The number that ruins barbecues
Almost every guide to this question answers it in cooked pounds and then stops. A third of a pound per person, they say, and they are right. The trouble is that nobody sells cooked brisket by the pound. You buy raw meat, you cook it for twelve hours, and somewhere in those twelve hours a large part of what you paid for leaves as steam and rendered fat and the trimmings in your bin.
Here is the whole problem in one line. A whole packer brisket comes off the pit at roughly half the weight it went on the scale at. So if you take the standard advice, work out that 40 guests need 13 pounds of cooked brisket, and buy 13 pounds of brisket, you will serve about 6 and a half pounds. You will be short by half, in front of 40 people, at a point in the afternoon when there is nothing at all you can do about it.
This calculator does that division for you, cut by cut, because different meats lose wildly different amounts. Cooked-ready hot links barely lose anything, because you are reheating something that was already cooked. A whole packer brisket loses about half. Spare ribs give back well under half, because you are buying a great deal of bone. A whole chicken, counting the frame you are never going to serve, lands around 40 percent. Applying one blanket rule of thumb across a mixed menu is how a plan that looked careful comes up short on the one meat everybody wanted.
The formula
children count as half an adult
raw meat to buy = cooked meat needed ÷ that cut's cooked yield
rounded UP to the next half pound, always up
whole pieces = raw pounds ÷ the heavy end of that cut's usual piece weight
Serving size is a third of a pound of cooked meat for a standard plate with sides, half a pound for a buffet or a hungry crowd, two thirds for a genuinely big eating crowd, and a quarter pound for sandwiches. Cooked yield is the fraction of the weight you paid for that reaches a plate, after trimming, after the bone, and after the cook. That last definition is the one this page is fussy about, and the section near the bottom explains why it matters more than any other number here.
The rounding only goes one way, and that is deliberate. Running out of food is the failure people remember for a decade. Leftover barbecue is a sandwich.
Worked example
40 guests, barbecue as the only main, a plate split evenly between brisket and pulled pork, two or three sides.
Each adult gets the standard 0.33 pounds (5.3 ounces) of cooked meat, so 40 guests need 13.33 pounds of cooked meat. Split evenly, that is 6.67 pounds of each.
Now divide by what survives the cook. Brisket at 50 percent: 6.67 ÷ 0.50 = 13.33 pounds raw, rounded up to 13.5 pounds. Pork shoulder at 52 percent: 6.67 ÷ 0.52 = 12.82 pounds raw, rounded up to 13 pounds. Total: 26.5 pounds of raw meat to serve 13.33 pounds cooked, which is very nearly 2 times the weight you will actually put on the table.
At the counter that is one whole packer brisket (they run 12 to 16 pounds, and you want one from the heavy end) and 2 bone-in Boston butts. Weight the plate 60 percent brisket instead and the total barely moves, to 16 pounds of brisket and 10.5 of pork, because those two cuts happen to lose almost the same fraction. Swap the pork for spare ribs and it moves a lot, because ribs are mostly bone.
The number to hold onto is the one nobody prints: 26.5 raw to serve 13.33 cooked. Every guide that tells you a third of a pound a head is telling you the truth about the second number and leaving you to guess the first.
How much meat per person, and what actually moves it
The planning figure the catering world agrees on is a third of a pound of cooked meat per adult when barbecue is the main course and there are sides on the table. Half a pound is the figure for a buffet, a long day outdoors, or a crowd you know eats. Two thirds of a pound is what gets quoted for a genuinely hungry crowd, and on this page you reach it by choosing hearty appetites with barely any sides. A quarter pound is a sandwich portion, which is the right number for a lunch service and the wrong number for a wedding.
Children under about 12 eat roughly half an adult portion. Sonny's puts it as half a portion of meat and a full portion of sides. Dickey's puts it as one adult meal feeding two kids under 12. It is not quite unanimous, and here is the honest wrinkle: USDA's own child nutrition meal patterns do not support a 50 percent cut, since the school lunch weekly minimums for younger children run about 0.8 of the high school figure and the CACFP pattern is identical for ages 6 to 12 and 13 to 18. Those are nutritional minimums rather than appetite estimates, so they do not refute the catering convention, but they are the closest thing to a measured standard and they disagree with it. We use half. Teenagers are not children for this purpose. Put them in the adults box and thank us later.
Sides do less than everybody says, and this is the one place we part company with every other calculator. You will find the figure "reduce meat by 20 to 25 percent if you serve three or more sides" on a dozen barbecue calculators. We went looking for its source and there isn't one. The sites cite each other, and the one that does explain itself attributes the number to "pitmaster consensus". Every credible source, meaning the people actually selling barbecue, says only that sides reduce meat consumption slightly, and none of them gives a number. Meanwhile the only peer-reviewed evidence on the question runs the other way: a 2021 meta-analysis in the American Journal of Clinical Nutrition found that greater food variety is associated with increased intake, not less. So our adjustment is capped at 10 percent, half of what everyone else applies, you choose it rather than having it applied behind your back, and now you know exactly how much confidence to place in it.
Another main course changes everything, and this one is actually quantified. If there is a pasta bake or a fish option or a taco bar beside the smoker, the barbecue is sharing the plate. Meathead Goldwyn puts a serving at 8 ounces of cooked meat with one main dish and 6 ounces when two or more main courses are offered, and 6 over 8 is exactly the 25 percent reduction this page applies. That is a published figure from a named authority rather than a guess of ours, which makes it the best-supported adjustment on the page. It is also the most dangerous setting here, because being wrong about it means being short. If you are not sure the other main will really pull its weight, leave this on "barbecue is the meal".
And the honest caveat that belongs on every page like this one: appetites are not a constant. A wedding eats less than a football tailgate. A hot afternoon eats less than a cold one. A crowd that has been drinking for four hours eats more than you planned. These are planning figures, not physics.
Why a caterer plans by share of the plate
Ask the internet how much brisket per person and you get a third of a pound. Ask how much pulled pork and you get a third of a pound. Ask about ribs and you get four bones. Serve all three and follow all three answers and you have bought about a pound of meat a head, which is roughly three times what anybody is going to eat.
The fix is the one professionals use: decide the size of the plate first, then decide how that one plate is divided. A guest who piles on brisket takes less pork. That is why this calculator asks for a share of the total for each cut rather than a separate serving figure per meat, and why leaving the shares blank simply splits the plate evenly between whatever you picked.
A word on how to choose shares. If one meat is the star, give it 50 to 60 percent and let the rest fill in. Smoked sausage is the classic filler because it is cheap, it is nearly impossible to ruin, and being sold already cooked it loses almost no weight, so a 20 percent sausage share is the cheapest insurance on the menu. Fresh bratwurst is a different proposition entirely and yields barely half as much. If you are cooking for a crowd that includes people who do not eat pork or do not eat beef, that is a reason to run two meats rather than one, and to skew the shares toward whichever one more of your guests can eat.
Ribs are counted in bones, not pounds
Nobody stands at a pit thinking in pounds of rib. They think in racks and bones, and the published rule is 3 to 4 bones a person when ribs are part of a spread, 5 to 6 when ribs are the main event. Baby backs run a little higher because the bones are smaller: half a rack, or about six ribs, is the usual quote.
Those rules and the weight arithmetic on this page were built from completely different directions, so it is worth showing you that they land in the same place. The federal buying guide puts pork spare ribs at 0.39 pounds of cooked lean meat per pound purchased, the lowest federally measured figure for any cut on this page. A St. Louis cut rack is about 3 pounds raw with about 12 bones, so 0.39 puts about 1.6 ounces of cooked meat on each bone. Divide a standard third-of-a-pound serving by that and you get 3.4 bones per person. Divide a hearty half-pound serving by it and you get 5.1. Those are the two published rules, 3 to 4 and 5 to 6, reproduced by arithmetic that never looked at them. When two independent methods land on the same answer it is usually because both are describing something real.
Two corrections while we are here, both from the actual federal meat specifications rather than from barbecue folklore. First, trimming spare ribs to St. Louis style removes no rib bones at all. The cut takes off the sternum, the costal cartilage, and the rib tips, which is cartilage and one flat bone, so a St. Louis rack has exactly the same rib count as the spare rack it came from and simply weighs less. Second, there is no such thing as a standard rack weight. The federal specification for spare ribs requires at least 11 ribs and says in plain words that the purchaser may specify the number, leaving weight entirely alone. Every "a rack weighs three pounds" claim, including the 3 pounds this page uses, is trade convention. Retail racks in a supermarket today routinely run heavier than the foodservice specs the folklore was built on, which is another good reason to buy ribs by the rack and count the bones yourself.
One thing the rib numbers make obvious: a rack of ribs is a light portion by weight and an expensive one by the pound, because you are buying a great deal of bone. That is not an argument against ribs. It is an argument for knowing that a rib plate costs more per satisfied guest than a brisket plate, and for pricing accordingly if you are selling it.
What the cook does to your plan
Hours per pound is not a real rule. Cook time follows thickness far more than weight, so a 16 pound packer is not twice the cook of an 8 pound flat, and two packers side by side take roughly as long as one. Meathead Goldwyn puts the determinants as meat thickness, cooker temperature, and whether you wrap, with humidity, air temperature, wind, rain, and the grade of the meat all playing a part, and then gives a 14 to 18 hour range for one brisket at 225 degrees. The constraint on a big cook is pit space, not hours. If you need double the meat, you need double the grate, not double the day.
Plan to be done early, never on time. This is the single most useful sentence in barbecue. Finished meat holds for hours wrapped and packed in a dry cooler, and most of what people call resting is really the tail end of cooking: in a long cook, the great majority of the tenderising happens in the final push and the early part of the hold, not in the first eight hours. Meat that stalls at hour eight with guests already parked outside cannot be rescued by anything. Aim to be finished two hours before you serve and treat that buffer as part of the recipe.
Hold it hot, and know the number. USDA puts the danger zone at 40 to 140 degrees F, so cooked meat should be held at 140 degrees F or above until it is served. Once it leaves the heat, do not let it sit out more than 2 hours, or more than 1 hour if it is above 90 degrees outside. There is a wrinkle worth knowing if you are cooking commercially: the FDA Food Code, which is what a health inspector actually enforces on a caterer, sets hot holding at 135 degrees and explicitly allows 130 for roasts, and a whole brisket or pork butt is a roast under that section. We publish 140 because it is the consumer guidance, it is the higher number, and the error here only runs one way. And notice what all of this means for the practice pitmasters already follow: holding a wrapped brisket at 150 to 160 in a cooler for three hours is comfortably above every one of those lines. In barbecue the danger is in the cooling, not the holding.
Rest, then slice, and only slice what you are serving. Sliced brisket dries out quickly, and pulled pork sitting in a pan loses moisture fast. Slice to order if you can, and pour the rendered juices back into the pulled pork rather than leaving them in the pan. That last trick is worth several percent of yield on its own, which on a big cook is a genuine pound or two of food.
Where the published yield numbers disagree, and why
You will see cooking-loss figures quoted everywhere, and a lot of them are the wrong kind of number for this job. The federal cooking-yield tables define yield as the hot cooked weight divided by the raw weight of that same trimmed sample. That measures the cook and only the cook. It does not include the fat cap you cut off, the deckle you threw away, or the bone you are never going to serve. So "meat loses 30 to 40 percent in cooking" is a perfectly true sentence about a trimmed sample, and applying it to the weight on the package quietly asks it to cover the fat cap and the bone as well. It is an easy join to make, and the error runs in only one direction: the one that leaves you short.
This page uses purchase to plate: cooked, edible meat as a fraction of the weight printed on the package you paid for. On that basis the published brisket figures run from 36 percent, which is what a working Austin pitmaster gets in sellable slices at commercial trim standards, to 69 percent, which is the federal buying guide's number for brisket bought practically free of fat. That is not a disagreement about physics. It is four different questions wearing the same word. The figures that are genuinely about a whole packer cluster between 43 and 57, and we use 50, with every result showing what the ends of that range would cost you. The federal buying guide disagrees with itself here in the most instructive way possible: the same cut is 0.46 pounds per pound purchased at quarter-inch trim and 0.69 practically free of fat. Trim spec, not cooking, is what moves brisket yield most.
Pulled pork is the one where two numbers get swapped constantly, and it is an easy swap to make, because it is also the place where the sources agree best once you sort them out. The federal buying guide gives 0.52 pounds of cooked lean meat per pound as purchased for a bone-in shoulder and 0.60 for boneless, and a documented cook weighed an 8.37 pound bone-in butt down to 4.42 pounds of pulled pork, which is 52.8 percent. Federal data and a man with a scale agree to within a point. The 60 percent figure printed on most barbecue pages is the boneless number. Apply it to the bone-in butt most people actually buy and you are about 15 percent short before you light the fire.
Sausage carries the other trap, and it is bigger than the brisket one. Fully cooked links get no yield reduction at all in the federal tables: a pound of frankfurters is listed as eight two-ounce servings, which is sixteen ounces per pound. Raw sausage is 0.47 pounds of cooked lean meat per pound purchased for market-style pork, 0.62 for Italian style. Those are the same word for two things that differ by a factor of nearly two, which is why this calculator lists them as separate cuts. Classic barbecue hot links are sold cooked and only reheated. Bratwurst is not.
Where we could not find a measured figure, we say so rather than inventing precision. The baby back and chicken quarter yields are ours, composed from rack geometry and federal cook-loss data. Beef plate ribs are the shakiest number on the page and we would rather admit it: the federal study for boneless beef short ribs reports a standard deviation of 11.76 with an observed range from 44.9 to 97.3 percent, against brisket's standard deviation of 3.04. A single confident number for beef ribs would be dishonest, so that cut gets the widest band here. What we have not done anywhere is round a yield in the flattering direction, because the error only hurts in one direction, and it hurts in front of everybody you invited.
If you are scaling a recipe for the sides rather than the meat, the recipe scaler handles the parts that do not multiply cleanly, and the cooking measurement converter sorts out an old recipe's units.