Soft Pretzels Lab173 recipes

Soft Pretzels Lab

Three overlapping styles in 173 soft pretzel recipes

Leave the recipe names aside and group the 173 recipes by their 5 ingredient ratios alone, so that recipes built alike land together. Each group is named only by the numbers: the ingredients where its recipes sit furthest from the middle of all 173.

The data splits best into three groups, but they overlap: a silhouette of 0.24, on a scale where 0 is no separation and 1 is complete. Read them as tendencies within one broad spread rather than separate kinds of soft pretzel.

The recipes on one map

Every recipe, placed on the two directions along which the ratios vary most (a principal component projection). Close dots are recipes built alike. The lettered rings mark the middle of each group.

ABC
Across →
More water and yeast, less salt to the right; 33% of the variation.
Up ↑
More sugar and butter upward; 22% of the variation.

The three groups

AMore salt, less water

82 recipes · 47% of all 173 · median oven 425°F (220°C), 15 min

Where it differs from all recipes

  • Saltmedian 2% vs 1.7% for all

    average rank: 67th percentile of all recipes

  • Watermedian 60% vs 63% for all

    average rank: 35th percentile of all recipes

  • Yeastmedian 1.2% vs 1.4% for all

    average rank: 36th percentile of all recipes

Its median recipe, at 500 g flour

Median recipe of group A in grams, against the median of all recipes
IngredientGroup AAll
Water300 g315 g
Butter46 g35 g
Sugar11 g12 g
Salt10 g9 g
Yeast6 g7 g

Each line is the group's median for that ingredient, scaled to 500 g of flour.

Closest to the middle of the group

BMore water, less salt

71 recipes · 41% of all 173 · median oven 415°F (215°C), 14 min

Where it differs from all recipes

  • Watermedian 71% vs 63% for all

    average rank: 71st percentile of all recipes

  • Saltmedian 1.2% vs 1.7% for all

    average rank: 31st percentile of all recipes

  • Buttermedian 2.8% vs 7% for all

    average rank: 32nd percentile of all recipes · none in 39% (all recipes 20%)

Its median recipe, at 500 g flour

Median recipe of group B in grams, against the median of all recipes
IngredientGroup BAll
Water355 g315 g
Butter14 g35 g
Sugar12 g12 g
Salt6 g9 g
Yeast7 g7 g

Each line is the group's median for that ingredient, scaled to 500 g of flour.

Closest to the middle of the group

CMore sugar and butter

20 recipes · 12% of all 173 · median oven 425°F (220°C), 15 min

Where it differs from all recipes

  • Sugarmedian 11% vs 2.4% for all

    average rank: 94th percentile of all recipes

  • Buttermedian 9.9% vs 7% for all

    average rank: 65th percentile of all recipes · none in 15% (all recipes 20%)

  • Watermedian 60% vs 63% for all

    average rank: 36th percentile of all recipes

Its median recipe, at 500 g flour

Median recipe of group C in grams, against the median of all recipes
IngredientGroup CAll
Water300 g315 g
Butter50 g35 g
Sugar55 g12 g
Salt8 g9 g
Yeast8 g7 g

Each line is the group's median for that ingredient, scaled to 500 g of flour.

Closest to the middle of the group

How the groups are found

Each ratio is converted to a z-score (how many standard deviations a recipe sits from the mean), capped at 3 so one extreme recipe cannot claim a group, and k-means sorts the recipes into k groups, trying k from 2 to 4. Each k runs from 16 fixed starting points and keeps the tightest result, so the page is the same on every build. No group may hold fewer than 18 recipes.

The number of groups is the one with the best silhouette, which compares how close each recipe is to its own group with how close it is to the next one. By number of groups: 2 groups 0.22 · 3 groups 0.24 · 4 groups 0.22. At 0.25 or more the page calls the groups separate; from 0.18 to 0.25, overlapping; under 0.18, no clear styles.

A group's name lists the one to three ingredients where its recipes rank furthest from the middle of all 173, on average. “More” and “less” are relative to the whole dataset, and the names say nothing about texture or taste, which the ratios alone do not show. The median recipe of a group is, like the typical recipe on the overview, the middle of each ingredient separately, not one real recipe. The recipes listed under each group are the ones nearest its center, at most two from any one site.