On the Catwoman series relaunched in 2011, the count of announcements on August 29, 2026 gives two medians: $14.00 for the 44 copies offered without certification, $214.99 for the 6 certified copies. A ratio of more than fifteen to one. This report does not measure anything: it is based on six advertisements, a number where the removal or addition of two sellers would move the median by half. Always look at the number before the amount, and refuse to cite a median calculated on less than ten advertisements without giving the number in the same sentence.
There is a category of numbers that have all the appearance of a measurement without actually being one. They are exact – the calculation is correct, the data is real, no one cheated – and yet they describe nothing stable. The survey carried out on August 29, 2026 on the current announcements of the Catwoman series relaunched in 2011 produces a textbook case. On one side, 44 listings for uncertified copies, ranging from $1.95 to $160.00, median at $14.00. On the other, 6 listings of certified copies, from $68.00 to $399.00, median at $214.99. The ratio between the two medians exceeds fifteen. Written like this, in a table, with a decimal place and a dollar sign, the number has the authority of a result. He only has the costume.
The problem is neither the source, nor the method of calculation, nor the honesty of the statement. The problem comes down to one number: six. A six-value median is the average of the third and fourth sorted values. It is therefore entirely determined by two announcements — two sellers, two personal decisions, two moods of the moment. If one of the two had published their ad the following week, or had withdrawn it the day before, the median would have shifted to neighboring values. This is not a property of the market of the series: it is a property of the accidental composition of a very small group, on a given day. The figure exists, it is reproducible up to the calculation, and it would not survive a second reading.
What six ads can establish, and what they can't
Six announcements are not nothing. They establish an existence. As of August 29, 2026, it is true that there are certified copies of this series on the online market offered for three-digit amounts, and that the highest displayed reaches $399.00. It is a verifiable, dated, and useful fact: it tells you that this segment exists, that it is not theoretical, that a collector looking for a certified copy will find one. An existence is demonstrated with a single example. Six is more than enough.
What six ads can't establish is a level. “Certified copies are around $215” is a phrase that the data does not carry. It assumes a regularity – a point around which the values are distributed – which six observations do not allow us to observe. The difference between $68.00 and $399.00, on the contrary, shows a very wide dispersion within the small group: the extremes are separated by a factor greater than five. On such a small number, such dispersion means that the “middle” is not a middle, it is a crossing point between two announcements which are not similar.
Four ways to lose control of a six-value median
The withdrawal of two advertisements
Out of six values, the median is the average of the third and the fourth. Remove two listings — a sale, a seller who changes his mind, a deleted listing — and only four remain: the median becomes the average of the second and third values of what remains. The calculation point moves mechanically, without any asking price having moved a single cent.
L'ajout de deux annonces
Symmetrically, two new listings are enough to increase the workforce from six to eight and shift the median rank. If the two newcomers position themselves near the bottom of the range, the median falls; if they position themselves near the top, it climbs. Nothing about the state of the series has changed: two people simply decided to sell in the same week.
The day of the statement
An ad count is a photograph. That of August 29, 2026 shows six certified copies. There is no guarantee that a reading taken three days earlier or later would have shown six, or the same six. On 44 announcements, a delay of a few days modifies the margins; out of six, he can recompose the entire group.
The confusion asked/observed
The two ranges noted are amounts claimed by sellers, not concluded transactions. An ad for $399.00 reflects the hope of the person posting it; it does not prove that a buyer accepted this amount. On a large number of announcements, this imprecision is diluted in the mass. Out of six, it is the very material of calculation.
These four mechanisms are not pessimistic scenarios, they are the ordinary conditions of a second-hand market between individuals. Ads appear, disappear, change price, are republished. On the 44 uncertified copies, this back and forth also exists, but it is dampened: it would take a coordinated movement of several announcements to significantly move a median of $14.00 calculated on 44 values. Out of six, only one movement is enough. The difference between the two figures is therefore not a difference in the quality of measurement, it is a difference in nature: one is a fragile measurement, the other is not a measurement at all.
The wording “the median of certified copies exceeds that of non-certified ones by fifteen times” is technically accurate and practically unusable. It deserves to be replaced, systematically, by: “out of six certified advertisements noted on August 29, 2026, the median comes to $214.99”.
The second version says exactly the same thing, plus the number — and the number is precisely what allows the reader to judge whether the number means something. Any median cited without its number is an assertion from which the only information allowing it to be contested has been removed.
Why small numbers systematically produce spectacular differences
A group recruited by hope
No one gets a random copy certified. We undertake a costly process on a copy from which we expect a lot. The small group of six is therefore not a sample drawn at random from the population of copies: it is the subset selected by those who hoped the most. It is oriented upwards by construction, even before we calculate anything about it.
A two-step selection
The bias comes into play twice: once when choosing which copy to submit, a second time when setting the asking price. A seller who has invested in a certification process rarely displays a modest amount. The six ads combine both filters; the other 44 suffer none.
Nothing to cushion
In the big picture, a very optimistic announcement is surrounded by ordinary announcements that put it into perspective. In a set of six, it occupies one sixth of the total and counts in the ranking which determines the median. There are not enough neighbors for the atypical value to become atypical again.
A comparison between unequal groups
Comparing 6 to 44 is comparing a sorted group to an unsorted group. The ratio of more than fifteen measures upstream selection at least as much as anything else. It would still be spectacular even if the two populations were originally very similar.
The round number of the gap
“Fifteen times” is remembered, quoted, copied. A sentence like “the median of the six certified announcements noted that day” does not hold back. It is the least reliable formulation which circulates the best, and it is exactly for this reason that it must be refused entry into collector's reasoning.
False decimal precision
$214.99 seems more serious than “around $200” because it has pennies in it. These cents come from a seller's posted price, not a refined estimate. The precision of writing is never the precision of measurement, and out of six values the difference between the two is maximum.
Identify an uninterpretable statistic before using it
The check takes ten seconds and is done in the reverse order of spontaneous reading. We read the amount first because it is written in large letters; you must first look for the number, often written in small letters, sometimes not written at all. A figure presented without numbers must be treated as a figure whose numbers are too small to be shown — it is a presumption, not a certainty, but it is verified often enough to be considered a reflex.
Next, ask the displacement question: How much would the value move if two observations changed? Of the 44 uncertified listings, the answer is “barely.” Of the six certified, the response is “of an order of magnitude comparable to the gap we want to comment on”. When the sensitivity of the figure is of the same order as the phenomenon it claims to describe, the figure describes nothing.
Third check: ask yourself how the observations entered the group. A set constituted by a voluntary and costly process does not have the same meaning as a set constituted by simple presence on the market. The first provides information on the expectations of those who formed it, the second on what is proposed. They are two different objects, and comparing them directly produces a difference that is mistakenly believed to measure a value.
Fourth control, specific to ad counts: distinguish what is requested from what is concluded. The two ranges noted on August 29, 2026 — from $1.95 to $160.00 on one side, from $68.00 to $399.00 on the other — are amounts claimed. An ad can remain online for months without finding a buyer; However, it continues to weigh as heavily in the calculation as an announcement published the day before. On a large workforce, we can at least hope that the unrealistic announcements compensate for each other. Out of six, there is nothing to compensate with.
The practical rule: the number before the amount, without exception
The discipline is contained in one sentence, and it applies to everything: catalog, forum, article, personal spreadsheet. Before reading an amount, read the number of observations on which it is calculated. If this number is not indicated, do not read the amount. If this number is less than ten, you can read the amount provided you never quote it without it — in the same sentence, not in a footnote, not in a parenthesis at the end of a paragraph.
This rule does not prohibit the use of all six ads. She forbids making them say a level. They say that there are certified copies offered at three figures for this series: this is real information, which is enough to direct a search, to decide to go watch this segment, to understand that the question deserves to be asked. It is not enough to decide how much to pay, nor to value a copy you own, nor to conclude anything about a market movement.
Applied to the statement of August 29, 2026, it gives a clean statement: out of 44 announcements of uncertified copies, the median of the amounts requested is $14.00, with proposals ranging from $1.95 to $160.00; on 6 listings of certified copies, it is $214.99, between $68.00 and $399.00. The first number supports reasoning, the second supports an observation. Treating them as equals because they are written in the same table is the error that this statement best illustrates.
What to note
The certified median of $214.99 is based on six ads, and six is a count where two ads decide the outcome.This is not a margin of error to be mentally added to the figure: it is a figure whose central value itself is determined by two particular sellers on that day. Removing it from the reasoning does not impoverish anything, since it did not provide any stable information.
The ratio of more than fifteen between the two medians is a symptom of the small number, not a discovery in the 2011 series.A group of six formed by a voluntary approach will almost always produce an impressive gap compared to a group of 44 formed by simple presence on the market. The discrepancy is expected before even opening the statement; seeing it appear therefore does not confirm anything.
Six announcements establish an existence and never a level.They prove that a certified segment exists for this series, which is useful and verifiable. They do not allow us to write "around $215", a formulation which assumes a regularity that six observations scattered between $68.00 and $399.00 show nowhere.
Both ranges are amounts claimed as of August 29, 2026, not transactions.The maximum of $399.00 reflects a seller's hope; the minimum of $1.95 on the non-certified side reflects another sales strategy. An ad count describes what is on offer on a given day, which remains legitimate information as long as it is not disguised as an actual price.
The operational rule is to read the number before the amount, and never quote a median under ten advertisements without giving the number in the same sentence.This writing constraint is what prevents a fragile number from circulating as a solid number. It costs a few words and it prevents a compositional accident from becoming, from copy to copy, a reference value.
The calculation is possible and exact, that is not the question. Out of six sorted values, the median is the average of the third and the fourth: two announcements determine it entirely. Add two or remove two and the median rank shifts, which is enough to move the result significantly without any asking price having changed. A value so sensitive to the composition of the group cannot be presented as the description of a level.
They serve to establish an existence, which is already a lot. As of August 29, 2026, we know that for this series relaunched in 2011, certified copies are being offered at prices ranging from $68.00 to $399.00. It is a verifiable fact that indicates that a segment exists and is worth exploring. The limit is elsewhere: from this existence we cannot deduce a reference amount.
There is no magic threshold, but around ten constitutes a reasonable floor below which the result remains dominated by two or three observations. The safest rule does not relate to a threshold: it consists of always citing the number of employees next to the amount. A reader informed of the workforce can judge for himself; a reader who only sees the amount cannot.
More robust, yes, because it would take a movement of several simultaneous announcements to move it significantly. It nevertheless remains a median of amounts claimed on a given day, no transactions concluded, and it says nothing of an evolution. It supports reasoning where that of the six certified copies does not support any.
By recording the workforce, the date and the nature of the data at the same time as the amount: six announcements, August 29, 2026, amounts requested. Thus stored, the observation remains consultable without ever being able to be reread as an estimate. It is the lack of context, much more than the figure itself, which transforms a one-off statement into a false reference.
Keep the context with the number
An amount recorded without its number or date becomes unusable in a few weeks. My Comics Collection lets you keep your numbers, states, and observations in one place, with notes that explain where each number comes from—so your decisions are based on data you can still interpret six months from now.