Wizz Air
All You Can Fly — Daily Route Archive
X snapshots
Routes available per day · scrub or click the timeline
SuMoTuWeThFrSa
Snapshot
Covers departures between
Added today
Removed today
Cities served
Why this day looks the way it does
Likely drivers
Departure board
Routes for :
Legend
Route status
Available
New today
Dropped today
Season
Winter
Summer
Spring
Autumn
Wizz Air
All You Can Fly — Route Map Archive
all
drag to pan · scroll to zoom · click a city for its departures & arrivals
Using this map
- All-time shows the full history; switch to a date to see that day's 3-day booking window.
- Click a city to filter its network into departures, arrivals, or both.
- Each route's strip shows its last 30 days at a glance.
- Tap the legend button for what each line's color & thickness mean.
The all-time network · one-way routes
Cities
One-way routes ever
Backbone ≥90%
Still active
Top hubs · all-time one-way routes
Legend & how arcs are coloured
Backbone — 90%+ of snapshots
Frequent — 50–90%
Occasional — 10–50%
Rare — under 10%
Retired — no longer offered
Departures from selected city
Arrivals into it
Both directions of a pair, merged into one line
Wizz Air
All You Can Fly — Insights & Records
derived from 577 snapshots
The pass itself · how All You Can Fly works
Launched
Aug 2024
Announced 13 Aug 2024. This archive starts 25 Sep 2024, the first eligible flight date.
Annual price
€499
Billed annually.
Per-flight fee
Flat booking fee per segment, on top of the annual fee.
Each release
10,000 passes
Capped per phase. Now in its third edition, across 34 countries.
Booking window
72 hours
Seats open 3 days before departure — hence the rolling 3-day PDF snapshots.
Typical use
~9 flights / yr
Wizz Air's stated average. List refreshes daily at 07:00 CET.
Key rules & caveats
- Personal item only — bags cost extra.
- No domestic Italy flights on the pass.
- Two no-shows ends the membership.
- Auto-renews yearly (not for Israeli residents).
- No changes — rebook instead.
- 30-day withdrawal if unused.
- Seats not guaranteed.
- First in Europe — like Frontier's US pass.
Figures reflect Wizz Air's published terms and reporting through early 2026; pricing and coverage
vary slightly by phase.
Records from the archive · all figures computed from the data
Seasonal rhythm · when the pass is most generous
Shares are of every directed route in each season, grouped by the latitude of the
destination city: Nordic/Baltic (55°N+), Central Europe (45–55°N),
Mediterranean (36–45°N) and Southern — North Africa, the Gulf and the
Middle East (below 36°N).
The network's reach · what's actually bookable
Pairs where one direction is far more available than the
return. On the pass these are easy to fly out on but hard to fly back —
you may need a different return route. Bars show each direction's share of all 577 snapshots.
The longest routes that are also dependable — present in
at least half of all snapshots. Unlike a one-off longest-route record, these are real
options for covering big distances on the pass.
How the network changed
Reliability shifts — routes trending up or down lately
A raw day-to-day route diff is almost all noise here — the PDF's
rolling booking window advances by one day at a time and touches hundreds of routes
regardless of any real change. This instead compares each route's presence over the last
snapshots against the before
that, so what's left is routes that genuinely went from reliable to rare.
Same comparison, the other direction — routes that went from rarely
available to showing up on most recent days.
Wizz Air
All You Can Fly — Route Count Forecast
experimental model
Read this first. This is an extrapolation, not a promise. The archive covers about
20 months — barely one full seasonal cycle — and the program grew and absorbed the Abu Dhabi
base closure during that time, so seasonal and trend effects are partly entangled. The model
assumes recent patterns simply continue: it cannot foresee new bases, route cuts, price changes
or another shock. It is most trustworthy 1–2 weeks out and the uncertainty band widens
fast beyond a month. Treat the monthly numbers as a rough sketch, not a fact.
Projected daily route count
History & forecast — routes available per day
Day-by-day — the reliable horizon
The next 14 days, where the model performs best. Each row is a
point estimate with an ~80% band. Weekend departure windows (amber) run higher because
they surface more leisure routes.
How accurate is it? · honest backtest
Before trusting any forward number, here is how the same model performed when asked
to predict days it had never seen. For each horizon the model was retrained on data up to
a cut-off, then scored against the actual count that many days later — repeated across the
last 90 days of the archive (walk-forward validation). The model's error is shown next to a
naive baseline — simply repeating the same weekday's count from the previous week —
so you can see whether the modelling actually earns its keep.
How the model works
Three multiplied components
A day's predicted count is trend × month factor × weekday factor.
The trend is a straight line fitted to the most recent ~8–9 months only — the
post-Abu-Dhabi regime — capturing steady network growth. The month and
weekday factors are multipliers measured from how far each month and weekday sat
above or below that trend line.
Weekday factor — solid signal
Backed by 80+ observations per weekday. Monday and Sunday departure windows
run highest, Thursday lowest.
Month factor — shrunk on purpose
With only one or two observations per calendar month, raw monthly figures
are noisy. Each month factor is pulled toward its season average (winter / spring
/ summer / autumn), which has far more data behind it — trading a little detail for
stability.
Uncertainty band
The shaded band is not a guess — it is the model's own backtest error
at each horizon. The 80% band spans roughly ±1.3× the typical miss; it widens with the
square root of how far ahead you look, because errors compound.
Recent-actuals anchor
A pure trend model snaps back to its trend line and ignores that the last
few actual days may sit above or below it. The forecast adds a
recency-weighted recent residual — how far the latest days landed from the model —
fading out over about three weeks.
Method. Multiplicative decomposition (trend × monthly × weekday) fitted only on the
archive's own daily counts; no external inputs. Its settings are chosen automatically by
walk-forward validation rather than hand-picked. Monthly factors are shrinkage-weighted
toward season-level averages. Bands come from walk-forward backtest RMSE, interpolated by
horizon. The model is deliberately simple — with roughly one seasonal cycle of data, a more
complex model would overfit. It carries no knowledge of Wizz Air's plans and should be read
as "if recent behaviour continues."
updated 17h ago