Realized vs Unrealized PnL in Crypto: What Each Number Means

When you look at a wallet's performance, you'll see two distinct figures: realized PnL and unrealized PnL. They measure the same underlying thing, how much a position has gained or lost, but at very different points in time. Mixing them up leads to bad reads on a wallet's actual track record.

Definitions and a worked example

Realized PnL is profit or loss that has been locked in by closing a trade. Once you sell, the number is fixed and cannot change regardless of where the token goes next.

Unrealized PnL (sometimes called "paper" PnL) is the gain or loss on a position you still hold. It moves with the price every second and becomes real only when you exit.

Here is a concrete example:

• You buy 1,000 tokens at $1.00 each. Cost basis: $1,000.
• The price rises to $3.00.
• You sell 500 tokens at $3.00, receiving $1,500.
• Realized PnL on the sale: $1,500 − $500 cost = +$1,000.
• You still hold 500 tokens, now worth $1,500 at current price.
• Unrealized PnL on the remaining position: 500 × ($3.00 − $1.00) = +$1,000.
• Total combined PnL: +$2,000, but only half of that is locked in. The other half is still on paper.

Now suppose the price drops back to $1.00. Your realized PnL stays at +$1,000. That trade settled. Your unrealized PnL collapses to $0. The two numbers tell completely different stories about the same position.

Why the difference matters

Realized PnL is the only number that tells you what a wallet actually took home. A wallet showing a $500,000 unrealized gain has simply not sold yet. That gain can vanish in hours on a low-liquidity token.

For evaluating whether a wallet is genuinely skilled, the question at the heart of smart money analysis, realized PnL is what counts: it shows what the wallet actually banked, not what a position is worth on paper. Skilled wallets tend to have high realized PnL relative to their unrealized exposure because they actually exit at strong prices rather than riding positions back to zero.

Unrealized PnL still matters as a signal: it tells you what a wallet currently has at stake. A wallet sitting on a large unrealized gain in a token is a potential seller. That is distribution risk. Watching how those positions resolve, whether they sell into strength or hold through the drawdown, sharpens your read on whether the wallet's timing is repeatable.

The broader context for all of this is covered in the On-Chain Analytics guide, which explains how on-chain data connects to market behavior.

Reading wallet PnL on-chain

Every swap event on a DEX is public. When a wallet buys, the blockchain records the token amount, the price paid (denominated in the paired asset), and the timestamp. When the same wallet sells, you can calculate realized PnL by comparing the exit value to the weighted average entry cost across all prior buys.

On Datablocks, the wallet view shows both figures for any address: realized PnL broken out by token, and the current unrealized exposure. You can see whether a wallet's realized history is a consistent upward line or a noisy mix of wins and losses that net close to flat.

A few things to check on any wallet:

• Realized PnL per token: did they profit on each individual trade, or did a few large wins mask many small losses?
• Win rate: the percentage of closed positions that were profitable.
• Average hold time: short-term flips versus longer accumulation cycles point to different strategies and risk profiles.

These metrics together give you a more honest view of skill than a single headline number.

Common mistakes

Counting unrealized as earned. A portfolio value that has tripled looks compelling until the token loses liquidity. Only realized PnL has actually landed in the wallet.

Ignoring gas and fees. Every swap on-chain costs gas. On high-frequency wallets, fees can meaningfully eat into realized PnL, especially on small positions. A wallet with 200 trades and thin margins might look profitable before fees and break-even after.

Comparing realized PnL without timeframe context. A wallet that made $100,000 realized PnL sounds strong. Over five years on a bear-market portfolio, it is modest. Over three months on a focused Base token rotation, it is exceptional. Always anchor the number to a specific period.

Treating paper gains as confirmation. Wallets with a strong track record sometimes show enormous unrealized gains before a token collapses. The unrealized number does not confirm a thesis was right. Only a completed exit validates the trade.

Reading a snapshot instead of the sequence. On-chain data lets you see every step: entry, partial exits, re-entries, final close. A wallet that looks like it held a perfect position might have re-bought three times on the way down. The full sequence of entries and exits, not a single snapshot, is what shows how the wallet really traded.