Is a 90% Win Rate Good for a Trader? Why It Can Mislead

Short answer: a 90% win rate is not an edge on its own

Is a 90% win rate good for a trader? Not by itself, and often not at all. A win rate tells you how often trades closed green, but it says nothing about how much was won when they won or lost when they lost, and it says nothing about how many trades produced the figure. A strategy can win nine times out of ten and still bleed money if the one loser is bigger than the nine winners combined.

That is why on PickVesting a figure never appears on its own. Every return is shown with how long the record has been verified and how many closed positions it covers. A win rate without those two numbers next to it is a headline, not evidence.

Why win rate hides the size of the losers

Win rate counts outcomes, not magnitude, so it treats a $5 win and a $5,000 loss as a one-and-one split. That asymmetry is where most high-win-rate strategies live.

Consider two traders with the same 90% win rate over 10 closed trades:

Same win rate. One is up, one is deeply down. The number that separated them was never the win rate, it was the size of the loser. This is the classic shape of strategies that sell a position for a small premium many times and occasionally get run over: selling far out-of-the-money options, averaging down into a falling position, or holding losers while cutting winners.

To judge a record you need to see the losers, not just hear that they were rare. That is also why a record that quietly drops its worst outcomes is worthless. On PickVesting the record covers the whole linked account, every position across stocks, options and crypto, including contracts that expired worthless. If you want the mechanics of how an expired option still counts against the record, see Do Verified Track Records Include Options That Expired Worthless?.

Sample size: 10 trades is a coin flip, not a career

The second thing a win rate cannot tell you on its own is whether the number is real or noise. Nine out of ten is easy to produce by luck over a short run; it is much harder to sustain over hundreds of closed positions and across different market conditions.

Ask three questions of any win rate:

• How many closed positions is it computed from? Ten trades is a sample so small that almost any result is plausible by chance.
• How long has the record been verified? A 30-day streak and a two-year record are not the same claim.
• Which window are you looking at? A number that is strong over 7 days may look very different over 90 days or all time.

PickVesting computes metrics nightly from closed positions, including time-weighted return, win rate, and max drawdown, over 7 days, 30 days, 90 days and all time. Seeing the same win rate hold up across those windows is far more meaningful than one glowing 30-day figure. The method behind each number is laid out in How PickVesting Calculates Return, Win Rate, and Max Drawdown.

The three numbers that actually matter together

A win rate becomes informative only when you read it alongside the numbers that describe magnitude and risk. Look for these as a set:

1. Win rate - how often positions closed green.
2. Return - time-weighted, so deposits and withdrawals do not flatter the figure, shown over multiple windows.
3. Max drawdown - the worst peak-to-trough fall the account actually took.

A high win rate paired with a large max drawdown is the tell for a strategy that wins often and loses badly. A high win rate paired with modest drawdown and a positive return over a long window is a different, more credible picture. Neither predicts the next trade, but the second has survived more.

Max drawdown matters because it is the part of the story a win rate is designed to bury. You cannot feel the one bad trade in a 90% figure. You can see it in the drawdown.

How verification stops the win rate from being staged

A number is only as honest as its source, and a screenshot of a 90% win rate proves nothing. The figure can be cropped, the losing account can be hidden, and the dates can be anything.

PickVesting records are computed from a read-only brokerage connection made through SnapTrade, not from screenshots and not from manual entry. That connection cannot place trades, move money, or change settings, and the trader signs in on their broker's own screen, so PickVesting never sees or stores a brokerage password. A few things follow from that which matter for a win rate specifically:

• You cannot cherry-pick trades. The record covers the whole linked account, so there is no separate hidden account absorbing the losses.
• History cannot be rewritten. Snapshots are append-only; a correction is a new row, so a bad month cannot be deleted later.
• Walking away is visible. A trader can disconnect, but the record keeps everything up to that day and shows the gap after it. Disconnecting never erases history.
• Missing accounts are visible. Every profile states how many brokerage accounts are linked, so an unlinked account is an absence you can see rather than a secret.

That last point is important for win rate claims. If a trader runs their losers in an account they never linked, the linked record will not show them, but it will show you that other accounts exist unverified. The absence is on the page.

Supported brokers shown on the home page include Robinhood, Charles Schwab, Fidelity, Interactive Brokers, Webull and tastytrade, and more, read-only via SnapTrade. You can browse verified records on Discover and read the win rate next to its sample size and drawdown rather than on its own.

So how should you read a 90% win rate?

Treat a 90% win rate as a question, not an answer. The question is: what did the 10% cost, and over how many trades.

A practical checklist:

• Find the number of closed positions. If it is small, the win rate is noise.
• Find the return over the longest available window. A high win rate with a negative long-run return is a losing strategy.
• Find the max drawdown. If it is large, the losers are large.
• Check how long the record has been verified, and whether other accounts are linked.
• Confirm the figure comes from a brokerage connection, not an image.

If all of those line up, a high win rate is one honest data point among several. If any of them is missing, the win rate is decoration.

What this still does not prove

A verified win rate, read with its sample size and drawdown, proves what happened in the linked accounts. It does not prove what happens next.

• A record does not predict the next trade or future results. A 90% win rate over two years can still break on the next position.
• A record proves what happened in the linked accounts only. It says nothing about accounts that are not linked.
• A record does not know your suitability, your position size, or your taxes.
• A record does not vouch for the quality of advice in a chat room, and a high win rate does not make a creator safe to copy.
• Your fill price would differ from the creator's. PickVesting does not place trades for anyone and does not offer copy trading.

If you want the full set of caveats, read the disclosures.

The point of a verified record is not to crown the highest win rate. It is to let you see the losers, count the sample, and judge the number in context instead of taking a screenshot on faith.

PickVesting is not an investment adviser and does not provide personalized investment advice. Past performance is not a prediction of future results.