Why a 90% win rate doesn't mean you'll make money
Win rate is the most-quoted number in crypto signals — and the most misleading. Here's what actually determines whether a signal service can make you money over time.
Open any "VIP crypto signals" Telegram group and you'll see the same thing. "95% win rate." "93%." "98% accuracy on our signals." Some of them post screenshots from CryptoSeason or BingX with green PnL bars.
Here's the uncomfortable truth: a 95% win rate tells you almost nothing about whether a signal service will make you money. You can have a 95% win rate and lose money. You can have a 40% win rate and become wealthy. The number people advertise is the one that's easiest to game and the one that hides the most.
Let me explain what actually matters, in math you can verify yourself.
The number that's missing
A trade has two outcomes that matter: how often you win, and how much you win versus how much you lose when you don't.
In trading we call these:
The expectancy of any trading strategy is:
``
Expectancy = (Win Rate × Average Win) − (Loss Rate × Average Loss)
That's the only number that determines whether you make money. Everything else is decoration.
Two services, same win rate, opposite outcomes
Imagine two signal services. Both advertise 80% win rate. You're considering subscribing to one. Here's what their actual track records look like:
Service A — "The Cherry Picker"
Service A loses money on every trade, on average. It "wins often" because they take profits at tiny moves and let losses run wide. Eight trades at +0.4% = +3.2%. Two trades at −2.1% = −4.2%. Net: −1.0% over 10 trades. They lose. You lose.
Service B — "The Patient Trader"
Service B makes ~1% per trade on average. Same win rate, completely different reality.
The win rate alone could not have told you which to subscribe to. It does not contain the information you need.
Why scammy services love high win rates
There's an easy way to manufacture a 95% win rate: take profits at +0.1%.
If your signal calls "BUY at $100, take profit at $100.10, stop loss at $98", you'll hit your tiny take-profit far more often than the wide stop-loss. The math:
You lose money on every trade. But the marketing says 95% win rate. The screenshots all show "TP HIT ✅". You only see the truth after weeks of compounding small losses.
This is what almost every Telegram VIP service does. The win rate is real. The mathematics is brutal.
The four numbers a serious signal service should publish
If a signal provider doesn't tell you all four of these, they're either incompetent or hiding something:
1. Win rate
Useful only as an input to expectancy. On its own, useless.2. Average win
The per-trade upside. Meaningful when paired with average loss.3. Average loss
The per-trade downside. The number nobody wants to publish — because it's the one that exposes whether the service is rigged for high win rate or for actual profitability.4. Profit factor
Calculated as (sum of all wins) / (absolute sum of all losses)`. A profit factor above 1 means you make money over time. Below 1 means you don't, regardless of win rate.A profit factor of 2.0 is a strong system. Above 3.0 is exceptional. Anything advertising 95% win rate with no profit factor disclosed is almost certainly below 1.5 — and probably below 1.0.
What "honest" win rates actually look like
The best discretionary traders in the world — Paul Tudor Jones, Stanley Druckenmiller, the people running Renaissance Technologies' equity book — run between 50% and 65% win rates. They make money because their average win is meaningfully larger than their average loss.
The systematic crypto futures strategies that actually work in published research (academic and institutional) cluster around the same range. A 55% win rate with a 1.5x risk:reward ratio is a profitable strategy. A 95% win rate with a 0.05x risk:reward ratio is a losing strategy that looks profitable.
Anyone selling you a 90%+ win rate is either:
The math doesn't allow for a fourth option.
What we publish on Ezath
This is why our public track record shows all four numbers — not just the headline win rate. Click into any individual signal and you'll see the entry, stop, take-profits, exit reason, and the realized P&L on a 1× leverage basis. We don't get to delete the embarrassing ones because every signal is hash-chained: editing a historical entry breaks the chain visibly.
Our current numbers are open and verifiable. Some weeks the win rate is high; some weeks it's lower. The number we focus on internally is profit factor, because that's the one that determines whether you'd make money following us.
If you want to evaluate any signal service — including ours — start there.
A 30-second test for any service
Before paying a dollar for crypto signals, ask the provider for:
1. A complete trade history (not screenshots — a downloadable file or a public page)
2. The four numbers listed above
3. Their methodology in plain language
If they refuse any of these, walk away. The reason they're not publishing them is the answer.
Why we wrote this
Ezath competes against Telegram VIP services that advertise 95% win rates, run on monthly subscriptions, and quietly delete losing trades when nobody's watching. We can't out-yell them. The only way we win is by making the underlying math obvious to the people we're trying to reach.
So now you know.
If you want to see what an honest track record looks like, it's right here. And if you want to read about how we actually think a crypto futures signal should be structured, the methodology post is here.
— The Ezath team