Wraps and portions: what prices show across cities

What we compare

Correction: the previous headline, ‘Thessaloniki is not expensive, it is generous’, presented a larger-portion hypothesis as a conclusion. We do not have the weight measurements needed to establish it. The headline and interpretation have been corrected; the price observations are preserved.

Comparing wraps and portions within a menu holds the shop constant, but does not eliminate differences in quantity, sides or pricing margins.

We measure a price ratio. Interpreting it as a meat ratio requires additional weight and composition evidence.

The explanation was already on our pages, but it rested on 18 price pairs, gathered one at a time from photographed menus across two decades. They gave Thessaloniki 0.41 and Athens 0.33.

The answer was already here

One of the three platforms we read every month also quotes a portion price, in the same shop on the same day. The test was already sitting in our data 1518 times, and nobody had asked it.

The result does not contradict the menus, it enlarges them: Athens 0.381 across 573 shops, Thessaloniki 0.522 across 116. The Thessaloniki price ratio is 37% higher.

And it costs 19% more. Assuming comparable portions and proportional prices and quantities, the hypothetical content adjustment gives: Thessaloniki comes out at 0.87, that is 13% CHEAPER than Athens.

The 2014 test, twelve years on

Our page cited as its cleanest case two menus from one year: Thessaloniki 2014, wrap 2.70 and portion 6.90; Ioannina 2014, wrap 2.10 and portion 6.50. The portion almost identical, the wrap 29% dearer.

Two menus are two menus. The very same test, across the 161 shops of those same two cities in 2026: the portion differs by 2%, the wrap by 19%.

Twelve years later, by a completely different method, the same shape. That is not proof, but it is as close to replication as price data will give you.

And then something we were not looking for

We put all 51 cities with enough sample on one chart: portion price across, the share the wrap holds up. The correlation is -0.80.

Look at what spreads and what does not. The PORTION runs from 7.60 € in Halkidiki to 15.95 € in Mykonos, very nearly doubling. The WRAP in those same cities moves by only 71%.

We observe less variation in wrap prices than in portion prices. This is compatible with different quantities, composition or pricing practices. It does not establish higher local costs or a smaller wrap; comparable quantity and cost measurements are missing.

Then we asked without the platform that gave it

All of the above once came from one platform. It was the only one giving a portion price, but not the only one that could: the other two simply were not looking. The very same rule went into all three, and the pairs went from 472 to 1,518.

The finding survived the tripling. Cities with enough sample went from 17 to 51 and the correlation from -0.81 to -0.80. The islands give a portion of 12.00 € and a ratio of 0.383, the northern mainland 8.50 € and 0.500.

The question of whether this is one channel's quirk does not disappear, though; it changes shape. It is now asked INSIDE the same reading: keep only the cities where one platform AND the other two each have four shops or more, and compare the two sides.

That is 31 cities, 630 shops on one side and 642 on the other. The ratios track each other at 0.85 with a median gap of 0.006. Athens 0.384 against 0.378. Thessaloniki 0.533 against 0.511. Volos 0.447 and 0.447, identical to the third decimal.

This form of the check is better than the previous one, and it is worth saying why. Before, we compared two different readings, so part of the difference could have been time or sample, and the shared cities were ten. Now both sides are the same file from the same day, and the cities are three times as many.

Two cities disagree noticeably and stay visible. Halkidiki gives 0.660 on one side and 0.531 on the other, on six and ten shops. Karditsa 0.494 against 0.615, on seven and four. On samples like that neither figure carries a conclusion, and they are not hidden.

The hollow shapes in the chart above are the second side. Where the line joining them to the filled circle is invisible, the two sides agree.

Why it matters

The index measures how many pita gyros a wage buys. If the wrap is not the same thing from city to city, the geographic comparison loses its meaning, and the worst part is that you cannot tell: the numbers come out looking fine.

That is why Thessaloniki's factor stays in the series as a price, with the explanation beside it rather than inside it. The per-content adjustment enters NO observation, because it assumes the portion is a comparable product in both cities. The within-shop ratio does not need that; the between-city comparison does.

The price ratio is an observation. Interpreting it as a quantity difference remains a hypothesis to test. The next evidence needs weighed cooked meat, full product composition, price, date and sales channel from the same shops.