Crypto Trading: The Concepts You Need Before Your First Order

There is an awkward asymmetry in crypto trading: the interface takes ten seconds to teach you how to buy, and nobody takes ten minutes to explain what exactly you are buying when you press that button. The result is people who think they are buying "at $79,560" and end up paying something else, or who place a stop-loss convinced it caps their loss at a precise percentage.

This is not a guide to making money trading. It is a guide to mechanics: what an order book is, what you are really choosing with each order type, how a liquidation price is calculated, and what trading costs once you add up a year of fees. Trading actively is speculation, with a serious probability of finishing with less money than you started with, and the numbers in this guide exist precisely to put a size on that.

Market data is from 28 August 2026 and will go stale. The arithmetic will not.

The order book, the spread and depth

A centralised exchange does not have "a price." It has two lists: pending buy orders (bids) and pending sell orders (asks), sorted by price. The number displayed in large type is simply the price of the last trade that crossed.

Between the best bid and the best ask there is a gap: the spread. And behind each price level there is a specific quantity: the depth. Those two things determine what entering will cost you.

Take this sell-side book for BTC (illustrative figures, with BTC around the $79,560 of August 2026):

You send a market order to buy 3 BTC. You do not buy 3 BTC at $79,560. You buy 0.8 at $79,560, 1 at $79,575, 0.7 at $79,610 and the remaining 0.5 at $79,680. Total: $238,790, or $79,596.67 on average. The gap against the price you saw, $36.67 per BTC, is slippage: 0.046%.

In bitcoin, with a deep book, that 0.046% is noise. The problem appears in thinly traded pairs. If the first ask level on a small token holds only $400 of supply and your order is $5,000, your order eats through the whole book and the average price you pay can land several percentage points above the one on screen. There, slippage is not noise: it is the dominant cost of the trade, larger than any fee.

Order types and the classic stop-market failure

The classic mistake goes like this: you place a stop-loss at $75,000 believing you have capped your loss at that level. A violent overnight drop arrives, the price collapses, and by the time your trigger fires the buy orders that sat at $75,000 and $74,500 have already vanished. Your market order chews down through the book and fills well below the level you set.

Stop-limit solves that problem and creates the opposite one: it guarantees the price but not the fill, so you can end up with a resting order and an open position while the price keeps falling. There is no free option here. There is a choice: accept an uncertain price, or accept an uncertain fill.

Maker, taker and what you actually pay

You are a maker when your order rests in the book: you supply liquidity. You are a taker when you execute against what was already there: you consume it. Exchanges charge the former less because the book is their product.

Base-tier spot fees, no discounts, as of 28 August 2026:

Two caveats that save grief. First: a single brand often runs several products with wildly different pricing. Coinbase's simple app and Coinbase Advanced are nowhere near the same, and Kraken's official 0.40%/0.80% figure differs from the historic Kraken Pro schedule (0.25%/0.40%) depending on which product you use. Check your fee in your own account, not in an article. Second: if you buy inside an app advertising "zero fees," the margin lives in the spread applied to you, and that one appears nowhere.

Before choosing where to trade, also check who is actually authorised to serve you. In the EU, MiCA licensing became the gate: Spain's transitional period ended on 1 July 2026, after which only firms with a CNMV licence or an EU passport may operate there. We go through the selection process in how to choose a safe exchange.

Leverage, margin, liquidation and funding

Leverage lets you move a position larger than your money. You post margin, the exchange lends the rest, and if the unrealised loss eats through your margin it closes the position for you: that is liquidation.

The approximate formula for a long is:

liquidation price ≈ entry price × (1 − 1/leverage + maintenance margin)

With numbers. You open a BTC long at $79,560 with $1,000 of margin and 5x leverage: a $5,000 position. Assuming a 0.5% maintenance margin:

• 1 / 5 = 0.20 → your cushion is a 20% fall.
• 79,560 × (1 − 0.20 + 0.005) = $64,045.

If BTC touches $64,045 you lose the full $1,000 plus the liquidation fee. And now the context that makes the number useful: in August 2026 bitcoin moved 26% in a month. A 20% move against you is not an extreme scenario, it is an ordinary month.

At 10x the cushion drops to 10%; at 20x, to 5%. Leverage does not increase your edge. It shortens the distance between your position and the price that kills it.

Perpetual futures add the funding rate: a periodic payment, typically every eight hours, between longs and shorts, designed to keep the perpetual pinned to spot. If buying pressure dominates, longs pay shorts; if selling pressure dominates, the reverse. A common reference rate of 0.01% every eight hours works out to 0.03% a day: close to 11% annualised, leaving your position without the price having moved a cent. In euphoric periods funding runs far above that reference.

What the 2026 numbers say

That table deserves a second look. Both assets are up sharply over the last month and both are down by high double digits over twelve. Someone who bought ETH in August 2025 and left it alone has less than half of it; someone who bought in July 2026 is up a third.

It is the empirical refutation of the sector's most repeated sentence: "long term it always goes up." Long term, sometimes it goes up. The specific window in which you enter and exit decides the outcome, and you do not get to choose it with information about the future.

Worked example: what daily trading really costs

This is the calculation almost nobody runs before starting.

Assume $3,000 of capital and one round trip every trading day: buy and sell. That is 250 market days and 500 orders a year, and assume you execute as a taker, which is what happens when you are in a hurry.

• On Binance (0.100% taker): $3,000 × 0.001 = $3 per order → 500 × $3 = $1,500. That is 50% of the account.
• On Coinbase Advanced (0.60% taker): $3,000 × 0.006 = $18 per order → 500 × $18 = $9,000. That is 300%.
• On Kraken (0.80% taker, base tier): $3,000 × 0.008 = $24 per order → $12,000. That is 400%.

And this excludes the spread and the slippage on every entry and exit.

Translated: on Coinbase Advanced, trading like that, your strategy needs to generate 300% gross a year just to break even. Fees are not an operational detail; in frequent trading, fees are the operation. Dropping from daily to weekly divides that cost by five, and using limit (maker) orders cuts it by roughly another third on platforms with tiered pricing.

DCA: a mechanical alternative, not a solution

Dollar cost averaging means buying a fixed amount at a fixed interval, ignoring the price. Its real virtue is that it removes the decision: you do not time the entry, but you do not mistime it either, and you drastically cut the number of trades and therefore the fees.

What DCA does not do: it does not protect you from a sustained decline. If you buy $100 of ETH every month through a year in which it falls 44%, you end up with more ETH and less money. Averaging your entry price does not turn a falling asset into a rising one; it only spreads out the moment of entry. And if you do it on a platform charging 1.99% on small fiat purchases, you are paying almost 2% every month for the privilege.

The biases that empty accounts

• Loss realisation aversion. You sell winners to "lock it in" and hold losers so you never admit the mistake. The systematic result is a portfolio made entirely of your worst decisions.
• Overtrading. Every trade has a guaranteed cost and an uncertain outcome. Multiplying the number of trades multiplies the only certain part of the equation: what you pay.
• Revenge trading. After a loss, doubling size to win it back fast. You need a 100% gain to recover a 50% loss; arithmetic does not forgive pride.
• Mistaking volatility for opportunity. An asset moving 30% in a month does not mean you will capture that move in the right direction.

Tax: the swap counts too

This is where most people get an unpleasant surprise. In most jurisdictions, a crypto-to-crypto swap is a disposal. Trading BTC for ETH generally crystallises a gain or loss at that moment, measured in your local currency, even though nothing reached your bank account. Someone doing 500 trades a year potentially has 500 taxable events to calculate and evidence.

The framing matters more than the label. Different countries slot the same transaction into different categories — capital gains in some, general income in others — and the applicable rates range from zero to well into the high tens of percent depending on residence, holding period and total income. Some jurisdictions also apply specific rules for matching acquisition costs across identical assets, which changes the gain on every one of those 500 trades. Do not assume a rate you read about somewhere applies to you.

Separately, there is a reporting layer that has nothing to do with what you owe. Many countries now oblige exchanges and custodians to report user balances and transactions to the tax authority directly, and several impose an additional obligation on the individual to declare foreign-held crypto balances above a threshold. That means the authority frequently already holds a picture of your activity from another source, and a mismatch is what triggers a review.

Before your first order

• Check in your own account which maker and taker fee actually applies to you, and on which product.
• Look at the depth of the pair, not just the price: if your order is large relative to the first level, split it.
• Use limit orders by default; save market orders for when execution matters more than price.
• If you are going to use leverage, calculate the liquidation price before opening and write it down.
• Decide in advance the total you are willing to lose, and treat it as money already spent.
• Log every trade with date, amount and local-currency value: that is your tax accounting.
• Verify the platform holds the authorisation required where you live, in the official register of your regulator.

If what you wanted was crypto exposure without trading daily, there are routes with a very different cost structure: from staking to simply holding the asset, by way of using stablecoins as an operational parking spot between trades, with their own issuer risk. None of these is a recommendation. They are different mechanisms with different costs, and it is worth knowing which one you are using.

Frequently asked questions

Sources and references

1. Binance — Trading Fee Schedule
2. Kraken — Spot fee schedule
3. Coinbase Help — Pricing and fees
4. mempool.space — Bitcoin network status
5. Etherscan — Gas Tracker
6. Regulation (EU) 2023/1114 (MiCA), consolidated text
7. CNMV — Register of crypto-asset service providers
8. SEC — Joint SEC/CFTC clarification on crypto assets (17 Mar 2026)

Related guides