Three Foods, One Method

• 01 Fromm Four-Star Diner Classics Milo Swap Remove
• 02 Fromm Gold Nutritionals Digestive Support Low Fat Pâté Complete Swap Remove
• 03 Darwin's Natural Pet Products Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen Swap Remove

Three Foods, One Method. Each row is one measure; the best figure in a row is marked.

	01 Fromm Four-Star Diner Classics Milo Swap Remove	02 Fromm Gold Nutritionals Digestive Support Low Fat Pâté Complete Swap Remove	03 Darwin's Natural Pet Products Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen Swap Remove	+ Add A Food
01 Protein, Dry Matter Higher Is Better 26-32% dm reads typical mainstream.	01 Four-Star Diner Classics Milo 36.4%	02 Gold Nutritionals Digestive Support Low Fat Pâté Complete 40.9%	03 Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen 56.5% Best in this row
02 Carbohydrate Load, Dry Matter Lower Is Better Estimated, 6% ash assumed. iEstimated, not measured. It is what is left of the guaranteed analysis once protein, fat, fiber and moisture are taken out, with ash assumed at 6%. That leftover is mostly starch but also carries sugars and the fiber a crude-fiber test misses, so read it as a close estimate rather than an exact figure. Measured as fed, the same basis as the bag.	01 Four-Star Diner Classics Milo 21% 0.5% as fed Best in this row	02 Gold Nutritionals Digestive Support Low Fat Pâté Complete 25% 1.5% as fed	03 Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen 13% 0.0% as fed
03 Named Proteins In The First Five Count of named animal proteins. iThe count matches any named animal word in the opening five ingredients, so it also counts animal fats and flavourings as proteins. It is shown because it is one of the five measured lines, but it is too rough to call a winner on.	01 Four-Star Diner Classics Milo 4 of 5	02 Gold Nutritionals Digestive Support Low Fat Pâté Complete 4 of 5	03 Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen 3 of 5
04 Flagged Additives Level Lower Is Better Matched against the flag list.	01 Four-Star Diner Classics Milo None	02 Gold Nutritionals Digestive Support Low Fat Pâté Complete None	03 Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen None
05 Brand Recall History Lower Is Better Manufacturer, not recipe.	01 Four-Star Diner Classics Milo 3 2 in the last 3 years	02 Gold Nutritionals Digestive Support Low Fat Pâté Complete 3 2 in the last 3 years	03 Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen None Best in this row
Starch, As Fed iEstimated, not measured. It is what is left of the guaranteed analysis once protein, fat, fiber and moisture are taken out, with ash assumed at 6%. That leftover is mostly starch but also carries sugars and the fiber a crude-fiber test misses, so read it as a close estimate rather than an exact figure. Measured as fed, the same basis as the bag.	01 Four-Star Diner Classics Milo 0.5%	02 Gold Nutritionals Digestive Support Low Fat Pâté Complete 1.5%	03 Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen 0.0%
Kcal / Cup	01 Four-Star Diner Classics Milo --	02 Gold Nutritionals Digestive Support Low Fat Pâté Complete --	03 Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen --
First Five Ingredients	01 Four-Star Diner Classics Milo Pork, Pork Broth, Beef Broth, Beef, Pearled Barley	02 Gold Nutritionals Digestive Support Low Fat Pâté Complete Tuna, pumpkin, chicken broth, whitefish, chicken liver	03 Darwin’s Natural Pet Products Intelligent Design for Dogs Joint Support Grain-Free Frozen White fish (Pollock), Turkey neck, Turkey heart, Broccoli, Sweet potato

+ Add A Food

Every figure here is the food's own. Protein and carbohydrate are converted to dry matter so foods of different moisture can be set beside each other; starch is estimated from the guaranteed analysis with ash assumed at 6.0%, and the recall record belongs to the manufacturer rather than to the recipe. Two rows are printed but never marked: the named-protein count, because the detection behind it is too rough to call a winner on, and calories, because neither more nor fewer is better without knowing the dog.