A Hamburger Today

The Food Lab: How to Cook a Perfect Prime Rib

It's time for another round of The Food Lab. Got a suggestion for an upcoming topic? Email Kenji here, and he'll do his best to answer your queries in a future post. Become a fan of The Food Lab on Facebook or follow it on Twitter for play-by-plays on future kitchen tests and recipe experiments.

20091218-rib-roast-opener.jpg

[Photographs: J. Kenji Lopez-Alt]

This is a four-pound joint of well-marbled prime beef rib—it is not cheap. And while my friends provide me with as many mental and philosophical riches as a man could ask for, and my wife supplies an adequate amount of emotional wealth, as a humble food writer, dollars and cents are not something I part with lightly.

20091218-rib-roast01.jpg

As such, when I buy a quality piece of beef—and honestly, does beef get any better than prime rib?—I have a great impetus not to mess it up, as do, I imagine, most of you.

So this week, at the Food Lab, I've decided to get through a lifetime's worth of messings-up so that in the future I (and hopefully you too!) will never again serve anything but a perfectly cooked roast.

So what exactly is a Perfect Prime Rib anyway? Whether you buy prime or select, fresh or dry-aged, corn-stuffed or grass-fed, if you don't cook it right, it ain't going to be good. Here is my definition or perfection, in three commandments:

Highs and Lows

Before I tried to start figuring out how to achieve all these goals simultaneousl,y it was helpful to note that when cooking beef to medium-rare, there are really only two temperatures that matter.

Ah—a dilemma revealed itself: In order to maximize browning I had to cook the meat in a sufficiently hot oven—I tried 400°F. At the same time, I didn't want the interior to come up above 125°F.

Since a big beef roast cooks from the outside in, by the time the center had reached 125°F (that is, 120°F in the oven, followed by a 5°F rise in temperature after resting), sure there was a perfectly browned exterior, but the outermost layers had risen closer to around 165°F or 180°F, rendering them overcooked, gray, and dry, their juices having been squeezed out.

I was left with something that looked like this:

20091218-rib-roast-gray edges copy.jpg

I know, I know—not pretty.

Score:

Okay, so what if I went to the opposite extreme, cooking the steak at a much lower temperature?

I cooked another roast in a 200°F oven until the center reached 125°F. Well, just like with boiled eggs, the temperature at which you cook is directly related to the difference in temperatures between the center, and the exterior layers.

In other words, by cooking it at a lower temperature, you make sure to minimize the volume of beef that comes above the ideal final temperature. I was able to almost completely eliminate the gray band of overcooked meat.

Of course, any browning that I was getting was also right out the window, leaving me with a roast with pale, flaccid exterior that looked like this:

20091218-rib-roast-pre-final-in-pan.jpg

I know, I know —again, not very pretty.

Score:

The Myth of the Sear

Jump back a couple of decades and the solution to my dilemma would have been obvious. It was a commonly held belief (and still is, by many home cooks and professional chefs alike), that in order to help a roast, steak, or chop retain moisture, your goal should be to first sear it, creating a crust that will "lock in the juices." Now anyone who reads their Harold McGee or has ever seen juices squeeze up through the seared side of a steak after you flip it over on the grill know that this can't possibly be completely true. But what about partially true?

Could a sear actually help retain at least some of the juices?

To test this, I cooked two roasts cut from the same rib sections, with comparable surface areas, weights, and fat contents according to the following processes:

If searing does in fact "lock in juices," then we would expect that the steak which was first seared then roasted should retain more juices that the steak that was first roasted then seared. Unfortunately for old wives' tales, the exact opposite is the case. I carefully weighed each roast at each step of the process to gauge the amount of moisture and fat lost during cooking. These are the results:

20091218-roasting-meat-graph.png

The meat that was seared first them roasted lost 1.68% more juices than the one that was roasted first then seared. It's not a particularly huge difference, but the knowledge that searing conclusively does not lock in juices was certainly liberating in the ways that it allowed my to think about the recipe.

Score:

Inside and Out

20091218-rib-roast-crust.jpg

So great, you may be thinking—you can sear first or you can sear after, and it makes no difference. What's the big deal?

20091218-rib-roast-seared-second.jpgWell the big deal, as some of the more astute readers may have noticed in the timing above, is that if you are starting with a completely raw roast, in order to get a well-browned crust, it takes around 15 minutes in the hot pan, during which time, the meat under the surface on the outer layers of the roast is busy heating up and overcooking, just like they did when roasted in a 400°F oven.

On the other hand, in order to get a well-browned crust after the prime rib has roasted, you need only around eight minutes in the pan. Why is this?

It all has to do with water.

In order for the surface of a roast to reach temperatures above the boiling point of water (212°F), it must first become completely desiccated. When searing raw meat, about half the time it spends in the skillet is spent just getting rid of excess moisture before browning can even begin to occur. You know that vigorous sizzling sound when a steak hits a pan? That's the sound of moisture evaporating and bubbling out from underneath the meat.

On the other hand, a prime rib that has first been roasted has had several hours in a hot oven, during which time the exterior has completely dried out, making searing much more efficient, and thus giving all but the very exterior of the meat less of a chance of overcooking.

Taking what I learned from both the oven temperature testing and the searing testing into account, I now knew what I had to do to fulfill all three commandments: My goals should be to cook the interior of the roast as slowly as possible (i.e., at as low a temperature as my oven could maintain), then sear it as quickly as possible (i.e., at as high a heat as possible). Searing in a pan is not that practical for a joint bigger than a couple of ribs wide, so I needed a way to do this all in the oven.

While some recipes will have you simply pump up the oven temperature towards the end of cooking, this is sub-optimal. An oven can take 20 or 30 minutes to go from its lowest temperature to its highest temperature setting, during which time, once again, the outer layers of beef are busy overcooking.

But then, I thought, 20 to 30 minutes is exactly how long a rib roast needs to rest anyhow. What if I were to first cook it at a low temperature (200°F or lower), take it out of the oven, allow it to rest while I heated the oven to its highest temperature (500 to 550°F), then pop it back in just long enough to achieve a crust?

What I achieved was nothing less than Prime Rib Perfection:

20091218-rib-roast-perfect.jpg

Score:

As you can see, no gray overcooked meat, a crisp brown crust, and a rosy pink from center to edge.

20091218-rib-roast-no-gray.jpg

But wait—there's more!

The best part? I found that by cooking with this two stage method, I had a much larger window of time to serve the beef. Once I got past the initial low-temperature phase of cooking, so long as I kept the roast covered in foil, it would stay warm for over an hour. All I had to do was pop it back into its 550°F oven 8 minutes before my guests were ready to eat, and the roast would emerge hot, sizzling, and ready to carve, no need to rest it after the 500, since the only part that is being affected is the very exterior.

Family gatherings will never be the same. Now if only I could find a way to expose the rosy center under my sister's crusty exterior, we'd really have something to celebrate at the holidays!

Printed from http://www.seriouseats.com/2009/12/the-food-lab-how-to-cook-roast-a-perfect-prime-rib.html

© Serious Eats