Picture two trades that close within the same hour, each one hitting its stop, the price level set in advance where a position closes automatically to cap the loss. Both lose exactly 1R, trader shorthand for one unit of the risk planned before entry, and yet one of them was handled perfectly while the other was a mistake from the first click.
A losing trade followed your written plan and still hit its stop, the normal cost of any strategy that doesn’t win every time. A bad trade broke the plan, with a moved stop or a position bigger than the rules allow, and it stays bad even if it makes money. In Xcelerate Trade terms, process gets graded before profit.
What counts as a losing trade?
A losing trade is a trade you were right to take that simply didn’t work out. The setup, meaning the specific chart conditions your plan requires before entry, was there, and the position size was calculated from the distance to the stop so that a hit stop would cost exactly the planned risk. Price then went the other way and closed you out at the level you picked in advance, and nothing in that sequence needs fixing. I think of it as rent, the regular cost a strategy pays to stay in business.
A small bakery makes a useful comparison. Some bread goes stale every week, and the owner keeps the baker anyway, because a few wasted loaves are the price of having fresh ones on the shelf when customers walk in. A trader with a 45% win rate, meaning 45 of every 100 trades close in profit, is in the same position, since more than half of the trades will lose by design.
The arithmetic that makes losses ordinary
Suppose a strategy wins 45% of the time, its winners average 2R and its losers cost 1R. Each trade is then worth about +0.35R on average, a figure traders call expectancy, and a positive expectancy held over a long series is what traders mean by an edge.
The uncomfortable part shows up in the sequence of results. A quick simulation I ran of that exact profile, 100 trades repeated 200,000 times, found a streak of seven or more consecutive losses in roughly half of the runs. Seven stops in a row can come from a strategy that is healthy on paper, so anyone who labels each of them a bad trade will usually start rewriting rules in the middle of a normal rough patch.
A stop that gets hit is the plan doing its job
A stop out, the moment the stop closes a position, can feel like a small personal insult. Read properly, it is information arriving on schedule, because the market reached the price where the idea was proven wrong and the plan took you out at a cost you had already accepted. A trade with no hard exit has no defined risk, so it can’t be measured in R or graded fairly in a later review.
What makes a trade bad, even when it wins
A bad trade is any trade where you broke your own rules, and the profit or loss at the end has no say in that label. Entering without the setup your plan describes counts, and so does sliding the stop further away once price comes close to it. Either one makes the trade bad before the market has moved a single tick.
The test I rely on is plain. Someone who has never met me should be able to read the written plan, look at the chart at the moment of entry and confirm that the trade was valid, and if that’s impossible, the trade was bad whatever it paid.
The bad trade that wins is the expensive one
Take a slow Friday afternoon as an example. A trader is bored after watching an index drift all day, jumps in on a hunch at double the usual size, and places the stop somewhere that merely looks sensible. The position runs to +3R in twenty minutes.
The damage comes in the following weeks, when the brain files that afternoon under skill and goes looking for a repeat. Trades taken on a hunch, without a stop tied to any invalidation, can lose several R at once, and a handful of those can erase more than a month of disciplined losing trades.
For that reason a winning bad trade deserves more attention in review than a losing good one. The losing good trade is already priced into the strategy and teaches nothing new, while the winning bad trade quietly trains you to repeat the mistake, usually at a worse moment and a bigger size.
Where the process usually breaks
Before entry, the typical break is a missing or vague setup, the feeling that a chart looks like it could go. During the trade, it is the stop that drifts, or a target that shrinks because a small open profit feels safer than the plan.
After a loss, the risk moves to the next trade. Revenge trading, a fast and oversized trade taken to win back a loss, is the usual form, and fatigue and late sessions make it more likely.
Why do results fool us so easily?
Outcome bias, the tendency to judge a decision by its result instead of by the information available when it was made, is the main reason results fool traders. Psychologists Jonathan Baron and John Hershey documented it in 1988 in the Journal of Personality and Social Psychology. Across five studies, participants rated the same decision as better reasoned, and its author as more competent, when the outcome happened to be favorable, and those who said outcomes shouldn’t count did it anyway.
The effect has held up over the decades. A pre-registered replication with 692 participants recruited online found the same pattern, again including people who stated that outcomes shouldn’t matter when judging a decision. Annie Duke, a former professional poker player, brought the poker term for this habit, resulting, to a much wider audience in her 2018 book Thinking in Bets.
What the base rates add
A base rate is the share of people in a large group who end up with a given result, and for retail trading those shares point in the same direction. In 2018 the European Securities and Markets Authority restricted CFDs for retail clients, contracts for difference being leveraged bets on price moves that never involve owning the underlying asset. It said national regulators had found that 74% to 89% of retail accounts typically lost money, with average losses per client between €1,600 and €29,000, even though every one of those accounts had winning trades along the way.
Short samples mislead even more. A study by Fernando Chague, Rodrigo De-Losso and Bruno Giovannetti, built on data from Brazil’s securities regulator, followed every individual who began day trading mini Ibovespa futures, contracts on Brazil’s main stock index, between 2013 and 2015, close to 20,000 people. Among those who day traded on a single day only, almost 30% made a net profit, while among those who kept going for more than 300 days, 97% lost money.
Sorting every trade into one of four boxes
Separating bad trades from losing trades works best on a grid with two axes, process and outcome, which produces four possible boxes. Only the two boxes where the plan was broken call for changes, and one of them usually looks like a success. This is the grid I’d use in any weekly review.
| Did the trade follow the plan? | Result | What it is | What the review does |
| Yes | Win | Earned win | Logs it and changes nothing |
| Yes | Loss | Losing trade | Logs it and changes nothing |
| No | Win | Lucky bad trade | Treats it as an error and looks for the trigger |
| No | Loss | Bad trade | Treats it as an error and looks for the trigger |
The two top rows get identical treatment. A planned trade that loses and a planned trade that wins are the same decision caught on different rolls of the dice, while the bottom rows are where the learning happens, and the lucky bad trade deserves the longest review notes of all.
Grading the decision with the result covered
The most useful review habit I know is also the simplest. Before looking at anything else, hide the profit and loss column and mark each trade with a yes or no to a single question, did it follow the plan. Only then uncover the numbers, which feels a bit childish but denies outcome bias the information it needs to work.
The plan adherence column
Plan adherence is the share of trades that followed the written plan, and a column in the trading journal, the log where each trade is recorded with its reasons and result, can track it with a simple Y or N. If adherence sits above roughly 90% and results are still poor, the strategy needs work or the market has changed character. If adherence is down at 70%, the results describe a mix of strategy and improvisation, so they can’t be used to judge the strategy at all.
How does a review routine keep the labels honest?
A trade review routine keeps the bad trade and losing trade labels honest by fixing the criteria before entry, so the result can’t rewrite them afterwards. Xcelerate.Trade organizes its Academy into learning paths, including separate ones on risk and on psychology, and the review routine is where those two subjects meet in practice.
Writing the invalidation before the entry
The invalidation is the price level or event that proves the trade idea wrong, and it belongs on paper before the order goes in, in a sentence such as this idea is wrong if price closes below X. The stop sits just beyond that level, and the position size follows from the distance between entry and stop, so a loss at the stop equals the planned risk, which I keep at 1% of the account or less. A trade whose invalidation can’t be written down simply doesn’t get taken, and in my view this single rule removes more bad trades than any other, because most bad trades never had an invalidation in the first place.
Measuring results in R
Measured in R, a loss at the stop is minus 1R and a winner that travels twice the stop distance is plus 2R. This strips the emotional weight out of currency amounts and makes trades of different sizes comparable. A loss of 1R is a planned loss, while a loss of 2.7R on a trade meant to risk 1R is a warning that the stop either moved or slipped badly, and the journal makes that visible at a glance.
When a run of losing trades does mean something
Strategies do stop working, and a string of correctly executed losing trades is how that shows up. The useful question is whether the current run is larger than the strategy’s own history says to expect.
A drawdown is the fall in an account from its last peak, and I compare the current one with the worst streak from backtesting, which applies the rules to historical data, and from replay mode, which plays past price action back bar by bar. The Practice area on Xcelerate.Trade offers replay mode alongside demo trading on virtual money. I don’t change rules on fewer than about 100 logged trades unless something structural has shifted, such as volatility doubling or a trending market turning into a range for weeks. Recovery arithmetic belongs in that comparison too, because a 20% drawdown needs a 25% gain to climb out of, while a 50% drawdown needs 100%.
Following other traders without borrowing their excuses
Copy trading, which means mirroring another trader’s positions in your own account, gets graded on process too, with two processes to check, the provider’s and yours. A signal, a published trade idea with a defined entry and stop, that hits its predefined stop is a losing trade for the provider. If you entered late at a worse price or deleted the stop because the provider surely knew what they were doing, your copy became a bad trade, even when the original was fine.
Xcelerate Trade is an online trading education platform, published in English, Romanian, Spanish and French, that pairs a structured Academy and a Practice area for demo and replay with a Marketplace for copy trading and signals across crypto, forex and futures. For copied trades, the useful part of any marketplace is whatever lets you check the process before entry, which a running balance on its own can’t show.
When I look through the Copy Trading Signals on the platform, I read them the way I’d read my own trades, checking first that the entry and the stop are stated before the position opens. The second check is whether the size suits my account, which may look nothing like the provider’s. The marketplace on Xcelerate Trade presents its signals as entries with defined risk, and defined risk is precisely what allows a copied trade that fails to be graded as an ordinary losing trade.
Judging a provider over a long sample
A provider’s last three trades are far too few to separate skill from luck. I look for track records long enough to include a proper losing streak, and I pay more attention to how a provider loses than to how they win. Consistent losses of about 1R are reassuring, while a tidy win rate with the occasional giant loss usually points to stops being moved somewhere out of sight.
Copying can work as a classroom too, provided the copied trades go into the journal with the same yes or no column used for your own. After a couple of months the column shows whether the results come from the provider’s process or from your own interference with it.
What changes when you stop punishing good losses
When losing trades stop feeling like personal failures, bad trades become easier to avoid, because most bad trades are emotional reactions to good losses. Revenge trades and sizing up to win money back both start with a legitimate loss that happened to feel like a verdict, so taking away that sting takes away much of the reaction.
Labelling every trade this way should also cut the number of trades, since setups that fail the stranger test no longer get through. The weekly review becomes less unpleasant as well, and a review nobody dreads is far more likely to happen.
None of this guarantees a profit, since a strategy without an edge will lose money however carefully its trades are labelled. What the labels protect is an edge that already exists, through the losing streaks it is bound to go through.
A practical start is the next 20 trades. Label each one Y or N for plan adherence before looking at its result, then compare the average loss in R of the Y trades with that of the N trades.
Questions traders ask about bad trades and losing trades
Is closing a trade early always a bad trade?
Closing a trade early is a bad trade only when your written plan has no rule that allows an early exit. If the plan says to exit when a candle closes back inside the range and you do exactly that, the process held. Closing because the open profit felt too good to give back is the version that turns a planned trade into a bad one.
Is a valid setup I skipped out of fear a mistake?
Yes, skipping a valid setup counts as a process error, even though it costs nothing on the account statement. A strategy’s edge depends on taking all of its signals, including the ones that look ugly, so skipped setups quietly distort its results. Logging them in the trading journal with their hypothetical outcome puts a visible price on the hesitation.
If a sudden news spike hits my stop, was it a bad trade?
It depends on whether the news was scheduled. Holding a position through a jobs report or a rate decision that your plan tells you to avoid makes the trade bad. A truly unexpected headline that hits a properly placed stop produces a losing trade, unpleasant but legitimate.
Does slippage turn a losing trade into a bad trade?
Ordinary slippage, the gap between the price you expected and the price you actually got, doesn’t turn a losing trade into a bad one. It becomes a process issue when you chose an order type or an illiquid hour that your plan rules out. A loss planned at 1R, one unit of planned risk, that ends up costing 1.2R is normal, while one that costs 3R deserves a close look at the trading conditions.
Should demo and replay trades be labelled the same way?
Demo and replay trades should be labelled exactly like live trades, and the habit matters even more there. Practice results feel weightless, so rule breaks creep in quietly, and the habits built in practice are the ones carried into a live account.
What if I honestly can’t tell whether I followed my plan?
Not being able to tell means the plan is too vague to audit. Rules such as enter on strength leave room for interpretation, so they need rewriting until a stranger could check them against a chart. A rule like enter when a 15 minute candle closes above the previous day’s high, with the stop below that candle’s low, can be graded without any discussion.
Does a high win rate mean I’m making fewer bad trades?
A high win rate says nothing about bad trades, because it only counts how many trades closed in profit. A trader can post a high one while breaking rules constantly, for example by pushing stops further away so that losing positions rarely close, a pattern that tends to end with one very large loss. Plan adherence and the average loss in R describe the process far better.
