baking
Baker's Percentages: Hydration, Starters, Yeast and Cake Pans
Baker's percentages measure everything against flour, not dough. How that settles hydration, starter math, pizza styles, yeast swaps and cake pan scaling.
Recipe math in baking looks like several unrelated problems: how wet a dough is, how to count a sourdough starter, why one pizza recipe has oil and another bans it, how to swap yeast, how to fit a cake into a different pan. Nearly all of it comes down to one convention and one ratio. Every dough ingredient is a percentage of the flour, not of the finished dough, and every pan conversion is the new pan’s batter volume divided by the old one’s. Get those two right and the rest is arithmetic. The wrong answers people give usually come from counting against the wrong total.
Flour is always 100 %
King Arthur Baking’s professional reference on baker’s percentage states the rule: “each ingredient in a formula is expressed as a percentage of the flour weight, and the flour weight is always expressed as 100%.” Hydration is the water’s share on the same basis, so their example of 50 lb flour and 33 lb water is 66 % hydration.
This is where people slip. 75 % hydration does not mean the dough is 75 % water. It means 75 g of water for every 100 g of flour:
500 g flour × 75 % = 375 g water
500 g flour × 2 % = 10 g salt
total dough = 500 × (100 + 75 + 2) % = 885 gWater is only 42 % of that dough; the percentages add up to 177, not 100. Because everything scales off the flour, working backwards from a target dough weight is one division: flour = dough weight ÷ (sum of percentages). A 1,000 g dough at 100 + 75 + 2 needs 1,000 ÷ 1.77 = 565 g of flour.
Our bread hydration calculator works in both directions. Enter flour and water and it gives the hydration, or pick a target and it gives the water to add. It also shows where that number falls on a rough guide. That guide is our own summary of common home-baking ranges (bagels under 60 %, sandwich loaves 65–70 %, ciabatta 75 %+), not a standard.
The starter trap
A sourdough starter is flour and water mixed ahead of time. The same King Arthur reference says what to do with any preferment when you work out the overall formula: “break the poolish up into its component parts, and add them in to the proper places.” A 100 %-hydration starter is equal weights of flour and water, so 100 g of it holds 50 g of each.
Ignore that and your hydration is wrong. Take a recipe that says 500 g flour, 375 g water, 100 g starter and calls itself 75 %. Run it through our calculator’s preferment option:
| Flour | Water | Hydration | |
|---|---|---|---|
| Naive (starter ignored) | 500 g | 375 g | 75 % |
| True (starter split 50/50) | 550 g | 425 g | 77.3 % |
The error grows with the starter percentage. Going the other way, if you really want 75 % with that 100 g of starter and 500 g of added flour, the calculator says to add 362.5 g of water, not 375 g. The starter already brings 50 g.
Starter hydration matters too. A stiff 50 % starter is two parts flour to one part water, so the same 100 g adds 66.7 g flour and 33.3 g water. The recipe above then works out to 72.1 %.
What the sourdough calculator does with it
Our sourdough calculator starts from the loaf you want and subtracts the starter’s flour and water from what you weigh out. For one 1,000 g loaf at 75 % hydration, 20 % starter (100 % hydration) and 2 % salt, it returns:
| Weigh out | Grams | % of total flour |
|---|---|---|
| Flour | 508 | 90 % |
| Water | 367 | 65 % |
| Starter | 113 | 20 % |
| Salt | 11 | 2 % |
| Totals incl. starter | 565 flour / 424 water | 75 % |
The flour and water you add look like 72 % hydration if you divide one by the other (367 ÷ 508). Once the starter’s 56.5 g of each goes back in, it’s exactly 75 %.
Recipes also disagree about what “20 % starter” is 20 % of. Some divide by total flour, including the flour inside the starter. Others divide by only the flour you add to the bowl, so “100 g starter to 500 g flour” counts as 20 %. The calculator has a toggle for this. With the same 1,000 g loaf, switching to the “flour you add” basis changes the starter from 113 g to 103 g and the added flour from 508 g to 514 g. Total flour is the same 565 g either way.
Why pizza styles use such different numbers
Pizza dough is the same arithmetic with a few more ingredients, and the styles really do differ. Here are published reference formulas next to the presets in our pizza dough calculator:
| Style | Published reference | Our preset |
|---|---|---|
| Neapolitan | AVPN: 1 L water per 1.6–1.8 kg flour (55.6–62.5 %), 40–60 g salt per litre (2.2–3.75 % of flour), “never add any fat or sugar” | 62 % water, 2.8 % salt, no oil or sugar |
| New York | King Arthur recipe: 360 g flour, 206 g water (57 %), 12 g each of sugar, salt and oil (3.3 % each) | 62 % water, 2 % salt, 2.5 % oil, 1.5 % sugar |
| Detroit | King Arthur weeknight recipe: 360 g flour, 270 g water (75 %), 9 g salt (2.5 %), 3 g instant yeast (0.83 %) | 72 % water, 2 % salt, 1 % oil |
The Neapolitan figures come from the Associazione Verace Pizza Napoletana’s 2024 international regulations. The same document specifies 200–280 g dough balls for 22–35 cm pies and a 60–90 second bake with the oven floor at roughly 380–430 °C and the dome at about 485 °C. That heat is a large part of why the style can do without oil and sugar. King Arthur’s New York-style recipe is built for a home oven, can be refrigerated for up to three days, and adds exactly the sugar and oil the AVPN rules out. Their Detroit recipe bakes in a greased 9 × 13” or Detroit pan. The pan holds up a dough far too wet to stretch by hand.
Our presets are starting points, not copies of those recipes. Our Neapolitan 62 % sits just inside the AVPN’s 62.5 % upper limit, and our New York preset is wetter and leaner than King Arthur’s version. Every field can be edited.
The calculator starts from dough-ball weight. For round pies it uses a thickness factor, meaning grams of dough per square inch of pizza. The Neapolitan preset uses 2.2 g/in², so a 12” pie gets π × 6² × 2.2 = 249 g, inside the AVPN range. The default four Neapolitan balls come out as:
Flour 604 g
Water 374 g (62 %)
Salt 17 g (2.8 %)
Instant yeast 0.9 g (0.15 %)Under 1 g of yeast for four pizzas is why the calculator shows anything below 10 g to a tenth of a gram.
Converting between yeast types
Yeast swaps are where published advice disagrees most. Here is what the sources actually say.
- King Arthur’s professional yeast reference, by weight: to go from fresh yeast to active dry, “multiply the fresh quantity by 0.4”; to instant, “multiply the fresh quantity by 0.33”.
- King Arthur’s home guide to yeast says instant and active dry “can be substituted for one another 1:1”, with active dry rising a little more slowly. The often-quoted “use 25 % less instant” comes from the same guide, but only for bread machines.
- Red Star’s conversion chart: one ¼ oz packet = 7 g = 2¼ tsp of dry yeast = 0.6 oz of cake yeast (a third of a 2 oz cake). Either dry type may be substituted.
- The AVPN regulations put dry yeast at one-third the weight of fresh (“1 g dry yeast equals 3 g of fresh”).
Our yeast converter uses the professional reference’s factors exactly: active dry = 0.4 × fresh, instant = ⅓ × fresh. That makes instant ≈ 0.83 × active dry by weight, and active dry = 1.2 × instant. So the widely repeated “instant = 0.75 × active dry” is not what our tool does, and we couldn’t find it as a general rule from King Arthur or Red Star. It’s King Arthur’s bread-machine adjustment. Here’s what one packet converts to in our tool:
| Starting from | Active dry | Instant | Fresh |
|---|---|---|---|
| 1 packet active dry (7 g) | 7 g | 5.8 g (1⅞ tsp) | 17.5 g |
| 1 packet instant (7 g) | 8.4 g (2¾ tsp) | 7 g | 21 g |
| 1 cake fresh (0.6 oz / 17 g) | 6.8 g | 5.7 g | 17 g |
“Fresh ≈ 3 × instant” holds exactly in our tool (21 ÷ 7), matching the AVPN’s 1:3. The pizza dough calculator uses the same factors when you switch its yeast type, so the baker’s percentage it shows is always for the yeast you’re actually weighing.
Cake pans: the ratio is volume, not diameter
Cake batter has no flour-based convention. What carries over is the idea of scaling by one ratio: scaling factor = batter the new pan needs ÷ batter the old pan holds. For straight-sided pans, volume is just geometry. A round pan is π × r² × depth, and a square or rectangle is width × length × depth.
King Arthur’s pan-size article gives the check figure: an 8” square and a 9” round, both 2” deep, each hold “about 128 cubic inches or 8.8 cups.” Our cake pan converter gets 8.87 and 8.81 cups for the same pans. Some conversions it gives:
| From → to | Factor | Why |
|---|---|---|
| 8” round → 9” round | 1.27 | 9² ÷ 8² = 81 ÷ 64. One inch of diameter adds 27 % area. |
| Two 9” rounds → 9 × 13” | 0.92 | 127 in² of pan vs 117 in². |
| 8” square → 9 × 13” | 1.83 | Nearly double. |
| 9 × 13 × 2” → half sheet 13 × 18 × 1” | 1.00 | Same volume, half the depth. |
The last row is why the converter also comments on baking time. The half sheet holds exactly the same batter spread half as deep, so it bakes faster. Our converter says to start checking 10–15 minutes early, or at about two-thirds of the time for a thin sheet cake. Going deeper works the other way. Moving a 9 × 2” round recipe into a 9 × 3” springform, the tool suggests 10–15 minutes longer, and for pans 3” deep or 12”+ wide, dropping the oven 25 °F. King Arthur puts the same shift at about 10–15 minutes in each direction. Thinner bakes faster, deeper bakes slower, and the minutes are a starting point for checking, not a guarantee.
The fill rule, and why Bundts are different
King Arthur says most butter and oil cakes should fill the pan “between half and two-thirds full” so the batter has room to rise. That’s a ceiling, though, not what a typical recipe makes. Our converter’s batter-depth setting defaults to 36 % of the pan’s depth, about ¾” in a 2” pan, and lets you change it. Between two straight-sided pans that setting cancels out, so it doesn’t change the factor.
A Bundt pan’s fluted shape can’t be worked out from a ruler, so the converter uses the capacity stamped on the pan, which is its brim-full volume. King Arthur’s Bundt pan size post separates that from bakeable capacity, the batter it can take without overflowing. A 10-cup Bundt bakes about 6 cups and a 12-cup about 7¼. The advice is to keep the batter 1¼–1½” below the rim. Our tool uses 60 % of the stamped capacity: 6 cups for a 10-cup pan, 7.2 for a 12-cup. If your pan isn’t stamped, fill it with water by the cup.
Here is where the batter-depth setting starts to matter. King Arthur says a typical 9 × 13 or two-layer recipe makes about 6 cups of batter and fits a 10-cup Bundt unchanged. That’s where the 36 % default comes from: 6 cups is 37 % of a 9 × 13 × 2” pan’s 16.2 cups and 34 % of two 9” rounds’ 17.6. At the default, the converter assumes a 9 × 13 held 5.8 cups and gives 1.03× for a 10-cup Bundt. It assumes two 9” rounds held 6.3 cups and gives 0.95×, and the same 0.95× for 24 standard cupcakes at ¼ cup each. Both are close enough to King Arthur’s “use it as written.” If you know your recipe makes more or less batter than usual, change the setting to match. If your Bundt cakes dome over the rim, raise it, which tells the converter the original pan held more batter, so it scales the recipe down for the Bundt. For time, the converter follows King Arthur: bake a Bundt at the same temperature for about 30 % longer than the layer or sheet time.
Where to start
- Checking a recipe you already have: put flour, water and any starter into the bread hydration calculator and see its real hydration.
- Planning a loaf: set the loaf weight, hydration and starter in the sourdough calculator and weigh what it tells you.
- Pizza night: pick a style and a size in the pizza dough calculator, then adjust hydration to your flour and oven.
- Wrong yeast in the cupboard: the yeast converter.
- Wrong pan: the cake pan converter. Then use the unit converter if the recipe is in cups and you’d rather weigh. For any other ratio, the percentage calculator does the division.
All five run entirely in your browser. Nothing you type is sent anywhere.
If the bread is for a holiday table, the roast has its own arithmetic: turkey thaw time and safe temperatures works back from dinner to the day the bird goes in the fridge.