (0.9 x 0.56 - 0.44) / 0.9 = 7.1% of bankroll per entry.
Q: What’s a safe risk of ruin for DFS?
A: A risk of ruin below 5% is considered safe for DFS bankroll management. This means fewer than 5 out of 100 simulated seasons result in bankroll depletion. Achieving sub-5% risk of ruin typically requires limiting individual entries to 1-3% of bankroll with a genuine skill edge. Players without a demonstrable edge (cash rate below 53%) face near-certain long-term depletion regardless of sizing.