Moong Dal Khichdi with Cucumber Raita

What it is

Yellow moong dal and white basmati rice, rinsed together and simmered in one pot until the dal melts and the rice goes soft. That's khichdi. It starts with a tadka: cumin seeds and a bay leaf sizzled in ghee, then ginger, green chilli, tomato and turmeric. Each bowl gets a teaspoon of ghee and some coriander on top, with a cold cucumber raita on the side.

Champion number one put rice and beans side by side on a Brazilian lunch plate. Number six puts them in one pot. Moong dal is the mung bean split, with its green hull taken off, and it goes in here with no overnight soak. Ghee and yogurt are its only animal foods, so it's lacto-vegetarian from first spoon to last, and it's the bowl I'd put in front of you after a long training week.

Khichdi has even been to the lab. When researchers in Bangalore needed a mung bean meal that toddlers would actually eat, khichdi is what they cooked. You'll meet them, and the hull, while the pot simmers.

Why you should eat it

The short version: soft and yellow, and still a protein champion. USDA FoodData Central lists 23.9 g of protein and 16.3 g of fiber per 100 g of raw mung beans. Each bowl starts from 62.5 g of dry dal, so the dal alone brings about 15 g of protein (14.9 g by that record), with its fiber riding along in the same spoon. One honest catch: that record is for whole beans, and taking the hull off takes some fiber with it, so I'd read the card's fiber as a little generous. On fat: USDA's ghee record is 100 g of fat per 100 g, 60 g of it saturated, which is why the ghee goes on by the teaspoon. The rice and the raita add protein of their own, and how much of all that protein your body actually collects is the twist I'll tell you while it simmers.

Tips

• Stir and scrape every 5 minutes: thickening khichdi likes to catch on the bottom, and a scorched base flavours the whole pot.
• It keeps thickening off the heat. Loosen it with hot water before serving, and again when you reheat.
• Squeeze the cucumber hard, or the raita turns watery before you sit down.
• More heat? Leave the slit chilli in the bowls. Less? Fish it out with the bay leaf.
• For the training week, freeze the leftovers in portions within the hour (see Storage), and a warm lunch is one reheat away.

Substitutions

• Yellow moong dal: Whole green mung beans, soaked overnight More fiber and a longer cook, so simmer until soft. In case one, whole beans gave up less of their protein than hulled ones (63.2% against 70.9%), and the card, built on the whole-bean record, matches them better.
• Heavy pot on the hob: A pressure cooker Faster. Follow your cooker's manual for the water, the timing and the release; no source I opened gives a khichdi pressure time, so I won't guess one.
• Plain whole-milk yogurt: Whole-milk Greek yogurt, gelatin-free (check the label) A thicker raita with more protein: USDA FoodData Central lists 9 g of protein per 100 g of whole-milk Greek yogurt (record 171304) against 3.47 g for plain whole-milk yogurt (171284). The card is built on the plain one, so it won't match.
• Ghee, for the tadka: A neutral vegetable oil, keeping the teaspoon of ghee on top if you like The tadka loses ghee's buttery note, and the card, built on ghee, won't match; check your oil's label.

Storage

Khichdi is a rice dish, so it follows the UK Food Standards Agency's rice rules: chill leftovers as quickly as you can, ideally within an hour, keep them in the fridge no more than a day, and reheat once only, until steaming hot all the way through. To freeze, portion it into small containers so it cools fast and goes into the freezer within an hour of cooking; defrost it in the fridge, reheat until steaming hot and eat it straight away. Loosen it with hot water as it warms. The raita is at its best the day you make it.

While it simmers: the present and the wrapping

Lid ajar, a stir every five minutes, and nothing else to do. Pull up a chair.

Every protein number so far has been a label number: what sits in the food. Your body plays a different game. Protein gets taken apart into amino acids in your small intestine, and that's where they're absorbed. What counts is the share collected there, and scientists call that share digestibility.

Case one measured it for mung beans in people. Whole mung beans gave up 63.2% of their essential amino acids. The same beans with the green hull taken off gave up 70.9%. The researchers point at the hull, which carries compounds that get in digestion's way. Same bean, a bigger share collected.

With my early-childhood education hat on, here's the picture: a present wrapped in far too much tape. You get most of it open, and some bits stay stuck in the wrapping. Hulling peels off one layer before the present ever reaches you, so more of it comes out.

The picture stops in two places. First, the wrapping wasn't only in the way: hulling took fiber with it too, from 17.54 g down to 13.37 g per 100 g of cooked bean in their lab. Second, not every gift came out easier. Lysine, the mung bean's strong suit, was absorbed at about 63% with or without the hull.

Now the lunchbox swap from champion number one. In USDA FoodData Central's records, mung bean protein carries 6.95 g of lysine per 100 g of protein, about the same share as black beans, but only 2.08 g of the sulfur pair, methionine and cysteine (black beans: 2.6). Rice protein runs the other way: 4.40 g of the sulfur pair, 3.62 g of lysine. Yogurt protein beats the mung bean on both: 8.96 g of lysine and 3.86 g of the sulfur pair.

Put this pot together and the sulfur pair's share of its protein climbs from 2.08 g with the dal alone to 2.38 g with the rice, then to 2.63 g with the raita, while lysine ends almost where the dal started, at 6.93 g. Label numbers again, before digestion has its say.

So, in one line: hulling lets you collect more of what the mung bean carries, and the rice and the raita top up the sulfur pair it runs short on.

Case one: the hull, measured in people

Six healthy young adults in Bangalore, India, three women and three men with an average age of 23.5, ate mung beans on two separate days: once whole, once hulled by hand. The beans had been grown on water carrying a heavy form of hydrogen, a tracer that let the team follow their amino acids into the blood. The meal sat on a local everyday base of ghee rice and tomato curry.

Whole mung beans gave up 63.2% of their essential amino acids by the end of the small intestine, the lowest of the three pulses tested (chickpea 74.6%, yellow pea 71.6%). Hulled, they reached 70.9%. That overall rise passed the statistics test, and so did the rises for threonine (42.5% to 54.5%), leucine and valine. Methionine, the scarce sulfur one, went from 52.2% to 64.3% but didn't pass on its own. Lysine barely moved: 63.0% to 63.4%.

For your plate: in these six people, hulled beans gave up more of their protein. Yellow moong dal is the hulled kind.

Where it stops: six people is a small group, and the authors say their method is new and still needs checking against direct sampling at the end of the small intestine. One thing I noticed: the hulled beans were also frozen before cooking and pressure cooked for 12 minutes instead of 15, so the hull wasn't the only change. The Wellcome Trust/DBT India Alliance and the International Atomic Energy Agency paid for it, and none of the authors declare competing financial interests.

Case two: khichdi, tested on toddlers

Eighteen children aged 18 to 24 months in Bangalore, none of them still breastfed. The team first checked which meals children like them would actually eat, and the mung bean meal they settled on was khichdi: rice and whole mung beans, 2.5 to 1, cooked with tomato, onion and garlic. Four boys ate it, and other children were tested on rice, finger millet or egg, all with case one's tracer method.

Mung bean came last: 65.2% of its essential amino acids absorbed, against 68.4% for finger millet, 78.5% for rice and 87.4% for egg. Methionine was mung's lowest, at 54.0%.

Then the team modelled national survey diets of Indian children aged 1 to 3. The pooled national diet scored 80 on DIAAS, a protein-quality score that counts digestion, where 100 or more meets a child's reference needs. Adding 200 g of milk a day lifted it to 101 and cut the share of children at risk of too little quality protein from 29% to 8%. Adding 45 g of legumes left it at 81, which the authors put down to legumes' poor methionine and lysine digestibility. Their wish list: studies on hulling or sprouting.

For your plate: in that arithmetic, milk was the teammate that moved the score. The raita is this plate's milk.

Where it stops: four toddler boys, some of the children anaemic or undernourished, and a milk figure modelled on state survey averages rather than fed to anyone. You're probably not 22 months old. The Wellcome Trust/DBT India Alliance and the Bill and Melinda Gates Foundation funded it, and the authors declared no competing financial interests.

Case three: mung bean dressed up as an egg

Thirty-seven vegetarian and vegan adults in Phoenix, Arizona, aged 18 to 55, 34 of them women, none exercising more than 150 minutes a week. For eight weeks, about half ate a mung bean protein product shaped as an egg replacement, JUST Egg, giving 18 g of protein a day; the rest ate a breakfast biscuit with 4 g. Nobody started training. Twenty-five people finished.

Lean mass barely moved (up 0.2 kg against 0.0 kg), and no single strength test differed. Only when three strength tests were averaged as percent changes did a gap appear: up 2.9% on mung bean against down 2.6% on biscuits, p = 0.05, right on the line. The paper's title says mung bean protein improves muscular strength. The data are shyer than the title. The authors estimate the 18 g carried about 1.35 g of leucine a day, which may have been too little to fully switch on muscle building.

For your plate: by my arithmetic, a bowl of this khichdi carries about 15 g of protein from the dal alone, close to that daily dose. Whether it moves your strength, this trial can't say.

Where it stops: a pilot that finished with fewer people than its own power calculation asked for (25 against 34), no blinding, and diet measured by 24-hour recall. An isolate isn't moong dal, and nobody lifted, so it says nothing about leg day. Arizona State University funds paid for it, JUST donated the product, and the authors declare no conflict of interest.

Case four: rice and dal at the glucose test

Ten healthy adults in Delhi, seven men and three women with an average age of 30.9. After an overnight fast, each ate a North Indian meal carrying 50 g of available carbohydrate: green gram dal (green gram is another name for mung bean) cooked with garlic and ginger, an okra and onion dish, and either rice or two wheat chapattis. On other mornings they drank glucose, and blood sugar was checked by finger-prick for two hours.

Glycemic index compares a meal's two-hour blood sugar curve with glucose's, which is set at 100. The rice meal scored 83.6 and the chapatti meal 85.5, with no significant difference (P = 0.71). Average peak blood sugar was 133.3 mg/dL after the rice meal and 168.5 mg/dL after glucose. The authors' point is about rice and wheat: in a mixed meal, swapping one for the other changed little.

For your plate: rice and dal together still reached about 84% of glucose's curve in these ten people. If you've seen khichdi called gentle on blood sugar, this is the fine print. This pot carries almost three times their dal per bowl, so this isn't the meal they tested.

Where it stops: ten people and one kind of meal, the authors' own limits. The meals came in a fixed order with no randomization described, and the two meals were compared with a test meant for separate groups although the same people ate both; that last point is my own observation. It had no outside funding, and the authors declare no conflicts of interest. If you manage your blood sugar, take this to your doctor.