Leverage and Costs -- Why Small Accounts Can Go to Zero Overnight
In L1.1, you learned that what most retail platforms offer is not spot forex, but CFDs. Behind the EUR/USD quote you see are brokers, counterparties, spreads, and execution rules.
Now we answer the other half: why can a retail account disappear at 100% speed on the same EUR/USD quote, while institutions can survive for decades in the same market? The answer is not as simple as "institutions are smarter." It comes down to three words: leverage, margin, and costs.
In this lesson, we look at Mike's story. A 100:1 leveraged account did not go to zero because every call was wrong. It was wiped out over three months by a very ordinary mathematical path.
How Leverage Amplifies Both Profit and Mistakes -- Mike's Story
Mike was 32, a front-end engineer at an internet company in Shanghai. In August, he joined a "forex trading group." Every day, the group owner posted profit screenshots: 80 pips caught on GBP/JPY today, gold doubled and withdrawn tomorrow. At first Mike just watched. Then one line got to him: "A small account without leverage will never grow."
He opened a $5000 account and chose 100:1 leverage. His reasoning sounded reasonable: "This is a system I already practiced on demo money. Not using leverage would be a waste." In the first month he made $800. In the second month he made another $1200. The group owner started tagging him in the chat: "Brother, you're getting it now. You've found the rhythm."
In the third month, on September 6, NFP was released. EUR/USD first rushed in the direction of his position, and his floating profit turned the account screen green. He did not close. He added another position. The next second, price snapped back, the spread widened, and his stop did not fill at the price shown on the screen. The screen flipped from green to red, then to black. Balance: $342.
Mike closed the laptop and drank a glass of water in the dark kitchen at 2 a.m. The apartment was quiet. Only one sentence kept repeating in his head: "It is not that I cannot trade. Leverage killed me."
100:1 leverage means something very direct: you use 1 dollar of margin to control 100 dollars of notional position. If price moves 1% against you, the notional position loses 1%, but relative to your capital that is 100%. This is not a metaphor. It is multiplication.
For EUR/USD, a daily move of 0.5%-1.5% is common. During high-impact news like NFP, CPI, or FOMC, a 50-100 pip jump in 1-3 seconds, roughly 0.5%-1%, is not rare. Under 100:1 leverage, a 1% adverse move is enough to break the account. Even if you reduce leverage to 50:1, a 2% adverse move can still zero the account; during an ECB rate decision or a surprise central bank speech, that kind of instant move does not exist only in textbooks.
What makes leverage more dangerous is that it disguises itself as "capital efficiency." When Mike saw $5000 controlling $500,000, his first reaction was not fear. He felt he finally had the right to make serious money. He forgot the other side: the market only needed a small price movement to create a large dollar swing in his account. When profits came, it felt like skill. Only when losses came did leverage show its real nature.
Large brokers have disclosed similar statistics themselves: the average retail client lifespan is under 12 months, and more than 60% of new clients lose enough in the first 3 months that they cannot keep trading. The reason is not necessarily that they are wrong on direction every time. It is that leverage leaves them room to be wrong only once or twice.
The next day, Mike left the group, withdrew the remaining $342, and opened a new account with 30:1 leverage. Later, he wrote on his blog: "The most expensive lesson leverage taught me was to admit I am not a genius."
Account $10,000. You want to risk 1% to go long EUR/USD at 1.0850 with stop at 1.0820. If 1 pip = $10 per standard lot, what lot size?
Margin / Margin Call / Liquidation Path -- How Your Account Gets Wiped Out
Margin is the money you put on the table for a position. 100:1 leverage = 1% margin requirement. A $5000 account can theoretically control up to $500,000 notional. It sounds like permission. In reality, it is a risk allowance.
After you open a position, you need to separate several numbers in the account. Used margin is the margin already occupied by open positions. Free margin is the room still available after account equity minus used margin and unrealized losses. Margin level usually equals equity / used margin x 100%. The lower this percentage gets, the closer the broker is to closing positions for you.
Different brokers use different thresholds, but a common path looks like this: when margin level falls below 100%, the system issues a margin call, reminding you to deposit more money or reduce positions; below 50%, it enters stop out, and the broker no longer waits for your confirmation. It starts forcibly closing positions, usually from the largest losing position.
Notice that a margin call is not a teacher grading your homework, and stop out is not an airbag built for you. They protect the broker first. In a CFD, you and the broker settle the price difference. If your account loses beyond its balance, the remaining gap becomes the broker's credit risk.
This is why many new traders see money still left in the account, then suddenly discover that their positions have been closed. They are watching balance, meaning realized account balance. The broker is watching equity, meaning balance plus unrealized P&L. During a fast market drop, balance may still show $5000 while equity has already fallen to $900. The system will not wait for you to accept that psychologically.
Look at Mike's path. When he opened the position, his account equity was about $5050, used margin was $50, and margin level was 10100%, which looked absurdly safe. After the first NFP jump, floating loss reached -$3500, equity was down to $1500, and margin level was still 3000%, so he thought it was only a drawdown. Then the spread widened and price kept moving against him. Floating loss approached -$4500, equity fell to $500, and margin level was 1000%. On the numbers alone, it had not yet triggered a 50% stop out. But the problem was that he had more than one position, and after adding, used margin rose quickly. The system began queuing liquidation while liquidity was disappearing.
By the time the forced liquidation filled, the price was dozens of pips worse than he expected. The position closed, and account equity was only $342. Forced liquidation did not lock the loss at some clean theoretical price. It simply threw the order into the market that existed at that moment.
This is why a hard stop is far safer than a mental stop. A hard stop can still slip, but at least you have signed a contract in advance: "past this point, I will not keep losing more." A mental stop requires you to stay calm, click, confirm, and wait for execution during the most chaotic second. Most people cannot do that. Mike could not either.
Drag the 3 sliders to see ruin probability live. "Ruin" = account drops below 50% of start — a hole almost no one climbs out of.
Adequate: this setup is comparatively durable, assuming you keep the rules consistent.
The Three-Part Cost Stack -- Spread / Swap / Slippage
Leverage determines how many times you can be wrong. Costs determine how much you have already lost before you even judge direction. In retail trading, the three most common costs are spread, swap, and slippage.
Spread was covered in L1.1: ask - bid, the cost you pay the instant you open a position. The problem is frequency. Suppose a day trader makes 30 EUR/USD trades per day, average spread 1.5 pip, 1 standard lot each trade, across 252 trading days in a year. Spread cost is approximately 252 x 30 x 1.5 x $10 = $113,400. Many retail traders do not make that much in a year, but the cost is deducted every day.
Swap is the interest-rate difference charged or paid when you hold a position overnight. In theory, if you are long the higher-yielding currency against the lower-yielding currency, such as USD/JPY or AUD/JPY, you may receive swap; the opposite direction may pay swap. But a CFD broker's swap table is not a textbook interest-rate differential. It includes markup. B-book brokers often give you negative swap in both directions: long pays, short also pays. A long-holding strategy that does not read the swap table is slowly being eaten.
Slippage is when you click a market order and the actual fill price is not the price shown on the screen. In normal conditions it may be 1-2 pips; during news it may be 5-20 pips. The real meaning of a market order is: I want execution now, and I am willing to accept the price the market gives me. Only a limit order can lock in the worst price you are willing to accept, but the cost is that it may not fill.
So do not understand "trading costs" as just the spread line on the broker's page. Real cost is the total of spread + swap + slippage. A retail trader using 100:1 leverage, trading 30 times a day, holding positions overnight, and not reading swap can lose 12%-25% of the account to costs in a year even if directional judgment is not bad.
The cruelest part of costs is that they do not require you to make a mistake. You can be right on direction, but entry costs 2 pips in spread, exit slips another 3 pips, and holding overnight deducts swap. After ten trades, you feel like you are "always just a little short of making money." That little gap is not luck. It is the fixed tax the execution structure takes from your account every day.
Your strategy's win rate must beat this number first before it has any right to talk about positive return. Otherwise, you are not competing with the market. You are paying rent to the execution environment.
100:1 leverage = 1% price movement = 100% account. Leverage does not amplify profit. It reduces the number of failures you can survive to 1.
The three-part cost stack (spread + swap + slippage) can eat 12-25% of the account in a year. Your strategy's win rate must beat this baseline first before you talk about directional judgment.
Next lesson L1.3: A First Look at Market Structure -- Chen's Story + Identifying HH/HL