Before
mile one.
Millions of estimated pre-orders. The first day in Leonida is still ahead. What can that demand actually tell us?
READ THE NUMBERS ↓estimated pre-order units
across PlayStation & Xbox
Not a publisher-confirmed sales total.
Has GTA 6 passed
six million pre-orders?
According to Sensor Tower’s estimate, yes. The company’s October 13 tracker shows 6.08 million units across PlayStation and Xbox. That is an analytics estimate, not a sales announcement from Rockstar or Take-Two.
We opened the original tracker and read its embedded figures and methodology note. Its revenue total is $594 million. Both headline metrics are labelled net of returns. The page is updated weekly; this article preserves the October 13 snapshot so the numbers you are reading and the visualization agree.
The milestone is striking because the game has not reached its announced launch. Rockstar’s current GTA VI page lists November 19, 2026 for PlayStation 5 and Xbox Series X|S. The tracker describes demand ahead of that date. It does not measure launch-day players, completed downloads or people simultaneously exploring Leonida.
A pre-order story starts with anticipation. It becomes useful when the unit behind the headline stays visible. Our release-date and platform guide covers the schedule; here, the question is what this particular count can support.
Two platforms.
One snapshot.
Figures displayed by Sensor Tower on October 13. Estimated units and revenue, net of returns.
| Platform | Units | Revenue |
|---|---|---|
PlayStation | 4.51M | $441M |
Xbox | 1.57M | $153M |
| Combined | 6.08M | $594M |
The displayed platform shares are 74% and 26%. Calculating from the rounded unit totals gives approximately 74.2% and 25.8%; that extra decimal is our arithmetic, not extra precision in the underlying dataset.
Open the original tracker ↗Give the number
some room.
608 tickets. Each stands for 10,000 estimated units.
Separate the platforms. Stack the same total. Inspect a group.
A number you can walk around.
Original 3D data sculpture · opens on demandMODELLED ESTIMATES / 13 OCT 2026
Each ticket represents 10,000 estimated units. Original geometry; no customer records, game map or GTA VI assets. The arrangement is a visual explanation, not a sales timeline.
$594M estimated revenue
Select a ticket in the 3D view, or use the buttons below.
Nothing new is sold.
Switch from one floor to platform stacks. The arrangement changes; the represented total does not. Height and area are design choices, so read the count before comparing the shapes.
A group is not a person.
Select a ticket. Its number locates a piece of this illustration. It does not identify an order, a household, a buyer or a console. Those records are not available to us.
No hidden geography.
The overhead grid is a reference map of the installation. Its positions do not represent countries, cities, stores or parts of Leonida. Every ticket has the same numerical weight.
“Net of returns”
is not “profit.”
The tracker names a deduction: returns. That phrase alone does not establish every other adjustment in the figure, or show the publisher’s costs. We have no basis for relabelling the total as Take-Two’s profit, cash on hand or recognized accounting revenue.
Think of a receipt and a company’s accounts as different documents. Even a perfectly counted purchase would not tell you the cost of developing the game, how the transaction is accounted for, or what a retailer keeps. An estimated aggregate supplies less detail still. Our GTA VI budget investigation addresses the separate problem of unsupported production-cost claims.
“Units” also deserves to stay “units.” A reader may buy more than one version; a household may share access to a purchase. Neither possibility supplies a correction we can apply without data. The safe move is simply to keep the source’s unit of measurement rather than rewrite it as a unique-person count.
Nor does this console snapshot create a PC total. For the platform reporting, read our PC release and specifications guide alongside the Q1 2027 PC report. The existence of a sales tracker and the confirmation of a platform are different evidence questions.
Does $97.70 mean
everyone bought Ultimate?
No. Dividing two totals produces an average. It does not recover the purchases that made it.
$97.70 per estimated unit
There is an obvious temptation here. Rockstar’s June 24 announcement lists US prices of $79.99 for the standard game and $99.99 for Ultimate. An aggregate near the higher price can look like an edition-share answer. It is not enough information.
The tracker is worldwide. To infer an edition mix, we would need the relevant price inputs and coverage: regions, currency conversion, any discounts, tax treatment and the methodology behind the estimate. We do not have a transaction-level table that matches those inputs to the headline totals.
We can demonstrate the problem without pretending to reconstruct those transactions. In a deliberately simplified basket containing only two prices, the higher-price share would equal the average minus the lower price, divided by the price difference. Change the assumed prices and the answer changes—even though the average stays fixed.
One average.
Many possible baskets.
Keep the calculated average fixed at $97.70. Move both hypothetical prices. A different mix can produce the same average.
- Lower-priced copy
- $79.99
- Higher-priced copy
- $99.99
Higher-priced share required in this simplified example only.
($97.70 − $79.99) ÷ $20The calculation uses unrounded inputs. It omits regions, exchange rates, tax treatment, discounts and the tracker’s estimation method. It does not infer real edition sales.Read the worked example without the controls
Using the unrounded average of $594M ÷ 6.08M, a hypothetical $79.99/$99.99 basket requires about 88.5% at the higher price. A hypothetical $89.99/$109.99 basket requires about 38.5%. The two examples share the same average by construction. They are not estimates of actual edition sales. An average outside both prices cannot be produced by any mix of them.
The useful result is the disagreement between the examples. One average can fit more than one story when the prices and coverage are unknown. Our price and edition guide helps with what the products contain; this experiment explains why dividing a global dollar total is not a substitute for an edition breakdown.
These estimated pre-orders are not a count of simultaneous players. Our GTA VI multiplayer report separates the announced single-player release from an online launch and explains why this sales snapshot cannot reveal server capacity.
The small note
that changes the comparison.
Sensor Tower says it updated its GTA VI pre-order estimation model on October 5 to better reflect estimated country splits. It notes a small effect on global unit sales, revenue and the console mix. That note appears inside the same tracker as the headline figures.
This matters when an old screenshot becomes the starting point for a growth calculation. The difference between an earlier display and today’s display may contain both estimated purchasing activity and a methodological revision. Without a comparable historical series, subtraction does not separate those contributions.
Imagine counting tickets with one model on Monday and a revised model on Tuesday. The second result can be more useful without every difference representing a new ticket purchased overnight. A dated screenshot preserves what a page showed; it does not guarantee the two readings were produced on the same basis.
For that reason, our room is a snapshot, not an animated timeline of pre-order growth. We have not invented daily points between published observations, or projected the total forward to launch. A line that looks smooth would add confidence the evidence has not supplied.
THE CAMPAIGN EFFECT / WITH A DATE CAVEAT
A reveal can
change the rhythm.
There is earlier evidence that attention and estimated orders moved together. In its September premium-games report, Sensor Tower describes typical August daily console pre-order revenue of roughly $2.4–3.8 million, a $16.3 million level around the Extended Look, and about $5.9 million two weeks later.
The page is inconsistent about the spike’s exact day: its takeaway says the day of the presentation, while its heading and body say the day after. We are retaining that caveat instead of assigning a precise date from conflicting text. We also have not downloaded the full underlying series.
The broad observation is an increase around a large marketing moment. It does not isolate the presentation’s causal effect from everything else happening then, and it does not provide a rule for predicting the next reveal. We should not reuse its September baseline to manufacture an October daily-sales chart.
CAMPAIGN ARCHIVE / EARLIER OFFICIAL FOOTAGE
Earlier campaign context. This is Trailer 2, not the August Extended Look used in Sensor Tower’s September analysis. Its original release-date messaging has been superseded.
Watch on YouTube ↗A big number.
Your own decision.
Does this decide my edition?
A popularity estimate cannot tell you whether the extra items matter to you. Compare the actual inclusions and your own price before choosing.
Compare editions ↗None of the totals decides whether you should pre-order. That depends on what you want, the edition you are considering and the terms shown by your own seller. A crowd estimate cannot inspect your order confirmation or tell you whether a bonus is attached to your account.
Our pre-order and code-redemption guide covers the practical route from purchase to access. The boxed-copies report explains why the format deserves its own check. Use those details for a purchase decision; use this tracker for a dated view of estimated demand.
Read every branch of the question map
Editions: compare inclusions rather than treating estimated popularity as a recommendation. PC: console figures do not confirm a PC date. Boxes: establish coverage before adding retailer and platform totals. Storage: orders are not downloads; check preload and file-size information separately.
WHAT WOULD MAKE THIS REPORT STRONGER?
Bring a clearer
measurement.
A publisher-issued unit total would be a different source. A transparent coverage note would clarify what the model includes. A historical series restated on one methodology would support cleaner comparisons. An actual edition split would answer the question our price experiment deliberately leaves open.
Until those details are established, the story is substantial enough on its own: an analytics company’s latest public snapshot places GTA VI above six million estimated pre-orders before launch. Keeping “estimated” in that sentence does not make it less interesting. It tells readers what they are looking at.