On Cooking Rice

Go to the bottom of the post to see my recommended methods for cooking rice.

This week, I resolved that for the new year I would start blogging more frequently. Given that I really haven't been blogging at all recently, that shouldn't be too hard. I won't bore you with the various reasons why blogging has been so slow recently, but it seems that starting a new job and a new life in a new city has upended my old routines. One activity that I have been focusing much more effort on in my new life, though, is cooking. Spurred in part by reading Ratio by Michael Ruhlman, I've been trying to elevate my cooking from a practicality to an art and even, in some cases, a science (at the very least I'm trying to be more systematic about it).

The other day, I was thinking quite a bit about one of the first items I learned to cook many years ago: rice. I spent the summer between my sophomore and junior years of college in Australia, and most of that time I lived in a hostel. For the first time in my life I found myself really out on my own (I had lived in the dorms my first two years of college), so I realized that I finally needed to learn how to cook for myself. Hostel cooking was more about practicality than anything else, but I picked up quite a few skills that have served me well since then.

Early on, one of my friends in the hostel showed me a pretty straightforward and effective way to cook rice. She would add her rice to unheated water, then heat both together to a low boil in an uncovered saucepan. Once the excess water had boiled off (about 15 minutes), she would then cover the pan, turn the heat very low, and then let the rice "steam" for another 15 minutes or so. This was a long time ago, so I may not be remembering her method totally correctly, but I believe she used a 2:1 volume ratio of water:rice (i.e. 1 cup rice, 2 cups water; this is also the ratio given on the package of rice I currently have in my cupboard). This technique--which I'll call the "open-pan" method--served me pretty well, but it didn't always work perfectly, for reasons I'll describe shortly. Cooking rice in general isn't too tricky, but you still have to get it right; if it's undercooked it will be hard and slightly crunch, and if it's overcooked it will be mushy and soupy--in either case totally inedible.

Although my friend's open-pan method didn't always work perfectly, it had its advantages. Firstly, it didn't require accurately timing the different steps. The rice was ready to be covered when the water had boiled off, and the amount of time it sat closed on low heat didn't seem to matter too much, as long as it was at least about 10 minutes and not much more than about 30. A more common method of cooking rice is the "closed-pan" method, which involves bringing the water to a boil, then adding the rice and cooking on low heat (for about 20 minutes, according to the packaging on my rice, until all excess water has boiled away). Using this method, however, you'll be much more likely to have problems with boiling over. Also, your results will depend quite a bit on how much steam escapes during cooking, whereas your results should be more consistent using the open-pan method. Traditionally, one would wait until the water begins boiling before adding the rice, so not having to wait for a boil is another advantage of my friend's open-pan method. Of course, one could devise a variety of hybrid methods by combining different aspects from either method, but in this post I'll just compare my friend's open-pan method with a traditional closed-pan method.

As mentioned above, the open-pan method also had its problems. Specifically, on occasion I found the rice undercooked toward the end of the process, causing me to have to add additional water and extend cooking. The other day, for example, I was cooking 1/2 cup of rice using 1 cup of water. After the excess water had boiled away, the rice was still very hard and undercooked. So, I added an additional 1/2 cup of water, let that boil until the excess was gone, and then I proceeded as usual. The final product came out just right.

Since I'm trying to be much more systematic about my cooking, I decided to figure out what exactly was going on here and to come up with a better formula for determining an optimal water:rice ratio. Previously, I had been working with the formula:

Vw = 2 Vr

where Vw corresponds to the volume of water and Vr to the volume of rice. But, conceptually, such a formula shouldn't work for the open-pan method anyway, because some of the water will be absorbed by the rice, and some will be boiled away. So, I hypothesized that a better formula would be:

Vw = A Vr + B

where A and B are constants. Based on my experience up to that point, I started with the values A = 1 and B = 8 oz., which meant that the rice would absorb a volume of water roughly equal to the volume of rice, and about 1 cup of water (8 oz.) would boil away. There wasn't a specific reason for choosing these numbers, other than that they melded with what I had previously observed. The cup of excess water would basically determine the cooking time of the rice, which I assumed would absorb a constant volume of water (relative to the amount of rice present), although this assumption turned out not to be totally correct.

Being the empiricist that I am, I decided to test this formula by cooking 1 cup of rice in 2 cups of water, using the open-pan method, weighing the rice before and after cooking. If my hypothesis was correct, the rice would come out perfectly cooked, weighing 8 ounces more than before it was cooked. This, of course, would not prove that my hypothesis was correct, and many additional trials using different volumes of rice and different water:rice ratios would be needed to have reasonable confidence in my final conclusion... but a single person can only consume so much rice! This experiment (along with all others reported here) was performed on white long grain rice. Results could differ for different types of rice, different cooking surfaces, and different pans. The rice was neither rinsed nor soaked before cooking, which would also change the results.

I found that in this trial ("Trial 1"), my 1 cup of uncooked rice weighed 6.1 oz. and weighed 17.9 oz. after cooking. Thus, it absorbed 11.8 oz. of water, roughly 1 1/2 times the original volume of rice. During this process, the rice roughly tripled in mass. Most importantly, this rice came out cooked just right! But, the rice absorbed more water than I had initially predicted.

I decided to perform a second trial ("Trial 2"), this time to see just how much water rice could absorb. For this trial, I cooked 1/2 cup of rice with 3 cups of water. This rice, which weighed 3.2 oz. absorbed an additional 10 oz. of water, quadrupling in mass. Not surprisingly, this rice was mushy, soupy, overcooked, and totally inedible.

So, I learned that rice can absorb quite a bit of water, but it's just about right when it absorbs about 1 1/2 times its initial volume (or 2 times its initial mass). So, as a general rule, for the formula:

Vw = A Vr + B

A = 1.5, and 4 oz.

More precisely, one could express these amounts in terms of mass, where, for the formula:

mw = C mr + D

C = 2, and 4 oz.

But, what about the traditional closed-pan method? In this case, I assumed that the value of the second constant (B or D) would be zero, since keeping the lid on the pan should prevent water from escaping, causing all of it to be absorbed by the rice. But, as a first test ("Trial 3"), I decided to follow the directions on my package of rice (2:1 water:rice ratio, bring to a boil, then keep at low heat for 20 minutes), cooking 1/2 cup of rice with 1 cup of water. In this case, the 3.1 oz. of rice absorbed 6.5 oz. of water--1.6 times the initial volume of rice. This rice came out almost right, although it was slightly wet and overcooked--but just slightly, and it was still edible.

So, I decided to conduct one last closed-pan trial ("Trial 4"), this time with 1/2 cup of rice and 3/4 cup of water. Here, 3.1 oz. of rice absorbed 5.1 oz. of water. This rice came out better than in Trial 3, but it was just very slightly hard.

It turned out, then, that even with the pan closed, a small amount of water was still lost, although perfectly cooked rice still absorbed about 1 1/2 times its initial volume in water. Thus, for this method, given the formula:

Vw = E Vr + F

E = 1.5, and F = 1 oz.

In terms of mass, given the formula:

mw = G mr + H

G = 2, and H = 1 oz.

These findings are, of course, only fully applicable to my own kitchen setup. However, they should be still be reasonably broadly applicable. I think that the value of the first constant (A, C, E, or G; 1.5 for volume formulas, 2 for mass formulas) should always be true, at least for this type of rice (white long grain). The mass formulas will be the most broadly applicable, but the volume formulas should generally work. The value of the second constant (B, D, F, or H) could vary more depending on your specific setup, so it might require more refining. In particular, this constant will need to be increased as you move to a larger pan with greater surface area.

The results of all of these trials, plus some of the "preliminary experiments" mentioned above, are summarized in the following table:

Trial Method Vr (fl.oz.) Vw (fl.oz.) mr (oz.) mr+w (oz.) mw,abs (oz.) mw,abs:mr Vw,abs:Vr Result
-2 open 4 8           undercooked
-1 open 4 12           good
1 open 8 16 6.1 17.9 11.8 1.9 1.5 good
2 open 4 24 3.2 13.2 10.0 3.1 2.5 overcooked
3 closed 4 8 3.1 9.6 6.5 2.1 1.6 OK, slightly wet
4 closed 4 6 3.1 8.2 5.1 1.6 1.3 good, just slightly hard

Based on the experiments above, I can recommend two methods for cooking rice. The key to both of these, though, is that they don't involve just a simple water:rice ratio, as described below. In either case, these are specifically for cooking white long grain rice.

1. Open-Pan Method

Add rice and unheated water to a small saucepan, and bring to a boil. Continue at a slow boil stirring occasionally just in the beginning. Once almost all of the excess water has boiled away, and the rice mixture has a dimpled appearance and no longer moves freely, turn the heat to very low and cover the pan. Do not uncover until you're ready to serve! Keep heating for about 15 minutes, but not more than about half an hour.

To determine how much water to add to your rice, take the volume of rice and multiply it by 1.5. Then, add an additional 1/2 to 3/4 cup of water. For example, to cook 1 cup of rice, you'll add 2 to 2 1/4 cups of water (1.5 x 1 cup + 1/2 [or 3/4] cup). For larger pans, you may need to increase the amount of excess water beyond 3/4 cup.

2. Closed-Pan Method

Bring water to a boil in a small saucepan, then add rice and cover the pan. Once boiling resumes, turn to low heat and heat at a low boil for 20 minutes (or until all visible excess water has been boiled away for at least about 5 minutes).

To determine how much water to add to your rice, take the volume of rice and multiply it by 1.5. Then, add an additional 1 oz. of water. For example, to cook 1 cup (8 fl. oz.) of rice, you'll add 13 oz. of water. These values are for a small saucepan, and for larger pans you will certainly need to increase the amount of excess water you add. You may also need to increase this amount if the lid does not form a tight seal with the pan.

More like this

What makes the perfect Thanksgiving turkey gravy? Depending on your preference, it might be the smooth texture, the rich flavor, or the glossy sheen. No matter what the end result, the magic of gravy lies in the science of starch. So, what is a starch, exactly? A starch is a complex carbohydrate;…
Here are the rest of the recipes for dishes that I'm making for Thanksgiving this year (with the exception of pumpkin pie -- I'm still shopping for a pumpkin pie recipe). I'll mention here (and should have mentioned in the previous post that all the measurements here are U.S. quantities (cups,…
I love to cook. I have a blog. Today, and probably a few more times in the future, I will post about cooking on my blog. Questions, concerns, and hangups may be directed to the management. I once had a friend who didn't understand the fetishization of chocolate. He so resented its common…
I got hit by a mutant meme; I don't remember who tagged me. I'm not terribly into these meme things, but I don't pass up excuses to post recipes. So below the fold are four recipes that I've created: seared duck breast with ancho chile sauce; saffron fish stew; smoked salmon hash; and spicy…

I have been searching for the qtty of water absorbed by rice during cooking to determine the output of a kg of rice grains. I need to clarify a few points on this experiment:

1. At what temperature/s were the volumes of the cooked rice taken?

2. I am used to experimenting by weight. Having ounces as the unit is new to me. Is it possible for you to give me a weight conversion of the absorbed water?

For a cook, the 1:1.5 proportion is convenient, however, for computation of %, it's really necessary to get the weights.

THanks in advance,

Carla Valencia

I use gas-stove. Aside from the monthly rent I pay, the owner of the room-for-rent I stay in have us declare our electronic devices and adds up a fixed monthly fee per device for us to use them. So, I reckoned it'd be better to use stove to save some money.

Not sure if this is the right choice, though. It'd surely help to know of the following:

- How much gas does it take to cook rice?
- How much electricity does it take to cook rice?

Gas is cheaper here than electricity. Any thoughts? :D

Having seen some of the places where rice is grown, stored and shipped I'd recommend rinsing rice before cooking it!
Love the post and I'm looking forward to reading the scientific side of cooking risotto...

I like to cook rice in the microwave in an oven-proof glass container with a lid. It works quite well, and there is no risk of burning the rice. Some microwaves even have a dedicated "rice button."

I wonder if this is more or less energy efficient than the stove-top method?

I use a rice cooker, just like about 1 billion other people on this planet that regularly eat rice. It comes with instructions, they work.

Nice article - one that I'm sure will be the final word on the matter. Probably not, however, until your math error in the final paragraph is fixed: (1.5 x 8 oz.) + 1 oz. is 13 oz., not 9 as you said.

I second the recommendation to get a rice cooker. Most rice cookers take advantage of the upper limit on the temperature of the water when it's being boiled off - when the temperature of the heating element rises above 100C, the liquid water has all boiled off.

Even with a rice cooker, you need to have an appropriate ration of water to rice. However, in addition to volume, I think you also have to consider surface area, which I think affects how much of the water vaporizes while bringing it to the boil. Thus, I think you'd use a different ratio making rice in your favorite saucepan (or rice cooker) vs making a paella in a wide vessel.

Rice Cookers all fine and dandy if you have the space and money, but for a student, knowing these ratios is wonderful!

I use a rice cooker it is the best, didn't quite believe it though until I used it myself. Any other way of cooking rice it just doesn't turn out right.

Forgot to add on last comment rinsing rice is essential though I think I do go a bit OTT on the rinsing and it ends up pappy...

Great Heavens, Horatio! There are more things in life than worrying about cooking rice. Just rinse, bang a batch into some boiling water and, two minutes before the boiling time stated on the packet take a little out and test it between your front teeth. Do this a couple of times until ready, i.e. just softer than al dente. Drain, rinse, fluff or whatever you fancy. Job done. Now let's talk about the agonies of getting squid right. What in physics makes it SUDDENLY turn rubbery and useless?

The first thing I bought when I first left home was a small rice cooker. So there was me, broke, studying, in a tiny bachelor with a murphy bed (which I somehow broke on the first day). My computer didn't even have a desk yet but I had a rice cooker!

I was raised on rice. The thought of a rice-less house horrifies my family. They consider it bad luck... (for the record, I come from a family of rice farmers).

I just have to point out that the overcooked/mushy can be made into congee; add water to get the consistency you like, and then season each bowl individually - add soy sauce, scallions, peanuts, chilis, tofu, cilantro to taste and it is lovely.

On the other hand, the crunchy undercooked...can't really do much there....

By Monisha Pasupathi (not verified) on 05 Jan 2010 #permalink

Given that water boils at a lower temperature as elevation increases, I have to ask: at what elevation did you do all of these experiments? That definitely affects the applicability of the results.

The amount of water needed actually depends on how dry the rice is. In this modern age the industrial drying machinery tends to get a fairly consistent water content; short shelf storage and packaging in airtight plastic bags also helps maintain a fairly consistent water content. Rice dried in the traditional way (under the sun) would have highly variable amounts of water between batches and this also depends on how long the grain is stored. Many recipes call for rice prepared a certain way - for example, dried and stored for ~3 years - such grains would have very low water content and require much more water for cooking. There are also many varieties of rice and they can't all be cooked to any single recipe. For example, were you using a short, medium, or long grained variety?

The chief effect of soaking is to reduce the cooking time; if you're going to cook rice in the evening and you're really keen on saving a few watts of power, in the morning you should soak the rice in the amount of water necessary for cooking and cover it well so than no water is lost to evaporation. Soaking ~2h before cooking as I've seen suggested in many places is of no practical value (unless you soak in hot water - and what's the point of that).

By MadScientist (not verified) on 05 Jan 2010 #permalink

Just 1 comment on the crunchy undercooked stuff - this often happens even if you do use the correct amount of water; the usual problem is that the heat should have been turned down much earlier, or the whole thing was heated up far too quickly to begin with. Water is rapidly lost as steam without much of a chance for the hot water to soak into the rice and change its texture. If you see this happening you can compensate by pouring a little more boiling water into the pot and reducing the heat. Of course it's better to get it right rather than have a need to fix it up.

By MadScientist (not verified) on 05 Jan 2010 #permalink

"Rice Cookers all fine and dandy if you have the space and money, but for a student, knowing these ratios is wonderful!"

There are perfectly adequate rice cookers that you can get for $10-15 (I bought one in that price range at Bed Bath and Beyond a couple of years ago when I was on sabbatical), so it's hard for me to buy the money argument. Plus, when I was a poor student, having a rice cooker freed up important burner space on the tiny stovetops that were found in the cheap apartments I could affort.

no way to measure the rice and water?..try this.add rice to the first joint of your finger,your-knuckle, and water to the third.salted water is best.this is an old canoe trippers trick.

Different rice varieties require different cooking methods and times. Arborio is best cooked into risotto, the liquid being added in small amounts as you go. Basmati can be soaked for an hour ahead of time and cooked for but a few minutes, covered, and then let to puff up, soak in the water in residual heat. I expect your experiments were all done using something like long-grain rice?

And regarding squid, either cook it forever or barely cook it. If you're adding it to a tomato sauce (seafood pasta), get the sauce to near boiling, put in the squid and turn off the heat! Residual heat will cook it, it's like poaching eggs. Or a super-quick deep fry. Or stew it forever. Go for the ends of the Bell curve, cook times in the middle will create shoe leather.

By zephyr haversack (not verified) on 05 Jan 2010 #permalink

Now that the concept of "void volume" has been invoked, I fear there's no turning back! It would be interesting to see, though, what sort of water:rice ratio you generally end up with, Jim, by eyeballing the water level in your rice cooker.

In the time that it took you to write this post, you could have gone to the store, bought the $15 rice cooker, found a place to put it in the tiny apartment and made rice twice. As a post-doc, your time is money and this just seems like a waste. Our kitchen is 6x6 and we have a crock pot and a rice cooker.

By A postdoc myself (not verified) on 06 Jan 2010 #permalink

I use an electric rice cooker. Always consistent. 3 times a day. Never have to stir, touch lid or keep watch. Just place 1 cup rice with 2 cups water, then go watch TV. It high-heats, then low heats then keeps warm then shuts off all by itself.

I once challenged a group of Pygmies in the Ituri Forest to cook rice without using a pot ... using only leaves. (They had claimed they could do anything with leaves)

Well, it was also OK to use fire. Fire and leaves and water and rice.

The did it easily. They also cooked some duiker stew and some beans using only the leaves and fire, for good measure.

I'll throw some more support to the rice-cooker faction. I have a little 6-cup one that I paid $12 for. The only downside I've found is that it's something of a pain to clean the bowl. If you don't need to make more rice right away, a couple of hours soaking will generally remove all of the cooked-on rice with a quick pass of the sponge.

I tend to cook jasmine rice; it's not quite as expensive but it has a very similar smell. (All rice tastes about the same to me, but the smells are very important and very distinctive.)

For white rice, put rice in pot and level off to even depth. Note depth of rice. Add water so that it comes to double the level of rice (but not, of course, double the volume). Bring to boil. Turn down to lowest heat. Give it a good stir and put the lid on. Wait 20 minutes.

Voila! Light fluffy rice.

Been doing it this way for decades. Works every time.

Tip: Use a thick bottom pot to spread the heat out more evenly, and I only use stainless or glass for cooking rice.

Some people turn the heat off at about 17 minutes, and leave the pot sit (with lid on) to let the rice cook from its own heat for another 2-3 minutes.

Works for brown rice too, except the water level is about 2.5 times and the cooking time is 40 minutes. I can't remember exactly, it's years since I did brown rice.

Another question on the method -- was this tested on a gas or electric stove? If electric, how did you compensate for the time it takes for the burner to cool down when reducing the heat?

Morning all

I graduated many years ago and am now in a position to be able to afford even better tools to cook rice. Currently it's in the form op an induction sealed japanese rice cooker, a mid sized model. It's got a coated steel vessel (pot) that sits above a copper coil. The lid closes down and seals the pot. A microprocessor controls the cooking cycles and there's a valved steam vent that is forced open during cooking cycle and remains closed during the keep-warm cycle. It's not a pressure cooking type, though those are also available.

Alot of tech for what it actually does, but it sure does it well. No more crunchy stuff on the bottom and the rice keeps warm in a palatable state for about 8 hours. It should also use less power but i haven't tested.

Nice thing about this cooker is its consistency. It's partially sealed, it's insulated, and the cooking is controlled by a microprocessor.

The mystery remains, how much water to rice? As I see it, here are only four variables:

mass of water
mass of rice
altitude, i'm at 950m (~3000')
type / quality of rice

The manual says that you should use the rice cup for the rice and the scale etched onto the side wall of pot for the water.

The cup is a "rice cup" 180ml, not a standard ~250ml cooking cup. The scale is intended to be used with the rice already in the vessel. Using this method for long grain and medium grained rice works every time. I've never had an undercooked or overcooked batch.

But why? Certainly because the ratios are correctly worked out by the manufacturer. Also my measurements are consistent, i.e., I take care when levelling rice in the cup and water scale and lastly because the cooker itself is consistent.

Using a weighing scale, 300g of rice went into the pot. Zeroed the scale and filled to the correct line in the pot, with 500g of water.

Rice to water is 1:1.66. Similar to your results.

This is at altitude, next month i'll be at sea level. Maybe there'll be a difference.

Great post, I'm also very interested in the science of cooking - I'm a fanatic of brown rice (love the flavor and nutritional value), and I'm trying to figure out how to make it best. I've made some great risottos and I'm now working on how to use the brown rice for totally open style entrees like paella.

Do you recommend the Ratio book?

I would recommend Ratio, but I would add a couple of qualifiers to this recommendation. The book offers only a limited number of recipes, so if you already have an extensive cooking repertoire, you may not find it as helpful. However, for someone (me) with more limited cooking experience--and someone with a scientific perspective on things in general--this book was perfect. Ratio is more of a teach-a-man-to-fish sort of thing, in that after reading it you should understand the fundamentals of various aspects of cooking (especially doughs and sauces)... allowing you to more effectively improvise on your own (and more effectively adopt other recipes to your needs and liking).

Dear Nick,

Rambling, rambling. Like my ole daddy used to say, "even a train stops sometime".
yours truly,

I've recently been reading about cooking with a thermos bottle (a relatively wide mouth stainless steel model is usually recommended).

Preheat the bottle with hot water, remove water to reheat, put rice (or beans or other things) into thermos, put the reheated water back in. Let it sit for a while (the reported times vary a lot).

Advantages are very low energy use (think propane stove during power outage) and no requirement to monitor cooking.

Anyone have experience with this method? I'm about to do some experiments.

By the way, I've been using a cheap rice cooker for several years. Great appliance, essentially 100% success rate, never burns, and keeps the rice well for about an hour or so after cooking.