Cost Basis: The Number That Decides If You're Actually Up

Ask a skin holder what their inventory is worth and they'll quote a number in two seconds. Ask what they paid for it, all-in, and most go quiet. Cost basis — your true average purchase price per item — is the only number that separates feeling rich from being up, and almost nobody tracks it properly.

Cost Basis: The Number That Decides If You're Actually Up
Cost Basis: The Number That Decides If You're Actually Up · source: cdn.dribbble.com

Inventory value is a screenshot, not a scoreboard

Every tracker site will happily total your inventory at current Steam Market prices and hand you an impressive figure. That figure misleads in three separate ways. First, it's quoted in Steam wallet prices, which run well above what the items fetch in cash — and wallet dollars aren't dollars to begin with, since Valve never lets them out. Second, it ignores the seller fee — roughly 13–15% on the Steam Market — that stands between the quote and any realized value. Third, and most fundamentally, it says nothing about what you paid.

Profit is a comparison between two numbers, and inventory value is only one of them. Without the other — your basis — "I'm up" is a vibe. Plenty of holders who bought aggressively into a hot market have inventories that look great in absolute terms and sit below their true average cost. The drawdown math is unforgiving about this: you can't manage what you never measured.

What a real cost basis has to include

The honest version of the number is: everything you actually spent, divided by every copy you actually hold. In the CS2 market that means catching details that a mental estimate always drops:

  • Per item, not per portfolio. A single blended average across five different cases tells you nothing actionable. You need "I hold 212 copies of this case at an average of X" for each line, because each case has its own price history and its own exit logic when the time comes to think about selling.
  • Per venue. A copy bought with Steam wallet balance and a copy bought with cash on a third-party marketplace are not the same trade, even at the same sticker price. Wallet money was already trapped; cash was not. Mixing them in one average hides which venue is actually treating you well — the theme of the Steam-versus-cash question.
  • Per currency, converted on the day. European marketplaces such as SkinBaron list in euros. If you log a €0.85 buy as "$0.85-ish," your basis drifts with every purchase. The conversion should happen at a dated rate, at buy time, and get written down — a small detail that compounds into real money across hundreds of fills.
  • Fees and friction included. Deposit fees, payment-processor cuts, and any premium you paid for convenience belong in the basis. They were part of the price of owning the item.

How averaging actually works when you buy daily

Daily accumulation makes basis tracking simultaneously more important and easier to automate. Say you buy one copy of a case most days. Some days it fills at $0.78, some at $0.84, one week a spike takes it to $0.95 and your price cap skips the buy entirely. Your basis is the running weighted average of every fill — and the skipped days matter too, because a cap that refuses expensive fills is quietly defending your average.

FillCopiesPriceRunning basis
Day 14$0.80$0.800
Day 24$0.74$0.770
Day 3skipped (cap)$0.770
Day 45$0.71$0.747

Illustrative numbers, but the shape is exactly what a real hundred days of DCA produces: the red days you hated buying through are the ones that pulled the average down. Dollar-cost averaging only earns its name if someone is actually computing the average — otherwise it's just "buying often."

Memory is not a ledger

The failure mode is universal. You start with a mental note, graduate to a spreadsheet, skip a week on holiday, backfill from fuzzy marketplace history pages, and within three months the spreadsheet is fiction. Marketplace purchase histories don't help much: they paginate awkwardly, quote in mixed currencies, and disappear behind account issues. If the record isn't written at the moment of purchase, by the thing that made the purchase, it decays.

This is the strongest practical argument for an append-only purchase ledger — every buy logged with timestamp, venue, item, quantity, price in original currency, the conversion rate used, and the resulting dollar figure. Append-only means rows are never edited after the fact, so the history can't quietly rewrite itself. The case for it is laid out in why ledgers beat memory, and it pays a second dividend every spring: a clean basis record turns tax season into a filter, not a forensic reconstruction.

What a performance report should actually compare

Once basis exists, a useful performance email writes itself, and it looks nothing like a tracker screenshot. Per item, it should show: copies held, total spent, average cost, a current reference price, and the spread between them — ideally with the honesty to note that the reference price is a listing price, not a guaranteed exit. Portfolio-wide, it should show total deployed capital against a conservatively-marked current value, so the one number you skim at breakfast is the real one.

That's the design cs2stack follows: every automated buy appends a money-exact row to a ledger committed to git, euro fills are converted at a dated rate, and the performance email compares your actual average to the market rather than flattering you with wallet-price totals. The three-email pattern exists precisely so that basis, fills, and balance each get their own honest report.

Cost Basis: The Number That Decides If You're Actually Up
Cost Basis: The Number That Decides If You're Actually Up · source: cdn.steaminventoryhelper.com

The discipline dividend

Knowing your basis changes behavior more than any market call. It makes corrections legible — spot dipping below your average is an accumulation signal for a DCA plan, not a reason to panic. It makes euphoria survivable, because "up 40% on basis" is a fact you can set exit rules against, while "my inventory looks huge" is not. And it makes you audit-proof to the most important auditor: yourself, six months from now, wondering whether this whole thing is actually working.

Track the number. Better yet, make the software that spends the money write it down for you — at purchase time, in a format nobody can fudge later.