How Likely Is a 10-Loss Streak?
How Likely Is a 10-Loss Streak?
One of the most important but misunderstood statistics in trading is the expected length of the longest losing streak. A trader with a 60% win rate expects to be right 60% of the time and wrong 40% of the time. But the outcomes are not alternating wins and losses. In any long sequence, clusters of losses emerge. A trader might experience 15 consecutive losses despite a 60% win rate, then 20 consecutive wins, then another brief losing streak.
This article answers: How likely is a 10-loss streak? How long is the worst streak you should expect? Why do winning traders go broke despite positive win rates? The answer reveals why bankroll and risk management matter more than traders initially believe.
Key takeaways
- A 10-loss streak has a probability of (1 - p)^10, where p is the win rate. For a 60% win trader, it is 0.4^10 ≈ 0.0001 (0.01%).
- But this is the probability of 10 losses starting from any one trade. The probability of seeing at least one 10-loss streak in 1,000 trades is much higher (~10%).
- The longest expected streak in N trades grows logarithmically: roughly log_2(N) × log_2(N) trades in the worst case.
- Losing streaks interact with ruin risk: A trader who can survive a 5-loss streak but not a 10-loss streak faces elevated ruin risk if the expected longest streak exceeds 5.
- Streaks are more common than traders expect: Professional traders assume longest streaks of 10–20 losses; undercapitalized traders are shocked when a 5-loss streak wipes them out.
The Basic Probability: A Single 10-Loss Streak
The simplest question: What is the probability of exactly 10 consecutive losses starting from trade 1?
If each trade is independent with win probability p and loss probability (1 - p), then the probability of 10 losses in a row is:
P(10 consecutive losses) = (1 - p)^10
For various win rates:
p = 0.50 (fair): P = 0.5^10 ≈ 0.001 (0.1%)
p = 0.55: P = 0.45^10 ≈ 0.0003 (0.03%)
p = 0.60: P = 0.4^10 ≈ 0.0001 (0.01%)
p = 0.70: P = 0.3^10 ≈ 0.000006 (0.0006%)
These probabilities are small—less than 0.1% for any win rate above 50%. This leads many traders to ignore losing streaks: "The probability is tiny; it will never happen to me."
This reasoning is dangerously flawed.
The Real Question: How Many Opportunities for a Streak?
The key insight is that in a long trading sequence, there are many opportunities for a losing streak to occur. Consider a trader who executes 1,000 trades over two years. A 10-loss streak can start at trade 1, trade 2, trade 3, ..., up to trade 991.
That is approximately 990 different starting points for a potential 10-loss streak. Even though each individual streak is unlikely, the probability of at least one streak occurring is much higher.
The formula for the probability of at least one N-loss streak in M trades is complex, but a useful approximation is:
P(at least one N-loss streak in M trades)
≈ 1 - exp(-M × (1 - p)^N)
For p = 0.60, N = 10, M = 1,000:
P(at least one 10-loss streak) ≈ 1 - exp(-1,000 × 0.4^10)
≈ 1 - exp(-1,000 × 0.0001)
≈ 1 - exp(-0.1)
≈ 1 - 0.905
≈ 0.095 or 9.5%
So a trader with a 60% win rate faces a 9.5% probability of experiencing at least one 10-loss streak in 1,000 trades. Over two years of daily trading (250 trading days per year, 4 trades per day = 1,000 trades), you should expect a 10-loss streak with roughly 10% probability.
Worked Examples: How Many Streaks Are Likely?
Example 1: The 55% Win-Rate Day Trader
- Win rate: p = 0.55 (2.5% edge per trade)
- Time horizon: 2,000 trades (about 2 years of active daily trading)
- Question: What is the probability of seeing a 10-loss streak?
P(at least one 10-loss streak in 2,000 trades)
≈ 1 - exp(-2,000 × 0.45^10)
≈ 1 - exp(-2,000 × 0.0003)
≈ 1 - exp(-0.6)
≈ 1 - 0.549
≈ 0.451 or 45%
Over a two-year trading career, a 55% win-rate trader faces approximately a 45% chance of experiencing a 10-loss streak. This is not a rare event; it is a likely event. A trader who sizes positions assuming no streak longer than 5 losses faces significant drawdown and potential ruin.
Example 2: The 65% Win-Rate System
- Win rate: p = 0.65 (documented over 500 backtested trades)
- Live trading: 1,500 trades (1.5 years)
- Question: Probability of a 6-loss streak?
P(at least one 6-loss streak in 1,500 trades)
≈ 1 - exp(-1,500 × 0.35^6)
≈ 1 - exp(-1,500 × 0.00179)
≈ 1 - exp(-2.69)
≈ 1 - 0.067
≈ 0.933 or 93%
A 65% win-rate trader should expect a 6-loss streak with 93% confidence. If the trader sizes positions such that a 6-loss streak causes a 10% drawdown, the trader is in sound shape. If a 6-loss streak causes a 30% drawdown, the trader is undercapitalized.
The Longest Expected Streak
Rather than asking "What is the probability of a 10-loss streak?" traders should ask "How long is the longest losing streak I should expect?" This is answered by the theory of runs in random sequences.
For a random sequence of N independent trials with success probability p, the expected length of the longest run of failures is approximately:
E[Longest losing streak] ≈ log(N) / log(1 / (1 - p))
For a trader with p = 0.60 (win rate), the loss probability is 0.40. Over N = 1,000 trades:
E[Longest streak] ≈ log(1,000) / log(1 / 0.40)
≈ 6.908 / log(2.5)
≈ 6.908 / 0.916
≈ 7.5 losses
So a 60% win-rate trader should expect the longest losing streak in 1,000 trades to be around 7–8 losses. For 2,000 trades:
E[Longest streak] ≈ log(2,000) / log(2.5) ≈ 7.6 / 0.916 ≈ 8.3 losses
Over two years, expect roughly 8–9 consecutive losses.
The Bankroll Implication: Sizing for Streaks
Knowing the expected longest losing streak is critical for position sizing. If you know the longest expected streak is 8 losses, and you risk R per trade, then you can expect a maximum drawdown (in a worst case, with losses clustered) of approximately 8R.
Example: A trader risks $1,000 per trade, expects a 9-loss streak, and wants the drawdown to be < 15% of a $100,000 account.
9 × $1,000 = $9,000 maximum expected drawdown
$9,000 / $100,000 = 9% drawdown (acceptable)
But if the trader risks $2,000 per trade:
9 × $2,000 = $18,000 maximum drawdown
$18,000 / $100,000 = 18% drawdown (exceeds 15% target)
Thus the trader must limit position size to $1,000 per trade, not $2,000.
This is the connection between losing streak statistics and practical risk management. A trader ignoring the possibility of long streaks will oversized positions and face unexpected drawdowns.
From Edge to Streak Expectation
Why Traders Underestimate Streaks
Reason 1: Recency bias. Traders often look at recent performance (last 100 trades, last 6 months) and extrapolate. A recent stretch of 3 consecutive losses leads them to believe the longest streak is 3–5. But over a 2-year trading career, 8–10 losses is more typical.
Reason 2: Backtesting artifacts. Many backtesting platforms show statistics on a sample of 200–500 trades. The longest streak in a 500-trade backtest is often 5–7. The trader extrapolates: "My worst case is a 7-loss streak." But over 2,000 live trades, a 10-loss streak is likely.
Reason 3: Conflating independent probability with sequential events. A trader thinks: "The probability of 10 losses is 0.0001, so it is unlikely." But this is the probability starting from any one point. Over 1,000 trading points, the aggregate probability is much higher.
Reason 4: Cluster illusion. Traders who experience a long streak often attribute it to regime change, bad luck, or broken strategy. They do not recognize it as a natural statistical feature of any trading sequence. As a result, they do not plan for it.
Streak Distribution: The Longest Is Not the Only One
An important nuance: there are many losing streaks, not just one longest streak. In 1,000 trades, a 60% win-rate trader expects:
- Approximately 9–10 streaks of 4+ consecutive losses
- Approximately 3–4 streaks of 6+ consecutive losses
- Approximately 1 streak of 8+ consecutive losses
This distribution is important because a trader cannot just plan for the longest streak in isolation. The trader must survive the stream of multiple streaks over time.
A position-sizing rule that works for one 8-loss streak might fail if two 8-loss streaks occur within 50 trades of each other. The trader experiences not 8 consecutive losses but 8 losses, then some wins, then 8 more losses. The intermediate wins are smaller than the losses (due to the 60% win rate), and the total drawdown is deeper.
Real-World Example: The High-Win-Rate Loser
A day trader backtests a strategy on 400 trades with a 65% win rate and finds a longest losing streak of 4. The trader is confident and allocates $50,000 to live trading, risking $1,000 per trade.
After 800 live trades (6 months), the trader experiences a 7-loss streak (higher than the backtest maximum). The drawdown is $7,000 (14% of $50,000). The trader survives, but psychologically it is shocking. The trader expected a maximum 4-loss streak based on backtests.
After 1,500 live trades (1 year), the trader hits a 9-loss streak. The account drawdown is $9,000 (18%), but the trader is mentally prepared (having experienced a 7-loss streak already) and executes stops correctly. The account survives.
Had the trader sized positions at $2,000 (2% per trade), the 9-loss streak would have caused an 18% drawdown ($18,000). With a $50,000 account that has also experienced some operational losses and commissions, the account might have hit the psychological breaking point, and the trader might have shut down or taken excessive risk to recover.
The trader's edge (65% win rate) was real. But the bankroll was marginally sized given the longest expected streak. One sizing rule higher, and ruin becomes a real possibility.
The Formula Recap: Quick Reference
For quick calculations:
Probability of exactly N consecutive losses:
P = (1 - p)^N
Probability of at least one N-loss streak in M trades:
P ≈ 1 - exp(-M × (1 - p)^N)
Expected longest losing streak in M trades:
E[max streak] ≈ log(M) / log(1 / (1 - p))
Maximum expected drawdown (worst case):
Drawdown = E[max streak] × Risk per trade
Summary
A 10-loss streak has a low probability in isolation (0.01%–0.1% per starting point) but a much higher probability across a long trading career (5–45% depending on win rate and career length). The expected longest losing streak in 1,000–2,000 trades ranges from 5 to 10 losses, depending on win rate. Traders must size positions to survive the longest expected streak without reaching drawdown or ruin thresholds. Underestimating streak length is a primary cause of unexpected drawdowns and position-sizing errors in professional trading.