Crypto day trading: strategies and rules [2026]

Day trading is a style where every position you open gets closed before the trading day ends. Not a single trade carries over to the next day. That's it.

The crypto market runs around the clock, there's no official session close here, so the trader defines the "day" himself. Most often it's the stretch between funding rate settlements, or it's tied to a working window of 4-8 hours. When the window ends, there are no positions left in the terminal.

What this gives you in practice:

• No carrying cost. The funding rate (the fee between longs and shorts on perpetual futures) is charged every 8 hours, and at a rate of 0.05% three days in a long costs roughly 0.45% of position size. For an intraday trade that expense simply never appears.
• You start the morning with a clean slate. There's no position hanging there that went 6% underwater overnight.
• A liquidation cascade at 3 a.m. is none of your business, because you aren't in the market.

There's a flip side too. A day trader gives up the big moves. If BTC covered 12% over a week, intraday crypto trading will pull maybe 4-5% of that in total across several trades, while a swing trader takes most of it with one position. It's a conscious trade-off: less potential per trade in exchange for control.

A typical day looks like this. In the morning you mark levels on H4 and H1 and pick 3-5 coins with decent volume. Then comes the waiting. There will be two real entries per session, three at most. The rest of the time the trader just watches.

Day trading vs scalping vs swing

The difference between the three styles isn't speed. It's which source of profit the trader exploits.

A scalper makes money on micro-inefficiencies: slippage, imbalance in the order book, the reaction to a large limit order. His target is 0.2-0.5% per trade, and he compensates for the small size with quantity. The approaches are covered in "What crypto scalping is".

A day trader makes money on intraday momentum. He needs a move the market can cover in 40 minutes or in three hours: a level being worked off, an exit from the overnight range, a reaction to news. Target 1.5-4%.

A swing trader works with the market phase and holds a position from two days to several weeks. Details are in "Crypto swing trading".

For years I've been watching the same picture: a beginner shows up on the market and tries to start with scalping, because that's where "the money comes faster." A month later he's down on commissions with his psyche in pieces. Intraday crypto trading is more honest in that sense as an entry point into the profession. The decision speed is human, and the logic is exactly the same.

If you're still short on basics, start with the free lesson from the "Trading from scratch" course on the Secret Terminal YouTube channel, it covers exchanges, futures and the mechanics of a trade. Choosing your working interval is laid out in "Timeframes in trading".

Day trading strategies

What follows are four approaches that produce a repeatable result within the day. All of them rest on one idea: price moves toward liquidity, and we look for confirmation of the move not in an indicator but in the real flow of trades. A general overview of systematic approaches is collected in "Crypto trading strategies".

Trading from levels

The basic and most workable option. Price approaches a zone where it has already reversed before, and we watch who's stronger there.

In the morning you mark the previous day's extremes on H4, round numbers and the boundaries of the overnight range. That gives you 4-6 lines per coin. Then you wait for price to approach one of them.

The level by itself means nothing, what matters is the reaction to it. If price approaches support from above, you need to see three things: a large limit order on the bid in the order book, a buyer showing up on the tape, buying dominance by delta in the cluster.

How large does the density level in the order book (a stack of limit orders at one price) need to be? Compare it with the average volume that goes through inside the cluster. If the cluster usually trades 5 million and there's a 10 million order sitting on the level, it's unlikely to be taken out in one move. If the density is 500 thousand against those same 5 million of turnover, it won't hold anything.

The stop goes behind the level, not "0.5% below entry." The take goes to the nearest opposing density level, not to a round number in your head.

Where it breaks: on days with strong trending momentum, levels get punched straight through. If BTC has covered 5% in one direction since morning with no pullbacks, counter-trend work from levels isn't your play that day.

Breakout + retest

Don't chase a breakout without a retest. Not ever.

The classic mistake: price flies out of the range to the upside, the candle is green and huge, the trader hits Buy at market and gets filled at the very high. Then a 60% retracement of the move, and the position is 1.2% underwater with a 1% stop.

The working scheme is different. We wait for the level to break, we wait for it to hold (price hasn't returned into the range within 2-3 M15 candles), we wait for the pullback to the broken level. Entry happens on the pullback, when former resistance starts working as support.

Again we look for confirmation in the flow. On the pullback, seller pressure on the tape should be fading. If selling volume on the pullback is growing instead, that's not a pullback but a reversal, and you shouldn't enter.

Trade example. The 3,420 level on ETH broke to the upside, price went to 3,468. The pullback comes to 3,424. Entry with a limit order at 3,426, stop at 3,408 (behind the level with room), target 3,490. Risk 18 points, potential 64, ratio roughly 1 to 3.5. Leverage 5x, size calculated so the stop costs 1% of the deposit. Result: target hit after 2 hours 40 minutes, half the position closed at 3,444.

Where it breaks: on coins with thin liquidity the breakout often turns out to be one algorithm at work. The order book is empty, the volume is painted, and the pullback turns into a return into the range.

Trading the news

Intraday news trading is built not on guessing the reaction but on working with volatility after the data comes out.

The main events that move crypto: Fed meetings and US inflation data, spot ETF decisions, major listings on Binance and OKX, hacks and withdrawal halts, token unlocks.

Tactic one, the safest. Don't enter before the news at all: 10-15 minutes before the release, close positions and pull your limit orders. At the moment the data drops the order book is empty, market makers pull their orders, the spread widens, and slippage on a market order eats half the day's profit. After the release we wait 5-10 minutes, let the market form the first M5 candle and build a scenario off its extremes.

Tactic two. Working the momentum burnout. A sharp move on news almost always retraces partially. If price shot up 8%, a 3-4% pullback plays out fairly regularly. The trader's job is to see the stall in the order book (opposing density levels appearing, prints fading) and enter against the impulse with small size and a tight stop.

Tactic three, for listings. The first minutes after trading opens give maximum volatility, and here size is calculated not from your deposit but from liquidity. If there are $200 orders sitting in the order book, an entry of 2,000 will hit the market on the way out and give up 5-10% just closing the position.

What not to do: enter at market at the moment of the release, hoping to guess the side. That's a lottery with negative expectancy because of the spread.

Trading volume

An approach where the signal comes not from price but from an anomaly in the flow.

First situation. Price is flat, volume is rising. 8 million goes through the cluster, and the M15 candle closed with a 0.1% body. That's absorption: someone is building a position, soaking up the opposing flow. We determine direction by delta, and enter in the direction of whoever is absorbing, after price first exits the zone.

Second. An abnormal volume spike on a breakout. If the level broke on volume 3-4 times above the average of the last 20 candles, the odds of it holding are substantially higher. Low volume on a breakout signals that momentum is fading.

Third. Working from POC (the price with the most traded volume over the period). The level pulls price toward it: if within the day price has moved 2.5% away from POC without a clear trend, a return to it plays out often.

Trade example. SOL, absorption in the 168.4-168.9 zone, over 40 minutes 11 million went through the cluster with price moving 0.3%, delta positive. Entry on the exit upward at 169.2, stop 167.8, target 173.0. Leverage 5x, risk 1% of the deposit, R:R roughly 1 to 2.7. Result: the target wasn't reached, I took 60% of the way at 171.6, because ahead of the target a 4 million opposing density level grew in the order book.

I tested this on BTC/USDT: the first variant works most consistently there, liquidity is sufficient and absorption is clean to see. On alts the picture is dirtier, you have to make an allowance for noise.

A day trader's tools

Timeframes: M15-H1

You need to work on at least three intervals at once, and each has its own role.

H4 and D1 — morning markup. You open them once before the session, mark the significant extremes, round numbers, the weekly range boundaries. You don't look there again during the day.

H1 — context. Here you can see what phase the coin is in: trend, consolidation, reversal. H1 is exactly what answers the question of whether we're trading with the trend today or off the range boundaries.

M15 — the main working interval. This is where you look for the setup: an approach to a level, a hold after a breakout, an absorption zone. The M15 candle closes 4 times an hour, and that pace is enough to make decisions without rushing.

M1 and M5 — only for refining the entry point. Not for the decision to trade, but to place a limit order 0.15% better on price.

A common beginner mistake: making the decision on M1 and justifying it with the context on H4. It works exactly the other way around. The decision is made on the higher interval, the execution is refined on the lower one.

Order book + tape for confirmation

The chart shows the result of a move. The order book and the tape show the process. The difference is enormous.

The order book answers the question of where liquidity sits: you see limit orders that haven't been filled yet, and where a large player is ready to defend price. The mechanics are covered in "The exchange order book".

The tape answers the question of who's more aggressive right now. These are already executed trades, real money, not intentions. Buying accelerating on the tape as price approaches support is an argument. Silence on the approach to the same support is a reason to skip the trade.

The cluster answers the question of where volume actually went through and who controlled the situation. It also helps tell a real order from spoofing (placing a large order with no intention of filling it). If the density level held, price came to it, and volume actually went through in the cluster, the order was real. If the order vanished before price arrived and there's no volume, someone was trying to fool you.

What Secret Terminal gives an intraday trader:

• A lifetime timer for density levels right in the order book. A density level that has held for 30 minutes and one that appeared 4 seconds ago are different things. The first can be built into your scenario, the second is most likely an algorithmic imitation.
• A global density module across all connected exchanges in one table. In the morning, two minutes shows you where large limit orders exist at all and where the order book is empty with nothing to work with.
• Density levels projected onto the chart as horizontal lines with the volume shown.
• A quotes module with columns for turnover, trade count and the presence of density levels. Picking coins for the day takes one pass.
• A journal with statistics: PnL calendar, cumulative curve, a second-by-second breakdown of every trade.
• A candle close timer. For strategies that need confirmation on the close this matters: you don't enter 40 seconds before the end of a bar that can still reverse.

Plus hotkeys: Z to place a stop and take, G to move the stop to breakeven, Space to cancel all active limit orders, LeftCtrl for an emergency market close of the position.

A day trader's rules

Daily loss limit

The daily loss limit is the one rule that separates a trader from a gambler. Without it, any system sooner or later ends in a blown account.

The working formulation: the trading day closes when you hit minus 3% of the deposit or after three stops in a row, whichever comes first.

Three stops in a row mean your reading of the market today doesn't match what the market is doing. The reasons vary: the character of volatility changed, news came out that you missed, or you're simply tired. The fourth trade in that state is almost always opened with increased size. That's tilt.

Let's count. Deposit 2,000 USDT, risk 1%, meaning 20 USDT per trade. Three stops in a row is minus 60 USDT, tomorrow you start with 1,940 and that's still a working deposit. Double your size to win it back and catch two more stops and the account is already 1,860, and to get back you need to make 7.5%.

The second element is a limit on the number of trades, no more than six per session. Closer to that number the quality of entries drops: first you were waiting for the setup, then you started drawing it in for yourself.

The third element is calculating size from the stop, not the other way around. First you determine where the stop technically has to sit (behind the level, behind the density level, behind the extreme), then you calculate size so the stop costs exactly 1% of the deposit. Fitting it backwards is what produces stops right under the entry that any noise takes out.

And the rule that gets broken most often: the stop doesn't get moved. Ever.

Common day trader mistakes

Five scenarios that blow up deposits most often.

• Carrying a trade overnight because "it's about to go green." The moment you carry a position once, you're no longer a day trader, you're a swing trader without a swing plan. The stops, the size and the risk management there are different.
• Entering at market at the moment news comes out. The spread is wide, the order book is empty, slippage on a market order eats the day's result before price even picks a direction.
• Trying to win it back after a stop. Size doubles, analysis disappears, the next two trades are opened on emotion. The daily limit exists precisely against this.
• Trading a coin with no liquidity. A pretty setup on the chart with an empty order book is worth nothing: you won't be able to get out at the price you need.
• Ignoring the funding rate schedule. Entering a long 10 minutes before settlement at a rate of 0.08% immediately takes away part of the trade's potential.

The free lesson from the "Trading from scratch" course helps you break these situations down on a live market, it shows how professionals read the order book and clusters before entering. The lesson is part of the full free course on YouTube.

Taking profit

Taking profit intraday works on inverted logic: you have to close the position earlier than you want to.

A scheme that holds up statistically. The first part, 40-50% of size, is taken at a move of 1R, meaning when profit equals risk. At the same time the stop goes to breakeven, the trade becomes free, and the pressure lifts. The remainder rides to the nearest opposing density level or to the opposite boundary of the daily range.

What counts as a target. Not round numbers out of your head, but concrete objects on the market: a large limit order higher up in the order book, the previous day's extreme, yesterday's session POC. The target has to be a place where someone else's money actually sits.

Three signals to exit early even if the target hasn't been reached:

• Delta in the cluster turns around. You're long, and the last two M15 candles closed with negative delta on rising volume. The buyer is spent.
• A large opposing density level has grown in the order book ahead of your target, one that wasn't there at entry.
• The scenario stopped working: you entered on a bounce from a level, the level got punched through, price came back below it.

Costs are what people forget most often. At a taker fee of 0.05%, entering and exiting at market costs 0.1% of position size, with 10x leverage that's 1% of margin, and five trades a day gives 5% of margin on fees alone. So it's better to enter with limit orders (maker fee 0.02%).

My personal filter over the last couple of years is simple: if I can't name the specific price where someone else's liquidity sits, I don't take the trade. It usually cuts out half the entries, and it's exactly that half that used to produce losses.

FAQ

• What is day trading in simple terms?

  Day trading is trading where all positions are opened and closed within a single trading day. Trades aren't carried overnight. Everything else follows from that. You don't pay a funding rate for a long hold, you don't get a morning price gap, you don't wake up with a position 8% underwater. Average trade duration runs from 20 minutes to a few hours, with 2-6 trades per session.

• How much money do you need for crypto day trading?

  The working minimum for futures is 1,000-2,000 USDT. You can do it with less, but the statistics stop working. The reason is the math of risk. With a 2,000 deposit and 1% risk you're risking 20 USDT, and that's enough for a stop placed behind structure. With a 200 deposit the risk comes to 2 USDT, and fees plus slippage will eat a noticeable part of that. You'd have to put the stop right up against the entry, and noise will keep taking you out.

• How is day trading different from scalping?

  The scale of the move and the number of trades. A scalper takes 0.2-0.5% and makes dozens of entries, a day trader takes 1.5-4% and makes 2-6. Hence the different requirements. Scalping needs minimal ping and constant eyes on the order book, day trading gets by with monitoring M15 and checking order flow at entry. For a scalper fees are the main expense item, for a day trader just a noticeable one.

• Which timeframe is best for day trading?

  The H1 plus M15 pairing, with the entry refined on M1 and the order book. H4 is needed in the morning to mark levels, H1 gives you the market phase, M15 shows the setup, M1 only refines the limit order. There's no point going below M5, that's already scalper noise.

• What daily loss limit should a day trader set?

  Minus 3% of the deposit or three stops in a row, whichever comes first. The limit isn't there to protect you from one loss, it's there to protect you from the series of decisions that follows it. After the third stop the brain switches into win-it-back mode, size grows and analysis disappears. You have to stop mechanically, without negotiating with yourself.

• Can you day trade from a phone?

  You can monitor an open position, you can't enter a trade. On a phone you don't have the order book, the tape or clusters, and those are the main confirming arguments. The decision gets made from the candle alone, which noticeably lowers entry accuracy. Marking levels from a phone in the morning is realistic, everything else needs a full terminal.

• How many trades a day is normal for a day trader?

  From zero to six. A day without a single trade is a normal result. The market isn't obliged to hand you a clean setup every day. In a low-volume range the best entry is no entry. If the counter has gone past ten, you're no longer trading a system, you're trading the urge to trade.

A terminal for intraday trading

An intraday trade lives on confirmation. A level with no reaction in the order book and on the tape is just a line on a chart.

Secret Terminal brings all the arguments into one window: an order book with automatic density highlighting and an order lifetime timer, the tape, clusters with delta and POC, a global density map across Binance, Bybit, OKX, MEXC and WhiteBIT, and a journal with statistics on every trade.

Download the terminal and see what the market looks like from the inside, instead of reading candles after the fact.