Neapolitan Margherita Gozney Arc Lite · 12" max · 12 mm cordierite Lausanne

Il Disciplinare

Everything needed to make the pizza, and nothing else.

Ball240 g
Diameter28 cm
Hydration65 %
Salt2.5 %
Stone390 °C
Bake90 s
One

The kit What to put the dough in, what to cover it with, and where in the house to stand it.

Vessels

ForWhatWhy
MixingWide, shallow bowl, 4–5 L for a 1 kg batch. Metal or glass.You need room to get a whole hand in and turn the dough. A deep narrow bowl fights you.
Bulk fermentStraight-sided lidded tub, roughly 3 L per kilo of flour.Straight sides let you read the rise against the wall; sloped sides make it meaningless. 3 L holds a 1 kg batch with room to double — headroom you should never actually use, since you ball at 30–50 %.
The ballsFlat rectangular lidded box, 5–6 cm deep. One 30 × 40 cm box holds seven.Balls must sit apart and lift out from above without deforming. Depth is useless; floor area is everything.
Aliquot jarNarrow straight-sided glass, 50–100 ml — a shot glass or spice jar. Rubber band to mark it.Narrow means a small rise moves the level a long way, so you can read 10 % differences.
PoolishTall container, at least the poolish volume. 2 L for a 600 g poolish.It roughly triples, then falls back. Too small and it climbs out overnight.
BigaWide tub, 2× volume, lid resting slightly ajar.Barely rises, but must breathe. The one thing here not sealed airtight.

If you have none of this, a mixing bowl, a plastic food box for the balls and a jam jar will do everything. What cannot be improvised is the scale and the two thermometers.

Covering

The enemy is a skin. An uncovered surface dries within twenty minutes, and the dry layer never rehydrates — it shows up as pale patches that refuse to colour, and as tearing when you open the ball.

  1. A proper lid. Always the best answer. Not clipped down airtight; just resting closed.
  2. Cling film across the rim, pulled taut and pressed to the outside of the container — not onto the dough itself, which tears the surface when you peel it off.
  3. A damp cloth under a plate, for a 20–30 minute rest only. Over a night it dries and then wicks moisture out of the dough.
  4. A dry tea towel is not a cover. It is a wick.

Two exceptions. The biga wants a small vent — lid ajar, or film pierced once. The poolish wants to be airtight, since its wet surface skins fast.

Oil is not a cover, but a thin film wiped round the inside of the bulk tub and over each ball stops sticking and buys a margin against drying. Wipe it on with a paper towel — a sheen, not a pool.

Finding your temperatures

The recipes ask for 20 °C and 17 °C, and most people guess. Take the probe thermometer round the house once, find the places, then use them every time.

Two readings matter and they are not the same. Where you mix sets the water temperature. Where the dough then sits sets the timings. In a heatwave the kitchen can be 8 °C above the cellar, and treating them as one number is enough to wreck a batch.

NeedUsually foundFor
16–18 °CCellar, garage, unheated north-facing room, or a cool box with a bottle of fridge water in itBiga
20–22 °CA living room away from radiators, sun and draughtsPoolish, bulk, ball proof
26–28 °CClosed oven with only the light on — measure it first, many run past 30 °C. Otherwise on top of the fridgeCatching up
4 °CFridge, middle shelfSlowing down
whatever it isThe kitchen, at the moment you mixThe prep temperature — feeds the water calculation only
Measure at the right hour

A kitchen can swing 5 °C between 8 pm and 6 am, and an overnight ferment lives through all of it. Take the reading in the spot you will actually use, at the time of night it will actually sit there — not at teatime with the oven on.

An empty cool box makes an excellent proofing chamber: it holds whatever you put in it for hours, and a bottle of warm or cold water nudges it either way.

Two

Reading the dough Four checks. They tell you when the dough is ready; the hours in the recipes only tell you when to start looking.

1 · Dough temperatureA1A2

Take the temperature of the dough the moment mixing finishes. Aim for 22 °C in a final dough, 18 °C in a biga. Adjust the water to hit it: multiply the target by three, then subtract the flour temperature, the prep room temperature and the friction of mixing.

Two rooms, two jobs

The prep room — where you mix — sets the water temperature, because it is one of the three thermal masses meeting in the bowl. The storage room — where the dough then sits for sixteen hours — sets the yeast dose and every duration in the schedule.

They are often the same place. When they are not, using the storage figure to calculate the water is the classic way to end up several degrees too warm: a kitchen at 28 °C and a cellar at 20 °C will give you water 8 °C hotter than you needed, and a dough that starts at 27 °C instead of 22 °C. That is nearly a full doubling of rate before the ferment has even begun.

Fermentation runs about twice as fast for every 8 °C. A dough that comes out at 26 °C instead of 22 °C is ready hours before the recipe says.

2 · The aliquot jar

At balling, press 40 g of the same dough into a narrow straight-sided jar, flatten the top, and mark the level with a rubber band. Stand it beside the proofing box. The jar rises exactly as the balls do, and the percentage reads off the glass.

Bake when it reads 50–75 %. Not doubled — a ball taken to double has spent the gas it needs for oven spring.

How far to let the bulk rise

30–50 %. Not double — and a tub with room to double is the right size, because you want headroom you never use.

Gas made during the bulk is gas you are about to press out. Dividing and balling degasses the dough, redistributes the yeast into fresh food and builds new surface tension, so anything inflated before that point is spent for nothing. Bulk is where flavour and extensibility develop; the rise that has to be right is the one in the balls.

Put a rubber band on the bulk tub at the starting level, the same trick as the jar. Then read three things:

  • Level against the band — 30–50 % up
  • Feel — two fingers pressed in meet almost no resistance. Firm and springy means not yet
  • Look — domed rather than flat, bubbles visible where the dough meets the wall

Relaxation outranks volume. If the dough has gone soft and slack but the level says only 25 %, ball it anyway. A dough that is still tight will fight you at balling and give loose, badly formed balls, which no amount of proofing afterwards will fix. The reverse also holds: at 50 % but still springy, give it another half hour.

Where to spend spare time

When a schedule has slack in it, put it in the bulk, not the ball proof.

The bulk is forgiving because balling rebuilds whatever you over-relaxed. The ball proof is not, because nothing comes after it — an over-proofed ball has spent its gas and lost its surface tension, and you find out at the moment it tears on the peel.

Limits in both directions: past about six hours at 22 °C the bulk starts going sticky and tearing when folded; under two hours of ball proof the balls stay tight and snap back.

3 · The dimple

Flour a fingertip and press 1 cm into the side of a ball.

  • Springs back fast and completely — not yet. Give it 30–45 minutes.
  • Fills slowly and not quite all the way — ready.
  • Doesn't fill, ball spread flat, smells boozy — past it. Bake anyway; it will be paler and flatter.

4 · The open test

Opening is stretching a ball out into a flat disc — stendere in Italian. It is done entirely by hand, and it is where a proofing fault shows up first.

A ready ball opens without resisting: you stretch it, and it stays stretched. If the disc pulls back at you, contracts after you let go, or bunches up rather than spreading, it needs more time. Tenacious dough is under-proofed dough until proven otherwise.

The full technique is in Six.

Running early or late

Early: into the fridge. At 4 °C the dough ferments at roughly a seventh of its 20 °C rate. Pull it out 90 minutes before baking.

Late: warm to 26–28 °C — an oven with only the light on. Never above 30 °C; the dough turns vinegary and the gluten weakens.

Ripeness, stage by stage

StageReadyNot yetPast it
Poolish Domed, finely bubbled, centre just beginning to dimple. Smells of apple and cider. Flat, barely foamed, smells of raw flour. Sunk and watery, coarse bubbles, sharply sour.
Biga A torn clump pulls into a web with visible pockets. Soft, not crumbly. Smells of yogurt and hazelnut. Dry, breaks short, no web, smells of flour. Greyish and slack, wet at the base, vinegary.
Bulk Relaxed and soft under the hand, domed, bubbles against the tub wall. Risen 30–50 %. Springy and resistant. Slack and sticky, tears when folded.
Balls Jar at 50–75 %. Domed, slightly wobbly. Opens without fighting. Tight dome, dimple springs back, disc snaps back. Spread flat and merged, surface blistered.
Three

Flour, water, salt, yeast Four ingredients, four decisions.

FlourA5

Match strength to time. Too strong for the schedule and the rim comes out chewy and rubbery; too weak and the dough slackens and tears.

FlourWFermentHydration
Caputo Pizzeria (blue)260–2808–24 h60–63 %
Caputo Cuoco (red)300–32024–48 h62–65 %
Caputo Nuvola270–29024–48 h65–70 %
Caputo Nuvola Super320–34048 h +70–75 %

In Lausanne: L'Épicerie Fine Sainte Lucie, or Galaxus online.

Hydration

Start at 62 %. Move to 65 % once you can open a ball without tearing it, and to 68 % when 65 % feels easy. Higher hydration buys a lighter, more open crumb and costs handling difficulty.

WaterA4

Lausanne’s water is hard — 295 ppm against an ideal of 100–150 — and calcium tightens the gluten. To soften it, dilute: 41 % tap to 59 % distilled lands at 120 ppm. Adding extra water instead does not remove minerals; it only makes a wetter dough.

Change the water last. Every other cause of tight dough is larger.

Salt

2.5 % of total flour, fine sea salt, added with the final water once the flour is hydrated.

YeastA3

Fresh yeast is the default here — it is what Italian formulas are written in, it disperses readily, and the doses land at weighable sizes. Keep the block wrapped in the fridge for about two weeks; never freeze it. Before committing a batch, check it is firm, pale cream and crumbles cleanly. Tan edges or a sharp smell mean it has lost activity, which is the commonest reason a dough never moves.

Quantities depend on time and temperature. Below are the doses for the three recipes; for any other schedule, room or preferment length, use the calculator.

StageTempFreshActive dryInstant
Poolish · 300 g flour · 18 h20 °C0.7 g0.22 g0.18 g
Biga · 1000 g flour · 17 h17 °C10 g3.3 g2.6 g
Direct · 1000 g flour · 23 h20 °C1.4 g0.5 g0.4 g
Weighing under a gram

The biga's 10 g weighs directly — about a quarter of a supermarket block. For the poolish and the direct dough, crumble 1 g of yeast into 100 g of water and wait ten minutes. Each gram of that liquid holds 0.01 g of yeast, so 0.7 g becomes 70 g of liquid, weighable on any scale. Subtract it from the recipe's water, and mix it fresh each time.

Four

The three doughs Each makes seven balls of 240 g — seven pizzas at 28 cm, the working size for the Arc Lite. Begin with the direct dough and stay there until the oven is familiar.

24 hours

Direct

1000 g flour · W260–280 · 65 %

Flour
1000 g
Water
650 g
Salt
25 g
Fresh yeast
1.4 g
Method
  1. Mix, rest 20 min, then knead to 22 °C — see Five.
  2. Bulk 18 h at 20 °C, covered.
  3. Divide into 7 × 240 g, ball tightly. Fill the aliquot jar.
  4. Proof 5 h at 20 °C, to 50–75 %.
24 hours

Poolish

1000 g flour · W270–300 · 65 % · 30 % prefermented

Poolish · 18 h at 20 °C
Flour
300 g
Water
300 g
Fresh yeast
0.7 g

Stir to a smooth batter, cover. Ready when domed, finely bubbled, centre just dimpling.

Dough
All the poolish
600 g
Flour
700 g
Water
350 g
Salt
25 g
  1. Loosen the poolish in the water, add flour, rest 20 min, salt, knead to 22 °C — see Five.
  2. Rest 30 min, covered.
  3. Divide into 7 × 240 g and ball.
  4. Proof 4 h at 20 °C, to 50–75 %.
20 hours

Biga 100 %

1000 g flour · W300–340 · 65 % · all prefermented

Biga · 17 h at 16–18 °C
Flour
1000 g
Water, 15 °C
450 g
Fresh yeast
10 g

Mix 3–4 minutes only, to dry lumps with no loose flour. Cover, leaving a small vent. Ready when a torn clump webs and smells of yogurt.

Dough
All the biga
1453 g
Cold water
200 g
Salt
25 g
  1. Break into walnut pieces. Add 150 g water, mix slowly 5 min — see Five.
  2. Add the salt, then the last 50 g water in stages. Mix 4–5 min, to 22 °C.
  3. Fold to smooth. Rest 20 min.
  4. Divide into 7 × 240 g and ball.
  5. Proof 2 h at 20 °C, to 50–75 %.

Size and the ovenA8

The Arc Lite is rated to 12 inches — 30 cm — on a cooking floor 32.5 cm wide, through a mouth 32.5 cm across and 8.5 cm high. At a full 30 cm the pizza leaves barely a centimetre either side, which is not enough room to get a turning peel under it cleanly.

Work at 28 cm. It gives you 2 cm of clearance all round, turns easily, and launches without catching the sides.

DiameterBallUse
22 cm150 gChildren's size
26 cm210 gEasy handling while learning
28 cm245 gThe working size
30 cm280 gThe oven's limit — tight to turn

Finished pizza runs about 0.40 g per cm² of surface, so any diameter sets its own ball weight.

Five

Mixing and kneading Two jobs, done in order: wet every particle of flour, then build the gluten. Each dough needs a different amount of the second.

The two jobsA11

Mixing wets the flour. Kneading links the proteins into a network. They are separate, and doing the first properly makes the second much shorter — a dough left to sit after wetting builds a surprising amount of gluten on its own, with no work from you.

So every method here follows the same shape: combine roughly, wait, then work. The waiting is not idleness; it is the cheapest gluten development available.

The two hand techniques

You only need these two. Both are for after the rest.

Slap and foldA12 — for the bench, and the faster of the two. Slide both hands under the far edge of the dough and lift it. Slap the hanging half onto the bench in front of you, then fold the part still in your hands over the top, away from you. Turn the dough 90° and repeat. Aim for about one cycle a second and keep going for 8–10 minutes. It is sticky and hopeless for the first two minutes, then abruptly starts holding together. Do not add flour through that stage — it always passes.

Stretch and fold — for the bowl, slower in clock time but almost no effort. Wet one hand, reach under one side of the dough, lift it until it stretches, and fold it over the centre. Turn the bowl 90° and repeat four to six times; that is one set. Do four sets, 20 minutes apart. Between sets the dough relaxes and does the work for you.

Direct dough — by hand

  1. Hold back 50 g of the water. Dissolve the yeast in the rest.
  2. Add the flour. Mix with one hand or a spoon for 2 minutes, to a shaggy mass with no dry flour left at the bottom of the bowl. It will look rough. That is correct.
  3. Rest 20 minutes, covered. Nothing to do. The flour finishes absorbing and the gluten begins forming by itself.
  4. Add the salt and the reserved 50 g of water. Squeeze the dough repeatedly through your fingers until the water disappears and there are no wet streaks — 2 minutes.
  5. Slap and fold 8–10 minutes, or four sets of stretch and fold.
  6. Rest 10 minutes, then shape into a tight smooth ball: drag it toward you on an unfloured bench, turning as you go, until the surface is taut.
  7. Windowpane, then take the temperature. Target 22 °C.

Poolish dough

The poolish itself is not kneaded. Whisk flour, water and yeast solution for 30 seconds — just until no dry lumps remain. It should pour like a thick batter. Nothing more.

The final dough needs noticeably less work than a direct dough, because the poolish arrives with its gluten already built.

  1. Loosen the mature poolish with the new water first — stir until it slackens into a milky liquid. This is the step people skip, and skipping it leaves lumps of poolish that never disperse.
  2. Add the flour. Mix 2 minutes to a shaggy mass.
  3. Rest 20 minutes, covered.
  4. Add the salt. Squeeze through until absorbed.
  5. Slap and fold 6–8 minutes. It comes together faster than a direct dough and feels smoother earlier — stop when it does, rather than working to the clock.
  6. Rest 30 minutes, then divide and ball.

Biga — day one

This is the one step in the manual that is not kneading in any sense, and the instinct to knead it must be resisted. You are damping flour, not making dough.

  1. Pour the yeast water over the flour in stages, tossing with your fingertips as if rubbing butter into pastry.
  2. 3–4 minutes maximum. The result is dry, lumpy and looks like a failure: irregular clumps from pea to walnut size, holding together only when squeezed.
  3. The single thing to check is that no dry flour remains at the bottom of the bowl. Dig down and look.

If it forms a coherent dough, you have overworked it. It will still make pizza, but with less of the volume the biga is for.

Biga — day two, the refresh

The hardest mixing in this manual: forcing 200 g of cold water into 1.45 kg of stiff, fully developed, seventeen-hour-old preferment. Use a mixer if you have one.

  1. Break the biga into walnut-sized pieces into the bowl. Do not tip it in whole.
  2. Add 150 g of the water. Mix on the lowest speed for 5 minutes. For the first three it will look like a bowl of wet gravel that will never combine. It combines.
  3. Add the salt. Raise to medium and mix 4–5 minutes, until the dough clears the sides of the bowl.
  4. Add the last 50 g of water in three additions, waiting for each to be fully taken up before the next. Rushing this is what turns a biga dough into a slurry.
  5. Turn out and fold by hand a few times to smooth. Rest 20 minutes, then divide and ball.

Watch the temperature, not the clock. A stiff dough heats fast in a mixer — 6 °C or more across those ten minutes. Check at the halfway point and stop early if it is climbing past 22 °C.

By hand it takes 12–15 minutes of genuinely hard work. Use the pincer method: squeeze the dough flat between thumb and fingers to cut through it, then fold it over, repeatedly. Expect to feel it is going nowhere for the first five minutes.

Knowing when to stop

The windowpane. Pinch off a piece the size of a walnut and stretch it slowly between the fingers of both hands. Ready dough thins into a translucent membrane you can see light through before it tears. Nuvola at 65 % should do this easily.

StateFeel and lookDo
Under-workedTears short and thick when stretched. Rough, dull surface. Slumps flat.Another 3 minutes, or one more set of folds.
ReadySmooth and satiny. Tacky but leaves your hands clean. Holds a domed shape. Windowpanes.Stop.
Over-workedSlack, wet, greyish, tears wetly rather than stretching. Only reachable in a mixer.Nothing recovers it. Bake it flat and start again.
Stickiness

Never add flour to a sticky dough. It changes the hydration you calculated and the dough turns out tight and dry. Wet your hands, or use a plastic scraper to lift and turn it. Stickiness at minute two is normal and passes on its own; stickiness at minute ten means the dough is under-worked, not over-hydrated.

Six

The day, step by step The direct dough in full, with nothing left out. The other two differ only at the start, and those differences are marked at the end.

Hour 0 · Mise en place

  1. Take two temperatures first: the kitchen you are standing in, and the flour itself. Put both into the calculator as the prep room and flour figures — the storage room is a separate setting and does not belong here.
  2. Weigh 1000 g flour into the mixing bowl and 650 g water into a jug. Adjust the water with hot or cold until it reads what the calculator asked for. If that is below about 14 °C, use fridge water or ice, and weigh what you actually pour.
  3. Weigh 25 g salt into a small dish.
  4. Make the yeast solution: crumble 1 g fresh yeast into 100 g water, stir, wait ten minutes. Weigh out 140 g for a 1.4 g dose, and subtract it from the 650 g — so 510 g plain water plus 140 g solution.
  5. Pour 50 g of the plain water into a separate cup and set it aside. This gets held back.

Hour 0 · The mix

  1. Pour the solution and the remaining water into the bowl with the flour.
  2. Mix with one cupped hand, turning the bowl with the other, for 2 minutes. Scrape the sides down. Dig to the bottom and check no dry flour is hiding there.
  3. It will be rough, lumpy and unpromising. Correct.
  4. Cover and leave 20 minutes on the bench.

Hour 0:20 · Salt and knead

  1. Scatter the salt over the dough and pour on the reserved 50 g of water.
  2. Squeeze the dough through your fingers for about 2 minutes, until no wet streaks or salt grains remain.
  3. Tip onto an unfloured bench. Slap and fold 8–10 minutes. It sticks badly for the first two. Do not add flour — scrape your hands with the bench scraper and carry on.
  4. Rest 10 minutes under the upturned bowl, then shape into a tight ball by dragging it toward you across the bench until the surface goes taut.
  5. Windowpane: stretch a walnut-sized piece thin enough to see light through. If it tears, 3 minutes more.
  6. Take the temperature and write it down. Target 22 °C.
  7. If it reads above 24 °C, put the covered bowl in the fridge for one hour before it goes to storage. A dough that starts hot stays hot for hours and ferments far ahead of the schedule. This one hour is the cheapest insurance in the whole process.

Hour 0:35 · Into bulk

  1. Wipe the inside of the bulk tub with olive oil on a paper towel.
  2. Put the dough in seam down and press it flat to the corners so the surface is level. This is what makes the rise readable.
  3. Lid on, resting closed.
  4. Stand it in your 20 °C spot. Not beside the oven, not in sun, not on a cold tiled floor.
  5. Put a rubber band round the tub at the level of the dough.
  6. Write the time and the dough temperature on tape on the lid.

Hour 18 · Divide and ball

Start checking at 16 hours. Ready means risen 30–50 % against the band, domed, bubbles visible through the side of the tub, and — the part that decides it — relaxed and soft under the hand. If it is still springy, wait, whatever the level says.

  1. Dust the bench lightly. Invert the tub and let the dough fall out under its own weight. Do not pull it — that tears the structure you spent the night building.
  2. Cut 7 pieces of 240 g with the bench scraper, weighing each. Cut straight down. Do not saw or tear.
  3. Ball each oneA12: fold the edges into the centre four or five times to make a parcel, turn it seam down, then cup both hands round it and drag it in small circles on the unfloured bench until the surface is taut and smooth. About 15 seconds each. A tight ball holds gas; a loose one spreads.
  4. Pinch the seam closed underneath.
  5. Press 40 g of dough into the aliquot jar. Push out the air pockets, flatten the top, mark the level with the rubber band. Stand it beside the box.
  6. Set the balls in the oiled box seam down, at least 4 cm apart — they roughly double their footprint.
  7. Lid on. Back to the 20 °C spot.

Hour 23 · Ready

Start checking at 4 hours. Three things must all be true:

  • Jar at 50–75 % above the band
  • A finger pressed into the side of a ball leaves a dimple that fills slowly and incompletely
  • A test ball opens without pulling back

If the first two say yes and the third says no, give it another 30 minutes. The third one wins.

Light the oven 45 minutes before you want to eat. The balls carry on proofing while it heats — expected, and already accounted for.

Lifting a ball out

Flour the top of the ball, slide a floured bench scraper flat underneath, and lift it out in one movement onto a bed of semolina. Never pull it out with your fingers: you stretch and degas it before it reaches the bench, and that is exactly where a flat cornicione comes from.

Opening the ballA12

Two stages: press the gas outward to build the rim, then stretch the middle. Never a rolling pin — it crushes the gas out of the cornicione, and no oven heat brings it back.

Work fast. A disc left standing sticks to the bench and starts to dry.

Stage one — the press. Builds the cornicione and sets the base thickness.

  1. Lift the ball with a floured scraper onto a bed of semolina. Turn it once so both faces are lightly coated, then shake the excess off.
  2. Place the pads of three fingers of each hand about 2 cm in from the edge. Press down firmly and push outward, driving the gas from the centre into the rim.
  3. Rotate a quarter turn, press again. Go round two or three times until you have a disc of about 15 cm with a visibly thick, puffy rim.
  4. Flip it over and repeat once on the other side.
  5. Never touch the outer 2 cm. Every finger that lands there costs you cornicione.

Stage two — the stretch. Takes it from 15 cm to 28. Three ways; use whichever you can do without tearing.

  • Knuckles and gravity — the easiest to learn. Drape the disc over the backs of both closed hands, lift it clear of the bench, and let its own weight pull it down. Rotate steadily. Use knuckles, never fingertips, which punch holes.
  • The steering wheel — hold the disc vertically by the rim, as if gripping the top of a steering wheel, and pass it hand to hand round the edge. Gravity does the stretching while the rim stays untouched.
  • The slap (schiaffo) — the Neapolitan method, fastest once learned. Hold the disc by the rim in one hand with the other half hanging. Slap the hanging half onto the bench, pulling gently toward you as it lands, then rotate a quarter turn and repeat. A skilled pizzaiolo takes a ball to 30 cm in under thirty seconds.

When to stop: 28 cm across, centre 2.5 mm, rim 15–20 mm. Hold it up against the light — the centre should glow evenly. Bright patches are thin spots and they will tear on launch.

Opening is also a diagnosis

Bunches up and resists — under-proofed, or too cold. Put it back and take another ball.

Tears in the centre — over-proofed, or you stretched too thin.

Springs back after each stretch — not ready. Rest the disc 5 minutes and try again; if it still fights, wait.

Balls do not ripen at the same rate. Take them in the order they are ready, not the order they went into the box.

Faults to avoid. Too much flour on the disc burns black on the stone — brush it off before topping. Do not stretch the pizza on the peel; open it on the bench, then transfer. And top it only once it is on the peel, working quickly, because a topped pizza left standing welds itself down.

Where the other two differ

Poolish. The night before, whisk 300 g flour, 300 g water and the yeast solution for 30 seconds in a tall airtight container, and stand it at 20 °C. Next day, when it is domed and just dimpling in the centre, tip it into the bowl and stir the new water into it first, until it slackens to a milky liquid. Then add the flour and pick up at Hour 0 · The mix, step 2. Bulk is only 30 minutes; ball as above and proof 4 hours.

Biga. The night before, toss 1000 g flour with 450 g of 15 °C yeast water using fingertips only, 3–4 minutes, to damp clumps. Lid ajar, at 17 °C. Next day break it into walnut pieces, mix with 150 g water on the lowest mixer speed 5 minutes, add the salt, medium 4–5 minutes, then the last 50 g of water in three additions. Rest 20 minutes, ball as above, proof 2 hours.

Seven

The oven Cordierite drives heat into dough faster than a traditional clay floor, so the Arc Lite bakes cooler and a little longer than a wood oven.

The chamber

Cooking floor 32.5 × 35.5 cm, 15 cm high, through a mouth 32.5 cm wide and 8.5 cm tall. Rated to 12 inches; work at 28 cm. One pizza at a time, so the evening runs as a production line — stretch, top, launch, turn, out, recharge.

PreheatA6

Light the oven on maximum 45 minutes before the first launch. The stone alone is hot in two minutes; the deck, body and dome around it take the rest, and until they are hot the stone keeps bleeding heat into them.

Read the centre of the stone with an infrared thermometer before every launch. Target 380–400 °C, and calibrate upward if the base comes out pale.

Flame

Turn the dial down to low immediately before launching. For the first 45 seconds the stone alone drives the oven spring; radiant heat at this stage burns the rim before the centre has cooked.

At 45 seconds the base is set. Raise to medium, then to high for the last 20–30 seconds to char the rim.

Turning

The burner runs along one side, so the near edge takes far more radiant heat than the far edge. Quarter-turn every 15–20 seconds — four or five turns across the bake. A single 180° flip only moves the burn to the other side.

Flame setting against time. The rim is charred at the end, on a base that has already set.

Between pizzas

Full flame for 2–3 minutes while the next ball is stretched and topped. Launching sooner gives a pale base.

Semolina

Use the least that stops the dough sticking, and brush the excess off the disc — loose grains burn on the stone in seconds and turn bitter. Shake the peel before launching. If the pizza doesn't slide freely, it will not launch.

Eight

Toppings Water on top of the dough holds that layer at 100 °C until it has all evaporated. Every rule here is about getting rid of it.

Tomato

San Marzano DOP, crushed by hand or through a food mill. Never a blade — shearing separates the water from the pectin holding it, and the sauce floods the base. Salt at 1–1.5 % by weight. Do not cook it; it cooks on the pizza.

CheeseA9

Fior di latte, ideally Agerola. Cut into batons 2–3 mm thick and 1–2 cm wide — cubes have too little surface to shed water and too little mass to survive the heat. Drain in a colander 45–60 minutes, and use at room temperature. Cold cheese chills the surface and forces a longer bake.

Mozzarella di bufala works but is wetter and fattier: tear it, drain on a rack for two hours, use a third less, and add half in the last twenty seconds.

In Lausanne: Aligro for professional formats, or a specialist cheesemonger.

Basil and oil

Tuck the leaves under the cheese before launching, or add them in the last ten seconds — bare on top for a full bake they turn to ash. Weigh the oil at 5 g in a thin spiral, and add a final thread once the pizza is out.

Topping

Open the ball on the bench — the technique is in Six — then transfer it to the peel and top it there. Sauce first, in a spiral from the centre, leaving the rim bare. Cheese, then basil tucked under it, then the oil. Work quickly and launch straight away.

Nine

When it goes wrong One symptom, one change. Change a single thing per bake and note it.

SymptomCauseChange
Gum line — doughy layer under the sauceA7Top heat too low, or too much wet toppingMore flame for the last 30 s. Then cut tomato to 50 g and drain the cheese longer.
Pale, even baseStone not recoveredFull flame 3 min between pizzas; verify by infrared.
Black patches on the baseStone too hot, or loose semolinaDrop the target 20 °C; brush the disc.
Rim burnt, centre rawFlame too high earlyLow for the first 45 s.
One side dark, one paleNot enough turnsQuarter-turn every 15–20 s.
Dough snaps backUnder-proofed ballsAnother 45 min.
Dense, fine crumbUnder-fermentedProof to a true 50–75 % before touching hydration.
Chewy, rubbery rimFlour too strong for the timeLonger ferment, or a W260–280 flour.
Balls spread flat, boozyToo much yeast, or too warmRecalculate at your real dough temperature.
Orange pool in the centreCheese not drained, or too much oilDrain 60 min; weigh the oil at 5 g.
No oven springGas pressed out of the rimWork only the centre; never the outer 2 cm.
Ten

Batching, scaling and freezing Everything scales linearly. Dough survives the freezer at a modest cost; par-baked bases survive it at almost none.

Scaling a batch

Halve or double everything, yeast included. Nothing about the method changes. Below 800 g of total dough a planetary mixer may not engage the hook properly — hand-kneading is easier at that size in any case.

Ball weight follows diameter, so a mixed table is simply a matter of dividing differently.

FlourDoughDivides into
500 g840 g2 × 245 g at 28 cm, plus 2 × 175 g at 23 cm — two adults and two children
500 g840 g3 × 280 g at 30 cm, or 4 × 210 g at 26 cm
1000 g1680 g7 × 240 g at 28 cm

Small pizzas bake the same way — same thickness, same stone, roughly ten seconds less. They are also easier to turn, which is worth something in an oven this size.

Weighing at half scale

A halved poolish needs 0.35 g of fresh yeast, below what most kitchen scales resolve. Use the solution: crumble 1 g into 100 g of water, then weigh 35 g of the liquid. The poolish becomes 150 g flour, 115 g water, 35 g solution.

Freezing dough ballsA10

Freeze immediately after balling, before any proof. The dough then proofs after thawing, so the yeast lost to freezing is repaid in time rather than guesswork.

  1. Ball as normal. Oil lightly and wrap each one airtight.
  2. Spread them apart on a tray in the coldest part of the freezer. Bag them together once solid.
  3. Freeze at 62–65 % hydration, not 68.
  4. Best within a month; usable to three.

Thaw 12–24 hours in the fridge, then 3–5 hours at 20 °C — longer than a fresh dough, because fewer yeast cells survive. Freeze a 40 g aliquot in its own small jar alongside the balls and read 50–75 % off the glass exactly as usual.

Expect roughly 10–15 % less rise and a slightly tighter crumb. Poolish and biga doughs freeze better than direct dough, because their flavour is already built into the preferment rather than waiting on the final proof. Never freeze the preferment itself.

Freezing par-baked bases

The better option for weeknights, and it costs almost nothing in quality.

  1. Open a ball as normal. No toppings.
  2. Bake 35–45 seconds — until the base is set and barely coloured.
  3. Cool completely on a rack, stack with parchment between, bag and freeze.
  4. Top from frozen and bake 60–75 seconds.

The crumb comes out marginally drier, since the oven spring has already happened. In exchange the structure is locked in before freezing, so nothing at all is lost to ice.

One session, two dinners

Make the full kilogram. Keep half fresh for tonight and par-bake the other half into bases at whatever sizes suit the table. Dinner on the second night is ninety seconds from the freezer.

Eleven

The bench card Print this and keep it by the oven.

Numbers
Stone380–400 °C Preheat45 min Recovery2–3 min Bake75–100 s Ball240 g Diameter28 cm Centre2.5 mm Hydration65 % Salt2.5 % Dough temp22 °C Jar50–75 %
Before the launch
  • Stone centre read by infrared
  • Ball opens without fighting
  • Rim untouched, centre 2.5 mm
  • Cheese drained, room temperature
  • Basil under the cheese
  • Semolina brushed off
  • Pizza slides on the peel
  • Flame down to low
The 90 seconds
  1. Launch to the centre, flame low
  2. Quarter-turn at 15, 30, 45 s
  3. 45 s — flame to medium
  4. 60 s — flame to high
  5. Quarter-turn at 75 s
  6. Out at 90 s. Oil.
  7. Full flame 3 min. Next.
From the freezer
  • Balls: 12–24 h fridge, then 3–5 h at 20 °C
  • Read the frozen aliquot jar: 50–75 %
  • Bases: top frozen, bake 60–75 s
Fixes
  • Pale base → longer recovery
  • Burnt base → −20 °C
  • Burnt rim → low for 45 s
  • Gum line → more flame at the end
  • Uneven → more turns
  • Snaps back → wait 45 min
Appendix

Why the numbers are what they are The reasoning behind each figure in the manual, for when you need to depart from it.

Fermentation and temperature

Fermentation is a chemical reaction, and like most of them its rate roughly doubles for every 8 °C. That single relationship is the whole of scheduling: a recipe's hours are meaningless without its temperature.

rate ≈ 2^((T − 20) / 8) T in °C, referenced to 20 °C
Dough tempRateAn 18 h step becomesUse
4 °C0.15≈ 5 daysStorage, not fermentation
12 °C0.5036 hVery long, very aromatic biga
16–18 °C0.71–0.8421–25 hBiga window
20–22 °C1.00–1.1915–18 hPoolish and dough window
25 °C1.5412 hSummer kitchen; watch closely
28 °C2.009 hRescue speed
above 30 °CVinegary, weak gluten

Below about 10 °C the doubling law over-predicts: yeast slows disproportionately as it approaches dormancy. The fridge figure carries an additional ×0.6 correction, fitted to observation rather than derived.

Setting the water temperature

Three thermal masses meet in the mixing bowl — the flour, the air around it, and the water — plus the heat that mixing itself puts in. The finished dough lands near their average, so to hit a target you solve for the one you control.

The air in that average is the air in the room where you are mixing. This is worth stating plainly because it is the single easiest term to get wrong: the room the dough will ferment in is irrelevant to the mixing-bowl arithmetic, and substituting it produces an error equal to the difference between the two rooms, passed straight into the dough.

water °C = 3 × target − flour °C − prep room °C − friction friction ≈ 1 °C by hand friction ≈ 6 °C planetary mixer, 8 min Target 22 · flour 19 · room 21 · by hand: 66 − 19 − 21 − 1 = 25 °C

Friction is the only term you cannot look up. Measure it once: mix a batch, note the dough temperature, and the gap between it and the prediction is your figure for that mixer at that speed and time.

Flour left in a cold cupboard, warm hands and a long mix routinely put a dough 4–6 °C away from the room — most of a doubling, and the usual reason a schedule runs early.

How the yeast dose is calculated

The yeast needed is inversely proportional to how long it works and how warm it is kept. Sum the fermentation across each stage in rate-adjusted hours, then divide a constant by it.

effective hours = Σ ( hours × rate(T) ) dry yeast % = K ÷ effective hours (% of that stage's flour) K = 1.1 direct dough K = 1.4 poolish K = 4.5 biga

The three constants differ for real reasons. A direct dough contains salt, which slows the yeast, so it needs more than an unsalted poolish of the same length. A biga needs roughly three times a poolish because at 45 % hydration there is very little free water for the yeast to work in — water availability, not food, is the limit.

The formula returns active dry as its unit. Convert from there: fresh = dry × 3, instant = dry × 0.8. Since Italian formulas are quoted in fresh yeast, a dose given as a baker's percentage in a published recipe compares directly against the fresh column.

Calibration

The constants are fitted to published professional formulas rather than chosen for tidiness.

StageModelPublishedSource
Biga, 17 h at 17 °C1.00 % fresh1.00 % freshGozney, 100 % Biga
Poolish, 16 h at 20 °C0.070 % instant0.03–0.08 %Weekend Bakery
Poolish, 12 h at 21 °C0.086 % instant0.08–0.1 %ChainBaker
Direct, 9 h at 23 °C0.28 % fresh0.2–0.3 %AVPN practice

The model holds well over 12–24 hour preferments. Below about 8 hours it under-predicts, because at short times the yeast population is still growing and simple rate scaling no longer describes it.

Hard water and gluten

Calcium and magnesium ions cross-link gluten strands where they would otherwise slide past one another. The dough becomes more elastic and less extensible: it holds shape well but fights being opened.

Lausanne's mains water measures 16.6 °dH. German degrees convert at about 17.85 ppm of calcium carbonate each, giving 295 ppm against an ideal for dough of 100–150.

ppm CaCO₃ = °dH × 17.85 fraction of tap water = target ÷ tap 120 ÷ 295 = 0.41 → 41 % tap, 59 % distilled

Adding extra water instead of diluting does not change the mineral concentration at all — the ions arrive with the water. It only produces a slacker dough that is also still cross-linked, which is two problems where there was one.

The effect is real but modest, which is why it comes fourth in the list of suspects. Under-proofing produces the same symptom several times more strongly.

W value, and why strength must match time

"Tipo 00" describes only how finely the flour is milled and how little bran remains. It says nothing about baking strength. That is the W value, measured on a Chopin alveograph: the work absorbed inflating a bubble of dough until it bursts. Higher W means a gluten network that holds gas for longer without collapsing.

The companion figure is P/L, the ratio of resistance to extensibility. Around 0.50–0.60 is right for pizza — the dough stretches without snapping back.

Strength has to be matched to time because protease enzymes in the flour degrade gluten bonds slowly throughout fermentation. That controlled breakdown is what makes a dough extensible and gives the crumb its melting quality. A W340 flour on a 12-hour schedule never gets the hours it needs, and the rim comes out rubbery. A W260 flour on a 48-hour schedule gets too many, and the dough slackens and tears.

The stone: preheat, recovery, and cordierite

Heat takes a time of roughly L² / α to diffuse through a slab of thickness L, where α is thermal diffusivity. For cordierite, α is about 1 × 10⁻⁶ m²/s:

t ≈ L² / α = (0.012 m)² ÷ 10⁻⁶ m²/s ≈ 144 s

The stone itself saturates in a little over two minutes. The 45-minute preheat is for everything around it — the deck, the steel body, the insulation, the dome. Until those reach steady state the stone keeps losing heat downward and sideways into them, and the surface reading falls as fast as you build it.

The same two-minute constant is why 2–3 minutes on full flame genuinely restores the surface between pizzas: you are recharging only the slab, not the whole machine.

Cordierite against biscotto

Traditional Neapolitan floors use a low-conductivity clay, biscotto, at roughly 0.4 W/(m·K) against cordierite's 2.5–3.0 — a ratio of six to eight. But conductivity is not the quantity that decides whether a surface scorches dough. That is thermal effusivity, √(kρc), which describes how quickly a material can deliver heat into whatever touches it. Effusivity goes as a square root, so the real penalty is nearer two to three times.

Hence the working rule: bake a little cooler and a little longer than a wood oven — 380–400 °C and 90 seconds rather than 450 °C and 60. Not dramatically different, but enough that copying wood-oven numbers directly will burn the base.

What happens in ninety seconds

On contact, conductive heat drives into the base. As the dough passes 60 °C starch granules absorb water and gelatinise, and the unbound water turns to steam. The gluten network is still pliable, so the expanding steam inflates it — that is oven spring, and it is nearly all over within the first 20 seconds.

Between 20 and 45 seconds the base dehydrates and sets into a crust. From then on the pizza is being coloured rather than raised. Browning needs 140 °C at the surface and reducing sugars for the Maillard reaction; those sugars come from amylase breaking down starch during fermentation, which is why an under-fermented dough stays pale however hot the oven.

The gum line

A gum line is starch that gelatinised but never dried. The layer under the sauce reached 60 °C and absorbed water, but not enough energy arrived afterwards to evaporate it, so it stays translucent and doughy.

Three things cause it, and they compound: too much wet topping acting as a shield; the flame left low for the whole bake, so radiant heat never drives the moisture off while the conductive base cooks fast enough to force an early exit; and gas left trapped in the centre of the disc, insulating the sauce from the stone below.

Ball weight and diameter

A ball weight on its own says nothing — 280 g makes a thin 32 cm pizza or a thick 26 cm one. What stays constant across a well-made range of sizes is surface density, the dough per unit area of finished pizza.

ball g = 0.40 × π × (diameter / 2)² 22 cm → 152 g 28 cm → 246 g 26 cm → 212 g 30 cm → 283 g

The 0.40 g/cm² figure comes from the Neapolitan standard itself: six 280 g balls from a kilogram of flour at 65 % hydration, opened to 30 cm.

Why 28 cm and not 30

The Arc Lite is rated to 12 inches, and 30 cm is genuinely achievable — but rated size and working size are different questions. The cooking floor is 32.5 cm wide, so a 30 cm pizza leaves 1.2 cm of stone either side. A turning peel needs more than that to slide underneath without lifting the edge or dragging the pizza off centre, and the mouth is the same 32.5 cm, so the margin at launch is identical.

Dropping to 28 cm costs 13 % of the surface and buys 2 cm of clearance all round. Since this oven bakes one at a time and demands four or five turns per pizza, that clearance is worth more than the area.

Cheese geometry and moisture

Fior di latte is 60–65 % water. Applied straight from the packet, that water boils on the dough surface and holds it at 100 °C until it has gone — which is the whole bake, at which point the base is done and the layer beneath the cheese never cooked.

Two properties are in tension. Water leaves at a rate set by surface area, so thin pieces dry fastest. Resistance to scorching is set by mass, so thick pieces survive longest. Cubes lose on both counts: too little area to shed water, too little mass to survive the dome. Batons 2–3 mm thick and 1–2 cm wide sit at the useful compromise.

Draining for an hour removes only a few percent of the weight, not the large fraction sometimes claimed. But it is the right few percent — the free surface water that would otherwise pool — and it costs nothing but time.

Temperature matters for the same reason. Cheese straight from the fridge has to be warmed 15 °C before it can even begin to melt, and that energy comes out of the pizza surface, extending the bake and drying the crust.

What freezing does to dough

Yeast cells are killed by ice, not by cold. As water freezes it forms crystals, and crystals that grow large enough puncture cell walls. The cell contents leak out, and among them is glutathione — a reducing agent that attacks the disulphide bonds holding the gluten network together. So a badly frozen dough loses twice: fewer yeast to raise it, and a weaker structure to hold what gas there is.

Two things follow directly.

Freeze fast. Crystal size depends on how long the dough spends in the range where water is freezing. Balls spread apart on a tray freeze in a fraction of the time a stacked bag takes, and the crystals stay correspondingly small. Bag them only once they are solid.

Freeze drier. Free water is what crystallises. At 62–65 % hydration more of the water is bound to starch and protein and less is available to form ice, so the damage is measurably lower than at 68–70 %. This is the one situation where lower hydration is straightforwardly better.

Why before the proof

A dough frozen after proofing has already spent its gas, and the cells that would replace it are the ones the ice just killed. Frozen before the proof, the surviving population — typically 80–90 % of it — simply needs longer to reach the same aliquot reading. The loss converts into time, which you can measure, rather than into volume, which you cannot recover.

Why par-baked bases lose nothing

Once the base has passed 60 °C the starch has gelatinised and the gluten has set. The structure is no longer biological, so there is nothing left for ice to damage — only water to redistribute. What little quality is lost is moisture from the crumb, not rise.

What kneading actually does

Flour holds two proteins that matter. Glutenin is long and springy and provides strength; gliadin is compact and sticky and provides extensibility. Neither is gluten until water arrives. Once wetted, they unfold and begin linking — chiefly through disulphide bonds between sulphur atoms on adjacent chains — into the continuous elastic network that traps carbon dioxide.

Kneading does two things to speed this up: it brings chains into contact so they can find each other, and it aligns them so the links form in useful directions rather than at random.

Why resting works

Those bonds form on their own wherever chains happen to touch. Twenty minutes of rest after wetting the flour builds a substantial amount of network for no effort, which is why the rest cuts the kneading time roughly in half. Bakers call this an autolyse, and it also relaxes the dough, making the subsequent work easier.

Salt is held back until after the rest because it tightens the network and slows protein movement — useful later, obstructive while chains are still trying to find one another.

Why the biga is barely mixed

A biga's job is fermentation, not structure. Developing gluten on day one means spending, over seventeen hours of enzyme activity, a network you will then have to build again on day two. Left as damp clumps, the flour ferments while the proteins stay largely unlinked, and all the structural work happens at the refresh, when it will survive to the oven.

Bassinage

Water added after a network exists behaves differently from water added at the start. Introduced slowly into a developed dough, it lubricates between chains without dissolving the structure — which is how a dough reaches 65 or 70 % hydration while still handling like something drier. Added all at once, the same water simply produces a slurry the gluten cannot organise.

Why hands cannot over-knead

Over-kneading breaks disulphide bonds faster than they re-form and oxidises the carotenoid pigments that carry flavour and colour. Both need sustained mechanical energy well beyond what hands deliver — you tire first. A planetary mixer does not, which is why the only real over-kneading risk in this manual is the biga refresh, and why the instruction there is to stop on temperature rather than on time.

Watching it done

Three techniques are far easier to see than to read. These are worth ten minutes each before you do them for the first time.

TechniqueWatchWhat to look for
Slap and fold Before and after, 0–10 min
Bertinet, full method
The moment around minute three when it stops sticking and starts holding together. Also the posture — the movement comes from the arms and body, not the wrists.
Balling Gozney, shaping dough balls How little the hands do. The bench does the work as the ball is dragged in circles; the seam is pinched shut underneath at the end.
Opening — the slap The slap method, on a Gozney
Iacopelli, tecnica a schiaffo
The rim is never touched. Watch where the fingers land during the press, and how the disc is held by the edge throughout the stretch.

One caveat on the kneading videos: nearly all of them use a bread dough at 75–80 % hydration, which is far wetter and floppier than a 65 % pizza dough. The motion is identical, but yours comes together sooner and looks far less alarming. Do not judge your progress against how theirs behaves.

And on the stretching videos: they are working with fully mature dough at a bench they use every day. A first attempt taking two minutes rather than twenty seconds is normal. Speed matters only because the dough dries — accuracy first.