Scalping leveraged perps is the most fun you can have in crypto trading and the most expensive way to learn discipline. The promise is appealing: take many small trades, get small wins, compound them. The reality is that most scalpers blow up because they cannot survive the cumulative cost of being wrong on small edge. This post is the handbook for scalping perps without getting rekt.
The premise: scalping is a positive-edge game only if your unit economics work. That means low fees, fast execution, tight spreads, and disciplined exits. Get any one of those wrong and the math turns against you across hundreds of trades.
what scalping actually is at high leverage
Scalping at 75x to 250x means taking trades that last 10 seconds to a few minutes, targeting 0.1 to 0.5% moves on the underlying, which translates to 5 to 50% on the position. The win rate is usually 50 to 65% on a good run, with average win sizes slightly smaller than average loss sizes (because stops get hit by wicks).
- Setup types: range fades, breakout retests, order block tests, liquidity sweeps.
- Timeframes: 1s to 1m charts. Anything bigger is not scalping.
- Hold times: 10 seconds to 3 minutes typical, 5 minutes max.
- Position sizes: 2 to 5% of bankroll per trade, since you take many.
when this works: the scalper setups that pay
A few setup types are reliably tradable at high leverage. None are magic. All require discipline on entry and faster discipline on exit.
- Range fade: BTC is grinding inside a 0.2 to 0.5% band. Fade the upper edge with a stop just outside, target the midpoint or lower edge.
- Liquidity sweep: price spikes through a clear level, sweeps stops, then reverses. Enter on the reversal candle close, exit at the prior level.
- Trend pullback: a clear trend is in place. Wait for a 30 to 90 second pullback to a moving average, enter on the bounce candle.
- News reaction: a scheduled headline drops, price overshoots, you fade the overshoot with a fast exit.
- Cascade catch: visible liquidation cascade is happening. Enter into the deepest wick with an immediate exit at the 50% retrace.
when this fails: the scalper death modes
Scalping fails reliably in a few specific ways. Each is preventable with rules.
- Chop without trend: low-volatility chop kills scalpers because every setup looks the same but none follow through.
- High-cost platform: if open + close fees add up to 0.1% of position per trade, you need 0.2%+ moves just to break even.
- No exit discipline: holding scalps "to see what happens" turns them into losing swing trades.
- Overtrading: taking every setup instead of waiting for the high-conviction ones. Variance grinds you down.
- Trading tired: scalping requires sub-second focus. Tired scalpers die.
rules of thumb for scalping perps
These rules are not negotiable for long-term survival. Pick them up early and they will save you a lot of money later.
- Define your exit before you open. Both stop and target. Written down or visualized.
- Honor your stop. The single skill that separates surviving scalpers from blown-up ones.
- Take partials at 1R minimum, even on scalps. Lock something on every winner.
- Cap daily trade count. 20 to 40 trades max for most people. More than that and quality drops.
- Cap daily loss. If you are down 30% on the day, close the tab.
- Cap daily session time. 90 minutes max of focused scalping. After that, fatigue kills.
- Track your trades. A simple count of wins and losses with sizes is enough. You cannot fix what you do not measure.
common mistakes scalpers make
The mistakes are predictable and they cost real money. Each has a clear fix.
- Trading every signal instead of waiting for high-conviction ones. Variance eats you.
- Sizing too big "to make the scalp worth it." If 5% of bankroll feels too small, your bankroll is too big or your trade idea is wrong.
- Holding losers past stops "to give it room." This is a swing trade with no plan now.
- Taking revenge trades after a loss. The next trade should be objectively as good as the last one was, not "more aggressive."
- Scalping illiquid pairs. The slippage at high leverage on thin pairs eats your edge before you start.
- Chasing breakouts. Scalpers should fade, not chase. Breakout chasers die to retests.
the fee math: why scalping needs cheap rails
A scalper running 30 trades a day with 0.05% in fees per trade is paying 1.5% per day in fees alone. Over a month, that is 30% of bankroll in pure fee drag, before considering whether any individual trade won or lost. Scalping is impossible on expensive rails.
On uponly.win, there are no open fees and no fees on losses. The only fee is a small variable cut on net winnings. For a scalper, that means every losing trade costs only the loss, and every winning trade pays out minus a small cut. The unit economics of scalping work here because the platform does not tax each leg of the run.
session structure for scalpers
Discipline structures the session. A loose session with no plan turns into 40 trades of accumulated mistakes. A structured session sets up cleanly.
- Pre-session: identify the day's key levels on BTC and ETH. Five minutes of prep.
- Set caps: max trades, max loss, max time. Written down before you start.
- Trade only the levels you identified. If the market goes somewhere else, do not chase.
- Take a 2-minute break after every 5 trades. Reset.
- End session at the cap, not at the high. Walking away from a winning session is the hardest skill.
why uponly.win is a good scalping environment
Three structural pieces matter for scalpers. Zero open fees means you can take many trades without fee drag eating edge. Zero fees on losses means your unit economics are clean. The default 75x leverage and 500x ceiling cover the entire scalping leverage range. Combined with on-chain Avantis routing, the counterparty is the market, not a house with an incentive to widen spreads when you get hot.
For the leverage selection side of scalping, see our 500x vs 75x guide. For survival rules, see how to not blow up on max leverage. When you want to put a real scalping session on the clock, the rip button is at uponly.win.