Bringing Meat to Room Temperature Before Cooking: Essential or Not? We Tested It

Tempering meats by setting them out to warm at room temperature before cooking is one of those things that some cooks swear by and others swear off. Those who are in favor of tempering tend to insist on it with a level of conviction that can make you feel like this one step is the difference between success and failure—if you want a piece of meat to cook gently and evenly, let it warm up first. And then you have others, including J. Kenji López-Alt here on Serious Eats, who have long argued that tempering is futile—even after a lengthy rest, the meat will not have come to room temperature, nor will it make much of a difference in the final cook.

More recently, Chris Young—one of the coauthors of the Modernist Cuisine series, as well as the creator of the Combustion Inc. Predictive Thermometer—has argued that resting meat before cooking really does make a difference. In a convincing video, he demonstrates what seems to be a dramatic difference between a tempered and straight-from-the-fridge steak. The tempered steak looks far more even and has much less of a gray band of overcooked meat around the exterior—results that seem to speak for themselves.

Still, I had some questions. I noticed in Young's video that he didn't appear to flip the steaks repeatedly, a highly recommended technique that can help even out the internal cooking without compromising the crust. Would a refrigerated steak look dramatically more unevenly cooked than a tempered one if you flipped them both repeatedly throughout cooking? And what about larger roasts? Is there any argument for or against tempering as the cuts get bigger? That's worth knowing, since a prime rib isn't the kind of thing you want to accidentally screw up (although, you have way more latitude than most people claim on one of those).

Much of the dissonance between Kenji's and Chris's explanations centers on questions of thermodynamics. Kenji's argument focuses on how slowly meat actually warms to room temperature (often, he says, too slowly to make much difference while staying within acceptable food-safety limits for leaving food out) and on the idea that, when roasting, the bigger hurdle for browning is the energy required to drive off surface moisture, not the starting temperature of the meat itself. Young, on the other hand, frames the issue in terms of temperature gradients: Because heat flow into food is proportional to the temperature difference between the heat source and the meat, a colder steak experiences a larger initial influx of heat at its surface, which steepens the gradient from surface to center and yields a thicker band of overcooked meat by the time the core reaches doneness.

Look, I'm a scientifically literate guy, and I've written my fair share of food science explainers before, but I don't have the level of expertise in thermodynamics to even attempt to sort out who's right here, or whether they're both right depending on sneaky variables and nuances, or what. What I can do is run my own tests, capture the data, and see what the result looks like on the plate.

A Note on Testing Methodology

For each of these tests, I left one sample at room temperature for a specified amount of time (listed below) and its counterpart in the fridge until it was time to cook. I did not pre-salt the steaks because they would not rest long enough for a proper dry-brine; instead, I salted all steaks immediately before putting them in the pan. I did pre-salt the chickens and pork roast, since I had time to let them dry-brine for several hours.

To track and record temperatures, I used a Combustion Inc. probe thermometer inserted into each sample. The Combustion Inc. thermometers have multiple thermocouples along their length, allowing them to take multiple readings on each probe. Connected software is then able to calculate a more accurate estimate of the true core temperature via the multiple data points.

For the steak tests, I seared them in a cast iron pan set on a Breville Control Freak induction burner, which is capable of precise and consistent temperature control, largely eliminating the risk of variation in the heat source. For the steak test that involved flipping the samples in 30-second intervals, I used a metronome set to beep every 30 seconds, ensuring my flipping was consistent and not haphazard.

Test 1: Steak, 30-Minute Temper, Flipped Once

In my first test, I attempted to duplicate Chris Young's experiment in his video by searing 1.5-inch-thick strip steaks, only turning them once during cooking. Young says in his video that he found meaningful results after a minimum 15-minute temper, so I let one steak sample sit out for 30 minutes, then brought the other out from the fridge and cooked them together.

In the graph below, you can see the core temperature data for each sample (I'm omitting the surface temp data because the probe was sticking out the side of the steak, making the reading less useful—it wasn't tracking temps at the searing interface but instead reflecting whatever heat was being convected and conducted up the side of the steak, which is subject to variations in airflow around the steak and the placement of probe where it exited the steak).

What we see in the core temperature readings above is interesting: The tempered steak starts out roughly 15°F warmer than its refrigerated counterpart, and for much of the cook, it seems to maintain that lead. But then toward the end, the refrigerated steak surges ahead and catches up with the tempered steak—they ended up reaching the target core temp of 115°F within less than 30 seconds of each other (close to where the two lines cross in the graph).

So in this test, the tempered steak saw no real cooking speed advantage, despite starting with a 15-degree lead.

But the effect of tempering isn't just about cooking speed; it's also about how much of a gray band of overcooked meat you get around the edges. On the plate, the differences in my samples were subtle. As you can see in the picture above, both steaks had a similar gray band around their exterior.

To the degree there was any difference at all, it was that the refrigerated one had a slightly more medium-rare-seeming center, even though both were removed from the pan at the same core temperature. But the difference was so minute that it's hard to say whether it was due to a tempering effect or variations in the steaks themselves (marginal thickness differences, slightly different marbling patterns, etc.) or subtle heat variations in the pan (despite my efforts to control it).

As we stood there scrutinizing the steaks and considering these almost imperceptible differences, writer Lee Musho, who was in the test kitchen with us that day, said the truest thing of all: "If we have to discuss it this much, there's probably not much of a difference."

Test 2: Steaks, 1-Hour Temper, Flipped Every 30 Seconds

Next, I repeated the steak test, except this time I flipped each steak every 30 seconds. Flipping steaks as they sear can help minimize temperature gradients within the meat and reduce the formation of a gray band under the surface. I also let the tempered sample rest for a full hour this time, but as we can see in the graph below, it didn't get a whole lot warmer in that time.

Once again, the temperature graph shows the tempered steak starting warmer and maintaining that difference throughout almost the entire cooking time. But as the steaks neared doneness, they began to converge, just as we'd seen in the first test. This time, the tempered steak finished first, but by only about 40 seconds. One full hour of resting at room temp to shave 40 seconds off the cooking time isn't a great trade-off to me.

The steaks themselves were even more indistinguishable than the first round, an indication that frequent flipping may more than compensate for whatever difference one might otherwise see between a tempered and cold steak. The photo below shows the frequently flipped steaks, and I'd be hard-pressed to pick the tempered one out of the lineup (it's on the left).

Test 3: Rolled Porchetta Roast, 2-Hour Temper

The steaks didn't show much evidence that tempering offered a clear benefit. What about a roast? I bought a large rolled porchetta roast, cut it exactly in half to keep the pieces the same size, and let one half rest for two hours out of the fridge.

In the graph below, which shows both the core temps and the surface temps for each roast, you can see that the tempered roast (red and green lines) is not much more than 10°F warmer than the one that stayed in the fridge, even after a full two hours at room temperature. As they cook, the fridge sample gradually closes the gap, and they finish within minutes of each other at the set target temp of 135°F.

When I cut into the roast to examine and taste it, I couldn't tell any difference. Both roasts were juicy inside, deeply browned on the outside, and indistinguishable from each other, as seen in the photo below. This was an indication that tempering larger roasts makes little difference in how they cook—the roast is too large, and the heat penetration too gradual to have much effect.

Test 4: Whole Chicken, 3-Hour Temper

Not wanting to draw any conclusions from just one pork roast, I decided to try chickens as well. Lots of people talk about the need to pull their holiday turkey out of the fridge well in advance of roasting, and I figured a chicken was a good enough stand-in. If it didn't work for chicken, it probably wasn't going to work for a larger turkey, which will warm to room temperature even more slowly and take longer to cook, minimizing the potential impact of tempering.

In the graph below, you can see the tempered sample, which I let sit out for three hours. It got the warmest of all my samples in these tests—a full 25°F head start in warming. As it cooked, it maintained its core temp head-start through to the end, finishing about eight minutes before the refrigerated bird.

After the chickens rested and were carved, though, I couldn't see any difference between the two birds. The leg meat was similarly cooked, the breast similarly juicy. The skin wasn't the best ever on either bird, but that was because I jammed two birds on a single baking sheet, which increased local humidity around their sides and decreased browning there on both birds.

Conclusions

My tests didn't convince me that tempering is a helpful technique for speeding cooking time (you certainly don't reduce the cooking time by anywhere near the time it takes to temper first) or for achieving a more even, gentle doneness gradient inside the meat. The only differences I really observed in the finished meats were in the steak, which I flipped only once; even then, it was not by any means obvious enough to declare it superior. Flipping the steak frequently in the pan was enough to mitigate even that minor difference.

As for roasts, I saw no difference between my tempered and refrigerated samples, except for shaving several minutes off the chicken roasting time. But that chicken had to sit out for three hours at room temperature first, which is not practical and not exactly appealing from a food-safety standpoint (more on that below).

I'm not sure why I was unable to replicate Young's results in my tests. I don't want to say he was wrong, because Young is an exacting food scientist, and there are enough variables across meat types, specific cuts and animals, cooking methodology, and more to make it possible that there could be more of an effect in some instances than others. Maybe tempering does help in some setups.

But these tests have not convinced me that tempering offers a clear or sufficiently repeatable benefit to make it a practice I'll routinely adopt or recommend, especially for roasts, which seem too large to benefit. There are more important things, like dry-brining, steak-flipping, and cooking to a proper doneness that will deliver the consistently good results we all want.

Is Tempering Safe?

One final note, since some folks may be wondering about how safe tempering really is. The general guidance for home kitchens is to not leave any food out in the "danger zone" of between 40°F and 140°F for more than two hours. There are slightly more flexible rules for commercial kitchens that allow holding food at room temp for up to four hours, but those are tricky to apply to home kitchens.

Large, intact cuts of meat like steaks, roasts, and chops tend to carry most of the risk on the exposed surfaces—bacteria and other pathogens have a difficult time penetrating solid slabs of uncut muscle. Because those surfaces are then exposed to high heat during cooking, you have less to worry about from the internal sections of the meat, even if they're not cooked hot enough to kill all pathogens. That doesn't mean they're sterile or pose no risk, but it's lower than, say, a patty of ground hamburger meat.

Still, leaving food out for long stretches increases the risk, especially once you go past the two-hour mark. If you want to temper meats, that's probably the most important thing to keep in mind.