Doneness Temperature Chart with Carryover Pull Temperatures

Doneness is a two-number problem. There is the finished temperature you want on the plate, and there is the pull temperature at which you take the meat off the heat so that carryover brings it up to that finish. Confuse them and you will overshoot every time. Medium rare finishes at 135°F, so a steak comes off around 130°F, while a large roast aiming for the same finish comes off nearer 120°F because it carries so much more stored heat.

A definition, stated plainly because everything below depends on it. Carryover cooking is the continued temperature rise after the meat leaves the heat, driven by heat already stored in the outer layers moving inward. It scales with mass, not with the doneness you are aiming for, and it is the reason a chart of finished temperatures alone is not usable.

None of this applies to barbecue cuts. 17 of the cuts in our data set target 195°F or above, and they are cooked to collagen breakdown rather than to a preference. A rare brisket is not a doneness level, it is a chewing problem.

How do you read this chart?

Work from right to left, not left to right. Start with the finished temperature you want, find the row for the size of cut you are actually cooking in the carryover table below, and subtract. The result is the number you watch for on the probe. The conventional pull column in the first table assumes a steak-sized piece of meat, which is why it only removes 5 degrees, and it will leave you overcooked on a roast.

• Pick the finish first. That is the eating decision, and it is the only part of this chart that is a preference.
• Match the size band honestly. A steaks and chops, under 2 lb rises 3°F to 5°F. A large roasts and whole birds, 8 to 15 lb rises 10°F to 15°F. That is a swing of ten degrees of final doneness on the same pull number.
• Subtract the top of the range, not the middle, if you care more about not overshooting than about hitting the exact centre. Undershooting is fixable with more heat. Overshooting is not fixable at all.
• Check the USDA column before you commit. If the band you picked does not meet the minimum, you are making a risk decision on behalf of everyone you are feeding.
• Probe twice, in different places. One reading is a data point, two is a measurement.

Two things fall out of that grid immediately. First, on a large roasts and whole birds, 8 to 15 lb the pull temperature for medium rare is 120°F to 125°F, which feels alarmingly low the first time you do it and is correct. Second, the well done column on a large roast asks you to pull at 150°F, which is barely above the poultry minimum, and it illustrates why big roasts are so easy to overcook: the window between bands is narrower than the carryover itself.

Rest also has to be real for carryover to complete. A steak wants five to ten minutes, a roast twenty to thirty, and a large bird or brisket longer again. Tenting loosely with foil is fine; wrapping tightly traps steam and softens any crust you built. If you are holding for a while, use a dry cooler packed with towels, which will keep a wrapped cut above 140°F for hours without cooking it further.

Which smoked cuts are cooked to a doneness preference?

Most of what goes on a smoker is not. Barbecue cuts are collagen conversions with a temperature floor of about 195°F. But 14 of the cuts in our data set are cooked like steaks, and those are the ones this chart is for: prime rib, tri-tip, top round, reverse-seared ribeye, leg and rack of lamb, venison, pork loin and thick pork chops.

Notice how many of those rows sit below their own USDA minimum. That is the convention in restaurants and it is what most published recipes assume, and it is still a risk decision every time. Pork loin and thick pork chops are the honourable exceptions: their conventional pull of 145°F is both the safety floor and the point past which they dry out, which makes them the easiest cuts on this page to get right.

How does the reverse sear change the arithmetic?

A reverse sear splits the cook into a low smoking phase and a short hot finish, and the finish adds its own temperature rise on top of carryover. On a ribeye steak, reverse seared the convention is to pull at around 125°F from the smoker, because the sear will add roughly 10°F to 15°F before carryover is even counted.

Practically that means three numbers instead of two: the smoker pull, the post-sear reading and the finished temperature. Rest the meat briefly between the smoke and the sear so the surface dries, because a dry surface browns and a wet one steams. Get the pan or grill genuinely hot before the meat touches it, and keep the sear short: sixty to ninety seconds a side is usually the whole job. A pair of long locking tongs and a set of certified heat resistant gloves make that transfer far less exciting than it otherwise is.

Prime rib is the version of this that people most often get wrong, because it combines the largest carryover band with the highest stakes. A 12 lb standing rib roast is squarely in the large roasts and whole birds, 8 to 15 lb band, so it will climb 10°F to 15°F after it comes off. Pull it at the number in the carryover table, not at the number you want to see when you carve.

Does altitude change any of these temperatures?

The doneness temperatures themselves do not change. Proteins denature at the same temperature at 9,000 feet as at sea level, so medium rare is still 135°F and the USDA minimum is still 145°F. Two practical things do change, and both of them affect how much you can trust your instrument and your schedule.

The first is calibration. Boiling water is only a valid reference at sea level, and the boiling point falls by roughly 1.8°F per 1,000 feet. If you calibrate a thermometer against boiling water in a mountain town and assume 212°F, you will bias the instrument several degrees high, which is exactly the wrong direction for food safety.

The second is the cook itself. Lower atmospheric pressure speeds evaporation, which strengthens the evaporative cooling that causes the stall, so long cooks at altitude commonly run longer than the same cook at sea level. That does not change the target, it changes the schedule, and it is another argument for a leave-in probe such as the MEATER Plus or a two-probe base like the ThermoMaven WiFi rather than a clock.

What are the common mistakes with a doneness chart?

• Using the finished temperature as the pull temperature. This is the single most common error and it costs you a whole doneness band on anything larger than a steak.
• Applying the steak carryover figure to a roast. A large roasts and whole birds, 8 to 15 lb carries two to three times the rise of a chop. Same chart, different row.
• Probing into fat or against bone. Both read high, and both have you pulling early with total confidence. Thickest part of the muscle, nothing else.
• Trusting the dial thermometer in a lid. Those measure air near the top of the chamber, not food, and they are frequently 25°F or more out. They are not doneness instruments and this site never asks you to use one.
• Never calibrating. A probe that has spent a season in grease and 250°F air drifts. Run an ice bath check before a big cook, or buy an instrument with a calibration function such as the TempPro TempSwift.
• Applying a doneness band to ground or injected meat. A medium rare burger and a medium rare injected roast are not the same category of decision as a medium rare steak. Both need 160°F.
• Applying it to poultry at all. There is no rare chicken. 165°F, every time, in the thickest part of the thigh.
• Skipping the rest. Carryover is not optional physics you can outrun by slicing quickly, and the 3 minute rest at 145°F is part of the USDA requirement for whole cuts.
• Cooking to colour. Smoked meat keeps a pink smoke ring well past well done, and cured meat stays pink permanently. Colour tells you nothing on a smoker. Only the probe does.

Why does colour lie on a smoker?

On a pan-seared steak, colour is a rough proxy for temperature. On a smoker it is not, and this catches people out constantly. Nitric oxide and carbon monoxide from combustion react with myoglobin near the meat surface to fix a pink layer that will not turn grey no matter how far you cook it. That is the smoke ring, and it is a picture rather than a flavour.

The reverse error also happens. Meat from an older animal, or meat that has been frozen and thawed, can look considerably darker at the same internal temperature. Cured products stay pink permanently because the curing salt fixes the colour chemically. In every one of those cases the only honest answer is the number on a calibrated probe, taken in two places, away from bone and fat.

Related charts and guides

• Internal temperature chart for the full USDA table and the texture targets alongside it
• Smoking time and temperature chart for how long it takes to reach these numbers
• Best meat thermometer for choosing the instrument
• Food safety for smoking for the danger zone and the 4 hour rule
• Temperature control guide for holding a steady chamber