Tax · 8 min · 2026-06-14

If you trade CS2 skins for long enough, you end up owning several copies of the same item bought at different prices. Maybe you grabbed an AK-47 Redline at $100 last year, another at $180 over the summer, and two more as the market climbed. When you sell two of them, a question that feels like pure bookkeeping turns out to set your tax bill: which two did you actually sell? You held four identical items, so there is no physical answer. The IRS lets you pick a rule, and the rule you pick decides the cost basis you subtract from your proceeds, which decides your gain, which decides your tax. Same trades, same dollars in and out, three very different numbers on your return. I built fy_nance partly because this trips up almost everyone, so let me walk through it with real figures.

This is general information, not tax advice. Talk to a CPA about your specific situation.

The setup: lots, basis, and why the method matters

Every time you buy a skin, you create a "lot": a unit with its own cost (the basis) and its own acquisition date. When all your lots are different items, basis tracking is easy, because each sale maps to exactly one purchase. The trouble starts with fungible-feeling inventory, several units of the same skin at different prices. Now a sale of two units could be matched against any two of your lots, and the match you choose changes everything downstream.

The cost-basis method is just the rule for choosing. The three you hear about most are FIFO, LIFO, and HIFO. They do not change how much cash you made. They change how much of that cash counts as a taxable gain this year, and whether that gain is taxed at the lower long-term rate or the higher short-term rate.

The three methods in plain terms

FIFO (first in, first out) sells your oldest lot first. It is the default that the IRS assumes if you do not specify and document something else, and it is the simplest to defend. Because it reaches for your oldest holdings, those lots are more likely to qualify as long-term (held over a year). The catch: in a market that has been rising, your oldest lots are also your cheapest, so FIFO tends to book the largest gain.

LIFO (last in, first out) sells your newest lot first. In a rising market the newest lots have the highest basis, so LIFO books a smaller gain this year. The trade-off is the holding period: your newest lots are the ones you have held the shortest time, so LIFO gains are more often short-term and taxed at the higher rate.

HIFO (highest in, first out) sells your highest-basis lot first, regardless of when you bought it. This minimizes the gain you realize this year, which usually produces the lowest current-year bill. But it is a deferral, not a free lunch: you are spending your high-basis lots now, so what is left in inventory is your cheap stuff, and a future sale will carry a bigger gain.

A worked example

Say I own four copies of the same skin, bought in this order and at these prices:

• Lot 1: $100
• Lot 2: $180
• Lot 3: $260
• Lot 4: $340

The market has run up and I sell two of them at $400 each, so my proceeds are $800. Here is how each method matches my sale against my lots.

Same eight hundred dollars in the door. FIFO reports a $520 gain, LIFO and HIFO each report $200. At a 24% marginal rate, that gap is about $77 of tax on one pair of sales. Scale that across a year of trading and it stops being rounding error.

Why LIFO and HIFO match here, and when they do not

In the table above LIFO and HIFO land on the exact same two lots. That is not a coincidence, but it is also not a rule. It happens because I bought in increasing price order, so my newest lots and my most expensive lots are the same lots. LIFO looks at recency, HIFO looks at price, and when price climbs monotonically with each purchase those two orderings agree.

They come apart the moment your buys are out of price order, which is normal when a market dips and recovers. Suppose instead I had bought in this sequence:

• Lot 1: $100
• Lot 2: $340
• Lot 3: $180
• Lot 4: $260

Selling two units now splits the methods:

LIFO grabs the two most recent purchases ($260 and $180). HIFO ignores when I bought and grabs the two most expensive ($340 and $260). The gains diverge by $160. The lesson worth internalizing: HIFO is about price, not recency. They only look alike when you happen to have bought in a straight line.

The catch nobody mentions: lowest gain is not always lowest tax

It is tempting to read the example and conclude HIFO always wins. It often does on the current-year gain, but the gain is only half of the tax. The other half is the rate, and the rate depends on how long you held the lot you sold.

Long-term gains (assets held more than a year) are taxed at preferential rates. Short-term gains are taxed as ordinary income, which for most active traders is a meaningfully higher rate. HIFO and LIFO both tend to reach for recently acquired lots, and recent lots are the ones most likely to still be short-term. So you can engineer the smallest gain and still hand over more tax than FIFO would have, because FIFO sold an older lot that qualified for the long-term rate.

A quick feel for it: a $200 short-term gain at a 32% ordinary rate is $64 of tax. A $520 long-term gain at 15% is $78. Close enough that the "obvious" lowest-gain choice is not obviously the cheapest once the holding period enters the math. You have to look at gain and rate together, per lot, not just the headline gain.

Practical guidance

A few rules of thumb I actually use:

• Rising market, want to defer tax: HIFO usually gives the lowest current-year gain. Just remember you are pushing gain into the future, not erasing it.
• You hold long-term lots you want to cash out at low rates: FIFO naturally surfaces your oldest (long-term) lots, which can be exactly what you want in a year where the long-term rate beats a small short-term gain.
• You want simplicity and the least audit friction: FIFO is the safe default. It is what the IRS assumes, it needs the least documentation, and it is the hardest to second-guess.

Two constraints to keep in mind across all of this. You generally have to apply your method consistently rather than cherry-picking per sale, and you have to be able to specifically identify the lots you sold, with records of basis and dates, to use anything other than FIFO. No records, no method choice. You are back to FIFO by default.

How fy_nance handles this

This is the kind of calculation that is easy to describe and miserable to do by hand across a year of Steam Market and third-party trades. fy_nance keeps lot-level basis for every skin you own, with the acquisition date attached to each lot. When you record a sale, it runs all four methods (FIFO, LIFO, HIFO, and average cost) and shows you the realized gain and the resulting Form 8949 entries side by side, so you can see the difference before you commit to anything. You get to weigh the lowest gain against the holding-period effect with the real numbers in front of you, instead of guessing and discovering the cost at filing time.

Pick the method that fits your year, document it, and apply it consistently. And when the stakes are real, run it past a CPA.