Most collectors misread KAWS prices in exactly the same way, and it costs them money. They open a marketplace, sort by “sold,” see a single number — say, a Companion that “went for $3,600” — and anchor their entire sense of value to that one figure. Then they either overpay to match it or walk away from a fair deal because the real market sits far below that headline. The number they latched onto was almost never the market. It was one data point: possibly an outlier, possibly a hype spike, possibly a mislabeled or damaged piece, possibly a private sale that has nothing to do with what a clean example trades for on a normal Tuesday.
Reading a comp is not reading a price. It is reading a distribution — a spread of real sales, weighted by how many of them exist, adjusted for condition, completeness, and the noise that always contaminates a thin market. A professional buyer never looks at a single sold listing and calls it value. They look at four things together: the median observed sale, the value range (value_low to value_high), the number of real sold comps, and a confidence score that tells them how much to trust the first three. Only when those four numbers agree does a “comp” become a decision you can stand behind.
This is a buyer’s guide to doing exactly that. Every figure in this article comes from the Gauntlet Gallery KAWS index — a catalog of 214 catalogued releases spanning a value range from $209 to $68,000, with a median release value of $500 and a mean of $1,983. We’ll walk through the four data fields one at a time, explain our confidence scoring in plain language, and then run four fully worked examples — one from each confidence tier — using real numbers from the index. By the end you’ll have a reusable checklist you can apply to any KAWS figure you’re about to buy, sell, or insure, and you’ll know precisely when a “comp” deserves your trust and when it deserves your skepticism.
If you want to see how these numbers map onto specific releases as you read, keep the KAWS Figurine Index open in another tab. It’s the same dataset the examples below are drawn from.
Why a Single Sold Price Is Almost Never the Answer
Start with the failure mode, because understanding it is half the battle. The secondary market for designer toys is thin relative to equities or even fine-art prints. A given KAWS colorway might trade a few dozen times a year in the open, tracked market — and many trade far less often than that. When a market is thin, every individual sale carries enormous leverage over your perception, and that leverage cuts both ways.
Consider what a single sold price actually contains. It bundles together: the specific condition of that specific unit, whether the box and authentication materials were present, whether the seller shot good photos, whether two determined bidders happened to collide on the same night, whether the listing was mistitled and got found by only one buyer, whether it was a genuine arm’s-length sale or a wash between friends, and whether the piece sold into a rising or falling market mood. None of that context travels with the number when you copy it into your head. You just remember “$3,600,” and that ghost figure starts driving your decisions.
Take a real example from our index. The No Future Companion — Black Chrome (2009) carries a current fair value of $3,600 in the catalog. But look at what the actual sales say: the observed range runs from a value_low of $531 to a value_high of $1,599, across 8 recorded comps, with a median observed sale of $1,250. The headline “value” and the median sit thousands of dollars apart. A buyer who anchored to $3,600 would either overpay wildly or dismiss every fairly priced example as “too cheap to be real.” The median — the middle of the actual transactions — tells a very different, and far more actionable, story. And crucially, our confidence label on that release is low, which is the index’s way of flashing a caution light: don’t treat any of these numbers as gospel.
That gap between a single number and the truth is why we built the four-field framework. Each field answers a different question, and only together do they describe the market.
The Four Numbers That Actually Matter
Every release in the index carries the same core pricing fields. Here’s what each one means and what question it answers.
Median Observed Sale — “What does the middle of the market actually pay?”
The median observed sale is the middle value of all recorded public sales of $200 or more for that release. Half the sales came in above it, half below. This is the single most useful number in the entire record, and it is not the same as the average.
Why the median and not the mean? Because thin, noisy markets are full of outliers, and the mean (average) gets yanked around by them. One crazy $3,600 sale in a pool of otherwise $700 sales will drag the average up by hundreds of dollars, painting a picture of a market that doesn’t exist. The median ignores the magnitude of the outliers and just asks: what’s the middle transaction? That resistance to outliers is exactly what you want when you’re trying to find the “normal” clearing price for a clean example.
When a dealer quotes you a fair price, the median observed sale should be the spine of that quote. Everything else — condition, box, COA, current momentum — is an adjustment on top of the median, not a replacement for it.
Value Low / Value High — “How wide is the spread, and why?”
The value_low and value_high fields bracket the observed sale range: the cheapest and the most expensive real sales on record. This pair is your volatility gauge. A tight range means the market agrees with itself; a wide range means something is introducing variance — and your job is to figure out what.
A wide value range almost always comes from one of three sources: 1. Condition variance — mint-in-box examples selling for multiples of loose, played-with, or damaged ones. 2. Completeness variance — pieces with original box, hang tag, and authentication trading well above naked figures. 3. Outliers and hype spikes — one or two euphoric sales (or one desperate fire-sale) sitting far from the pack.
The range never tells you the answer by itself. It tells you a question exists. When value_low and value_high are far apart, you must find out which of the three causes is driving it before you trust the median — because if the spread is condition-driven, then your piece’s condition determines where in that range you should land.
Sold Comp Count — “How much data is this even built on?”
The sold_comp_count is the number of real, recorded sold comps behind the record. This is the field most amateurs ignore entirely, and it’s the one professionals check first. A median built on 40 sales and a median built on 1 sale are not the same kind of object, even if they’re the same dollar figure. The first is a market. The second is an anecdote wearing a market’s clothes.
Comp count is the raw material of confidence. The more genuine arm’s-length sales feed a record, the more the median and range converge on something real and the less any single outlier can distort the picture. Low comp counts don’t make a number wrong — they make it fragile. Treat a one-comp median the way you’d treat a stock price set by a single trade: directionally interesting, but not a market.
Confidence Label — “How much should I trust the three numbers above?”
Finally, the comp_confidence_label — high, medium, low, or insufficient — is our summary judgment of comp depth. It exists so you don’t have to eyeball the comp count and reverse-engineer reliability yourself. It reflects, above all, how many real sold comps stand behind the record. High means many comps and a reliable picture. Insufficient means a single sale or near-zero data — treat it with real caution. It is the meta-number: the number that tells you how much to trust the numbers.
How Gauntlet Gallery’s Confidence Scoring Works
Let’s be explicit, because this is the heart of the framework. Our confidence label is fundamentally a measure of comp depth — the number of genuine sold comps behind a release. Across the full 214-release index, the labels break down like this: high: 27 releases, medium: 21, low: 54, insufficient: 112. That distribution is itself a lesson. More than half the catalog (112 of 214) sits in the insufficient tier. The KAWS secondary market is, for most individual releases, data-poor. That is not a flaw in the data — it is the reality of the market, and pretending otherwise is how buyers get hurt.
Here is how to read each tier as a buyer:
| Confidence label | What it means about comp depth | How to treat the numbers | Buyer posture |
|---|---|---|---|
| High | Many real sold comps; median and range are stable and reliable | Trust the median as a genuine clearing price; the range reflects real condition/completeness spread | Buy or sell with confidence near the median; negotiate on condition |
| Medium | A solid but not deep pool of comps; picture is credible but softer | Use the median as a strong anchor, but expect some drift; verify with current listings | Reasonable to transact near median; leave a margin of safety |
| Low | Few comps; the median is directional, not definitive | Treat the median as a hypothesis; the range may be dominated by one or two sales | Proceed carefully; demand condition/authentication proof; don’t chase |
| Insufficient | A single sale, near-zero data, or value derived indirectly | Do not treat any figure as “the market”; it’s a placeholder, not a price | Maximum caution; price on fundamentals, condition, and dealer judgment — not the comp |
The single most important discipline this table encodes: confidence gates how you use everything else. A $1,500 median at “high” confidence is a price. The identical $1,500 median at “insufficient” confidence is a guess. Same dollar figure, completely different decisions. Amateurs treat those two numbers as interchangeable. Professionals never do.
Now let’s prove it with four real releases — one per tier.
Worked Example 1 — HIGH Confidence: Accomplice (Original) — Pink
Our high-confidence example is the Accomplice (Original) — Pink (2002), the pink rabbit-eared Companion cousin that anchors the early-editions market. Here’s the full record:
- Current value: $1,860
- Value range: $350 (low) to $3,050 (high)
- Median observed sale: $1,575
- Sold comp count: 21
- Confidence: high
Twenty-one real sold comps is a genuine market. With that depth, the median of $1,575 is trustworthy — it’s the middle of a substantial, repeated set of transactions, not a one-off. When a dealer prices a clean Accomplice Pink, the $1,575 median is the honest spine of the quote, with the $1,860 current value reflecting where a strong, complete example sits at the top of the recent trend.
Now read the range like a pro. That spread — $350 all the way to $3,050 — is enormous, nearly a 9x gap from bottom to top. Does a wide range at high confidence mean the number is unreliable? No. It means the opposite: with 21 comps, you can trust that the range is real and explainable, not an artifact of thin data. This is a twenty-plus-year-old vinyl figure. The $350 sales are almost certainly loose, played-with, sun-faded, or box-less examples. The $3,050 sales are pristine, complete, well-documented pieces. The median sits comfortably in the middle because most sales are ordinary — some wear, maybe the box, maybe not.
The buyer lesson: at high confidence, a wide range is a menu, not a warning. It tells you exactly how much condition and completeness are worth in this specific market — here, the difference between a rough example and a mint one can be several thousand dollars. So your entire negotiation should be about establishing where your specific piece sits between $350 and $3,050. If you’re buying near the $1,575 median, you should expect a solid, presentable, reasonably complete figure. If you’re being asked to pay near $3,050, you should demand mint condition with full documentation — and if it’s not mint, that price is unjustified regardless of what the “high” sale was.
Worked Example 2 — MEDIUM Confidence: Dissected Milo — Brown
Step down a tier. The Dissected Milo — Brown (2011) is our medium-confidence example:
- Current value: $932
- Value range: $200 (low) to $2,000 (high)
- Median observed sale: $625
- Sold comp count: 13
- Confidence: medium
Thirteen comps is a real but shallower pool than the 21 behind the Accomplice. The picture is credible — you can believe the shape of this market — but you should hold it a little more loosely. Notice that the median observed sale ($625) sits below the current value ($932). That’s an important tell: it suggests the release has firmed up recently, with the current fair value pricing in an upward trend that the trailing median hasn’t fully caught up to. At medium confidence, you treat that gap as a signal to verify, not to assume. Is the market genuinely rising, or is the current value optimistic? Pull up live listings and recent sales to confirm before you pay toward the $932 end.
The range here — $200 to $2,000 — is again very wide, a 10x spread. But at medium confidence you can’t lean on it as hard as you could with the Accomplice. With only 13 comps, that $2,000 high could be one genuinely exceptional sale (mint, boxed, lucky night) rather than a repeatable ceiling, and the $200 low could be a single damaged or incomplete unit. So you read the range as plausible boundaries rather than a precise condition ladder.
The buyer lesson: at medium confidence, anchor to the median ($625) as your strong reference point, treat the current value ($932) as a rising-market hypothesis worth verifying, and build in a margin of safety. A fair transaction lives in the zone between the median and the current value, tilted by condition and by what live listings are actually doing right now. Don’t pay the top of the range on the strength of one or two comps you can’t fully explain.
Worked Example 3 — LOW Confidence: Chum (Original) — Clear
Now the terrain gets rockier. The Chum (Original) — Clear (2002) is our low-confidence example:
- Current value: $399
- Value range: $285 (low) to $1,200 (high)
- Median observed sale: $399
- Sold comp count: 7
- Confidence: low
Seven comps is thin. The median ($399) and current value ($399) agree, which is reassuring, but seven sales is not enough to call this a settled market — it’s directional. And look hard at that range: $285 to $1,200. The high is more than four times the low, and more than three times the median. That $1,200 figure is a classic outlier signature. In a seven-comp pool, a single euphoric or unusual sale — a rare crystal-clear example, a bidding war, a mislabeled listing that a determined collector found — can create a value_high that has almost nothing to do with what the next clean Chum Clear will actually fetch.
This is where the discipline of ignoring the high number pays off. An amateur sees “$1,200” in the range and starts dreaming — or worse, prices their own Chum Clear at $1,200 and then can’t understand why it won’t sell. A professional sees a $399 median against a $1,200 high across only 7 comps and concludes: the ceiling is noise; the middle is the market. The realistic clearing price for a normal example is right around $399, and the $1,200 sale is not a target — it’s a warning about how much variance a thin market can generate.
The buyer lesson: at low confidence, treat the median as a hypothesis and the high end of the range as suspect until proven otherwise. Demand condition and authentication proof before paying anywhere near the top of the range, and never use an outlier high as your anchor. If someone’s asking near $1,200 for a Chum Clear, the burden is entirely on them to prove — with a specific, exceptional, fully documented example — why this one deserves to be the outlier. Absent that proof, you’re buying a $399 figure, full stop.
Worked Example 4 — INSUFFICIENT Confidence: Companion (Original) — Brown
Finally, the tier where more than half the index lives. The Companion (Original) — Brown (1999) — the piece that started it all, the first Companion — is our insufficient-confidence example:
- Current value: $1,500
- Value range: $1,500 (low) to $1,500 (high)
- Median observed sale: $1,500
- Sold comp count: 1
- Confidence: insufficient
This is the purest illustration in the whole framework. Every number is $1,500 — value, low, high, and median all identical — because they are all the same single sale. The range has zero width not because the market is perfectly stable, but because there’s only one data point, and a single point has no spread. The comp count of 1 is the tell, and the “insufficient” label is the index refusing to pretend otherwise.
A one-comp record is not a market. It is an anecdote. That $1,500 could be a fair figure, or it could be high, or low — we genuinely cannot know from a single sale, especially for a landmark 1999 piece where condition and provenance swing value dramatically and clean examples surface rarely. The flat, zero-width range is not a sign of a calm, agreed market; it is a sign of no market data at all. This is the single most dangerous trap in comp reading, because a zero-width range looks superficially like the most confident number on the page — when it’s actually the least.
The buyer lesson: at insufficient confidence, do not treat any figure as “the market.” Do not let a flat range fool you into false confidence. Price on fundamentals instead — the release’s historical importance, its rarity, its condition, its documentation, and the judgment of a dealer who actually handles the category. For a first-generation 1999 Companion, that means leaning hard on authentication and provenance rather than a comp that doesn’t really exist. The number is a placeholder, not a price. When you see the numbers all stack to the same figure with a comp count of 1, your reaction should be more skepticism, not less.
What Deep-Comp Releases Teach Us: The Open Editions
The four worked examples above span the confidence tiers, but they share a limitation: none of them is drawn from the deepest end of the data. To really understand what “high confidence” can look like, it’s worth studying the releases in our index with the most comps of all — the open editions — because they show you what a mature market looks like, and they sharpen your intuition for everything below them.
Consider the Companion Flayed (Open Edition) — Black (2016). It carries a current value of $595, a median observed sale of $600, a value range from $200 to $1,300, and — remarkably — 105 sold comps, the deepest record in the entire catalog, at high confidence. One hundred and five real arm’s-length sales is not a market you have to squint at. It’s a market you can practically set your watch by. When the median ($600) and the current value ($595) sit essentially on top of each other and 105 sales back them up, you are looking at about as reliable a clearing price as the KAWS secondary market ever offers. And notice the range still runs wide — $200 to $1,300 — which reinforces the earlier lesson: even in the most liquid, best-documented release we track, condition and completeness produce a genuine spread. The $200 sales are rough or naked; the $1,300 sales are pristine and complete. The 105-comp depth is what lets you trust that the median, not either extreme, is the real market.
Now compare its sibling colorways. The Companion Flayed (Open Edition) — Brown (2016) shows a $452 value, a $430 median, a $200–$1,400 range, and 86 comps at high confidence. The Grey version shows a $440 value, a $400 median, a $218–$700 range, and 38 comps, also high. Line those three up and a pattern emerges that every KAWS buyer should internalize: within a single release, colorway matters, and comp depth varies by colorway. Black is the deepest and most valued ($600 median, 105 comps); Brown sits just below ($430 median, 86 comps); Grey is shallower and cheaper ($400 median, 38 comps). All three are high-confidence — but “high confidence” at 105 comps is a different animal than “high confidence” at 38. The label is a floor, not a ceiling: within the same tier, more comps still means more certainty.
This is why you never let a comp for one colorway stand in for another. If someone quotes you the Black Flayed’s numbers while selling you a Grey, they’re using the deepest, most-valued colorway’s data to price a shallower, cheaper one. The index keeps them separate precisely because the market does.
The regular open-edition Companions tell the same story at a lower price point. The Companion (Open Edition) — Black (2016) posts a $340 value, a $325 median, a $200–$599 range, and 61 comps at high confidence. Its Brown counterpart shows $360 value, $350 median, $218–$750 range, 42 comps, high. The Grey: $335 value, $300 median, $200–$550 range, 46 comps, high. Three deeply-comped, high-confidence records clustered tightly in the $300–$360 zone. When a whole family of colorways agrees this closely and each is backed by 40-plus sales, you are looking at a genuinely settled sub-$400 market — the accessible tier of KAWS collecting, where the data is thick and the surprises are few. These are the releases where a buyer can transact near the median with real confidence and spend their energy entirely on condition and authentication rather than on second-guessing the price.
The takeaway from the deep end: the more comps a release has, the more the median, the current value, and the sibling colorways converge — and the more you can trust all of them. Once you’ve seen what a 105-comp or 61-comp market looks like, a 7-comp or 1-comp record can never fool you again. You’ve calibrated your eye to what real depth looks like, and everything thinner reveals itself for what it is.
A Note on the Anniversary and Chum Families: Reading a Cohort
There’s one more reading skill worth building before we consolidate: reading a cohort of related releases together, because patterns across a family are often more trustworthy than any single record within it.
Take the Chum (2022, 20th Anniversary) colorways, which are unusually well-documented as a group. The Black posts a $364 value, $349 median, $225–$380 range, 25 comps, high. The White: $328 value, $305 median, $200–$798 range, 47 comps, high. The Orange: $348 value, $327 median, $225–$500 range, 22 comps, high. The Yellow: $298 value, $260 median, $212–$380 range, 24 comps, high. And the Pink, slightly shallower: $420 value, $400 median, $250–$600 range, 11 comps, medium.
Read as a cohort, this family tells a coherent story: a cluster of high-confidence releases with medians in the $260–$349 band, tight-ish ranges, and one slightly pricier, shallower Pink standing a little apart. When four or five sibling colorways all land in the same neighborhood with deep comp counts, that neighborhood is the market for the cohort, and any individual quote that strays far from it should be interrogated. The Pink’s higher $400 median with only 11 comps (medium) is exactly the kind of figure you’d verify before paying up — it might reflect genuine scarcity, or it might be a shallower record that hasn’t fully settled.
Contrast that with the original 2002 Chum family, which is far thinner and mostly low or insufficient: the White at $454 value / $454 median / $222–$999 / 14 comps / low; the Yellow at $428 / $428 / $288–$1,875 / 14 comps / low; the Clear (our low-confidence worked example) at $399 / $399 / $285–$1,200 / 7 comps / low; and the Red at $392 / $392 / $299–$700 / 0 comps / insufficient. Notice how the original 2002 Chums carry wilder value_high figures — $1,875 on the Yellow, $1,200 on the Clear — against medians in the low-$400s. Those towering highs across low-comp records are exactly the outlier signatures we warned about. Read the original Chum family as a cohort and the lesson is unmistakable: the medians cluster sensibly in the $390–$454 range, while the highs are scattered noise that you should not anchor to.
The cohort-reading skill, distilled: when you can see a family of related releases at once, trust the cluster of medians and distrust the scattered highs. A single record can lie to you; a well-behaved cohort of medians rarely does. And when one member of a family carries a suspiciously high value_high on thin comps, the cohort itself tells you it’s an outlier — because its siblings don’t agree.
Putting the Four Examples Side by Side
Here’s the whole framework in one view. Read across each row and watch how the same kinds of numbers demand completely different postures depending on comp depth.
| Release (year) | Value | Range (low–high) | Median observed | Comps | Confidence | How to read it |
|---|---|---|---|---|---|---|
| Accomplice (Original) — Pink (2002) | $1,860 | $350 – $3,050 | $1,575 | 21 | High | Trust the median; wide range = real condition ladder to negotiate on |
| Dissected Milo — Brown (2011) | $932 | $200 – $2,000 | $625 | 13 | Medium | Anchor to $625 median; verify the rising current value; margin of safety |
| Chum (Original) — Clear (2002) | $399 | $285 – $1,200 | $399 | 7 | Low | Median is the market; the $1,200 high is an outlier — ignore it |
| Companion (Original) — Brown (1999) | $1,500 | $1,500 – $1,500 | $1,500 | 1 | Insufficient | One sale — an anecdote, not a market; price on fundamentals |
The pattern is unmistakable. As comp depth falls from 21 to 1, the numbers get less trustworthy even when they look cleaner. The insufficient row has the tidiest, most agreeable-looking figures on the entire table — everything equals $1,500 — and it is the row you should trust the least. The high row has the messiest, widest range — and it is the row you can trust the most. That inversion is the entire point of confidence scoring, and internalizing it is what separates a buyer who reads data from a buyer who gets read by data.
The Traps — and How Each Field Defends You
Now let’s name the specific ways comp reading goes wrong, and which of the four fields protects you against each.
Trap 1: The Thin-Comp Illusion
The most common trap: mistaking a low-comp median for a real market price. Defended by the comp count and confidence label. Whenever a figure looks authoritative, check the comp count first. If it’s in the low single digits — or worse, exactly 1 — downgrade your trust immediately no matter how clean the number looks. The Companion (Original) — Brown’s four-way $1,500 tie is the poster child: seductive on the surface, hollow underneath.
Trap 2: Condition Variance Masquerading as Volatility
A wide range panics inexperienced buyers into thinking a market is unstable or unpredictable. Often it’s nothing of the kind — it’s condition and completeness doing exactly what they should. Defended by reading the range against the confidence label. At high confidence (Accomplice Pink, $350–$3,050 across 21 comps), a wide range is a reliable condition ladder you can navigate. At low confidence (Chum Clear, $285–$1,200 across 7), the same width is untrustworthy noise. Same-looking spread, opposite meaning — the confidence label is what tells them apart.
Trap 3: The Outlier / Hype Spike
One euphoric sale becomes the number everyone quotes. Defended by the median and by cross-checking the value_high against comp count. The median structurally ignores outlier magnitude — that’s its whole job. When you see a value_high that towers over the median (the Chum Clear’s $1,200 vs. $399 median), and the comp count is thin, treat that high as a spike, not a ceiling. Never anchor to the most exciting sale; anchor to the middle one.
Trap 4: The Zero-Width Mirage
A range where low, high, and median are all identical feels like maximum certainty. It’s usually maximum ignorance — a single data point with nothing to spread against. Defended by the comp count, always. A zero-width range with one comp (Companion Original Brown) is the least reliable figure on the page, not the most. Reflexively check comp count whenever a range collapses to a single value.
Trap 5: Ignoring Momentum in the Value-vs-Median Gap
When current value sits well above the trailing median (Dissected Milo Brown: $932 value vs. $625 median), that gap encodes a market that may be moving. Defended by reading value and median together and verifying with live data. A large gap is a prompt to check whether the market is genuinely rising before you pay toward the higher figure — not an invitation to assume it is.
The Reusable Buyer’s Checklist
Here is the framework distilled into a checklist you can run on any KAWS figure before you transact. Screenshot it, print it, tape it to your monitor.
- Find the comp count first. Before you even look at the price, ask how many real sold comps back it. This single number sets your trust ceiling for everything that follows.
- Read the confidence label as a gate. High → trust the median as a clearing price. Medium → anchor but verify. Low → treat the median as a hypothesis and distrust the range’s extremes. Insufficient → do not treat any figure as “the market.”
- Anchor to the median, never the high. The median observed sale is your spine. The value_high is the most likely place to find an outlier or hype spike. Never let the exciting number become your anchor.
- Interrogate the range. Ask why value_low and value_high are far apart. Condition? Completeness? Outliers? At high confidence the answer is usually a real condition ladder you can negotiate on. At low confidence it’s usually noise.
- Mind the value-vs-median gap. If current value sits well above the trailing median, the market may be moving — verify with live listings before paying up.
- Distrust zero-width ranges. When low, high, and median are all identical, check the comp count. If it’s 1, you’re looking at an anecdote, not a market.
- Adjust for condition, box, and COA last. Only after you’ve established where the median and confidence put the market do you layer on the premium or discount for this specific piece’s condition and completeness.
Run those seven steps and you will never again anchor to a ghost figure. You’ll know, for every piece, whether the number in front of you is a price or a guess.
How Comps Interact With Condition, Box, and COA
A comp is a starting line, not a finish line. The final number for a specific piece is the median-and-confidence read adjusted for that piece’s condition, completeness, and authentication. This is where the wide ranges in our examples become actionable rather than abstract.
Condition is the single biggest swing factor, and it’s usually what’s driving a wide value range. Mint-in-box examples cluster near the value_high; loose, faded, scuffed, or repaired examples cluster near the value_low. The Accomplice Pink’s $350-to-$3,050 span is condition and completeness made visible. So when you assess a piece, you’re not asking “what’s the value” — you’re asking “where in this documented range does this example belong, given its condition?”
Box and completeness move value materially. Original packaging, inserts, hang tags, and any release-specific materials push a piece up the range. A naked figure — even a mint one — sits lower than a complete set. For many releases the box is not a nicety; it’s a meaningful fraction of the value, and its absence is a real discount, not a rounding error. If you want to understand exactly what “complete” looks like for a given release, our KAWS Packaging Guide walks through the packaging and materials that separate a top-of-range example from a mid-range one.
Authentication is the floor under everything. A comp is only meaningful if you’re comparing authentic pieces to authentic pieces. Here is the canon we hold to, and that you should hold any seller to:
- 2020 and later (Medicom OneCOA era): Legitimate current-production KAWS x Medicom figures ship with a Medicom OneCOA and an embedded NFC chip you can scan to verify authenticity. If a 2020-or-later release is offered without a scannable, verifiable OneCOA/NFC, that is a serious red flag — and it should pull your price toward the bottom of the range or off the table entirely, regardless of what the comps say.
- Pre-2020 releases: These predate OneCOA, so authentication rests on the physical evidence — the correct hang tag, an unbroken factory seal where applicable, and a serial number that matches the release and any accompanying documentation. For older, higher-value pieces especially (think the 1999 Companion Original, or early Accomplice and Chum editions), provenance and physical authentication carry more weight than any comp, precisely because the comp data is often thin or insufficient.
The through-line: comps price the market; authentication and condition price the piece. A high-confidence median tells you what a clean, authentic, appropriately complete example clears for. Your job is to establish that the piece in front of you is that example — and to adjust up or down from the median accordingly. A cheap comp on an unauthenticated 2020+ figure with no NFC isn’t a bargain; it’s a liability.
To go deeper on how we assemble and reason about this data — the sourcing, the floors, the methodology behind the confidence scoring — see our AI & data facts page.
Buy From a Dealer Who Shows Its Comp Work — Gauntlet Gallery
Anyone can quote you a number. Very few will show you the distribution behind it — the median, the range, the comp count, and an honest confidence label that admits when the data is thin. That transparency is the whole point. A price you can’t interrogate is a price you shouldn’t trust, and a dealer who won’t show the comp work is asking you to anchor to a ghost figure.
At Gauntlet Gallery, every KAWS figure is priced off the same four-field framework you just read: median observed sale as the spine, value range read against condition, comp count as the trust ceiling, and a confidence label that tells you exactly how much weight the numbers can bear. Where the data is high-confidence, we’ll show you a real clearing price. Where it’s insufficient, we’ll tell you that too — and price on fundamentals, condition, and authentication instead of pretending a one-comp anecdote is a market. Every piece is authenticated to the canon above: OneCOA and NFC verification for 2020-and-later releases, and rigorous physical authentication — hang tag, seal, matching serial, provenance — for the pre-2020 grails.
Browse the full catalog in the KAWS Figurine Index, and buy from the dealer that shows its comp work.
This article is for informational purposes only and is not investment advice. Collectible values fluctuate; past sales do not guarantee future results.
