Cost Basis vs. Market Value: Which Number You're Looking At

Your tracker says the position is worth $4,180 and that you are up $4,180 on it. Those two numbers being identical is not a coincidence and it is not a display bug. It is the app telling you, in the only way it can, that it has no idea what you paid.

This is the most common broken number in crypto tracking, and it is invisible in the place people usually look. The balance is right. The quantity is right. The price is right. Only the profit is fiction, and profit is the number most people actually make decisions on.

What is the actual difference between the two numbers?

Market value is the easy one. Quantity multiplied by current price. If you hold 0.4 ETH and ETH trades at $3,100, your market value is $1,240. Every tracker computes this the same way, and the only thing they can disagree about is which price feed they trust, which is a separate problem covered in why the same coin shows two different prices.

Cost basis is what that 0.4 ETH cost you to acquire. If you bought it in one transaction at $1,900 with a $4 fee, your basis is $764. Your unrealised gain is $476. Simple, as long as there was exactly one purchase.

There is almost never exactly one purchase.

Buy ETH four times over two years at four different prices, sell a bit, buy again, and the question "what did this 0.4 ETH cost me" no longer has a single obvious answer. You did not buy this specific 0.4 ETH. You bought a series of amounts at a series of prices, and you now hold a fraction of the total. Which of those purchases is the one you are still holding?

Why does changing the accounting method change my profit but not my balance?

Because the method is the answer to that question, and there is more than one defensible answer.

Work an example. You bought 1 ETH at $1,500, then 1 ETH at $3,500. You now hold 2 ETH, and you sell 1 at $3,000.

Under FIFO — first in, first out — the ETH you sold is the one you bought at $1,500. Realised gain: $1,500. The ETH you still hold carries a basis of $3,500, so at a $3,000 price you are sitting on an unrealised loss of $500.

Under LIFO, last in first out, you sold the $3,500 lot. Realised loss: $500. The remaining ETH carries a $1,500 basis, so you hold an unrealised gain of $1,500.

Under average cost, both lots pooled to a $2,500 basis per unit. Realised gain: $500. Remaining unrealised gain: $500.

Three methods, three completely different profit figures, one identical outcome: you hold 1 ETH and you received $3,000 in cash. Your balance never moved. Nothing about your actual financial position changed. The only thing that changed was the story about which coins left.

That is worth sitting with, because it explains why two apps tracking the same accounts can show wildly different performance and both be correct. It also explains why a tracker's profit number is not a fact about the market. It is a fact about a bookkeeping convention. Which convention you are allowed to use is set by the tax authority where you file, not by you, and getting that wrong is a real problem rather than a cosmetic one. Where a tracker ends and tax software begins covers the line between the two tools.

Why do transfers destroy cost basis?

This is the failure that catches nearly everyone, and it has a mechanical cause worth understanding rather than just patching.

When a tracker reads an exchange through a read-only key, it can see your trade history. A trade is a two-sided event: you gave up dollars, you received an asset, at a price, on a date. All the information needed to establish basis is present.

When a tracker reads a wallet address on-chain, it sees no such thing. It sees quantity arriving. The blockchain records that 0.4 ETH moved from one address to another. It does not record why, or what you paid for it, or whether you paid anything at all, because none of that information exists on-chain. A purchase, an airdrop, a gift, a withdrawal from an exchange and a transfer between two of your own wallets all look approximately identical to an indexer.

Faced with an acquisition of unknown price, most trackers default to zero. It is the only assumption that does not invent a number. The consequence is that the entire market value of the position gets reported as gain.

You will see this most often in exactly the situation where it hurts most: coins moved off an exchange into self-custody. The person who did the responsible thing and took their long-term holdings to a hardware wallet is the person whose portfolio now claims an impossible profit. Tracking a hardware wallet without exposing it covers the watch-address approach, and the basis problem comes free with it.

The reverse also happens. Move coins between two wallets you both track, and some trackers count the incoming side as a fresh acquisition while the outgoing side reads as a disposal, which manufactures a realised gain out of a transaction that was economically nothing. Duplicate transactions after connecting an exchange covers spotting that pattern.

How do I repair a basis that has already broken?

The repair is bookkeeping, and it is finite. Most people need one evening.

1. Find the positions with a zero basis. They announce themselves: the profit figure equals the market value, or the gain percentage is a suspiciously round and very large number. Sort your holdings by profit percentage and the broken ones cluster at the top.
2. Establish the real acquisition date. A block explorer gives you the timestamp of the incoming transfer. If the coins came from an exchange, the withdrawal record there gives you the same date, and the buy that preceded it gives you the price you actually paid.
3. Get the price you paid, not the price on the day it landed. These differ. The acquisition happened when you bought, not when you moved the asset. Using the transfer-day price is a common shortcut and it quietly rewrites your history.
4. Enter the position manually with that price. Adding a manual position walks through the fields. Where a transfer created a phantom acquisition, remove it rather than editing it, so the ledger reflects one purchase rather than two.
5. Export and keep a copy. Once reconstructed, this data is expensive to rebuild a second time. Exporting your data covers getting a file you own out of any tracker before you need it.

Do this the same week you move coins, not three years later. The information you need is trivially available on the day of the transfer and genuinely hard to recover once an exchange account has been closed and a year of price history has to be looked up one date at a time.

The habit worth building is smaller than the repair: when coins leave an exchange, write the purchase price down somewhere before you close the tab. Thirty seconds then saves an evening later, and it keeps the one number you actually make decisions on connected to something real.