A Banana Before Its Time: The Case for Green Bananas

For most of my life, a green banana was not considered food. It was a banana awaiting further instructions. You set it on the kitchen counter until it turned yellow, then ate it during the brief interval before it developed brown spots and became the raw material for banana bread.

Science has now arrived to inform us that we had the matter backward.

The yellow banana may be sweeter and easier to chew, but the green banana contains considerably more resistant starch, a carbohydrate that behaves less like ordinary starch and more like fiber. Because its glucose molecules are packed together with unusual stubbornness, our digestive enzymes cannot easily break them apart. The starch passes through the small intestine largely intact, avoiding the customary rush of glucose into the bloodstream.

It then reaches the colon, where several trillion bacteria are waiting for lunch.

These microbes ferment the starch and turn it into short-chain fatty acids and other chemical messengers that appear to affect inflammation, insulin sensitivity, fat metabolism and possibly even appetite. It is a remarkably elaborate chain of events for something most of us would ordinarily throw at the grocer.

The most impressive evidence comes from a randomized crossover study published in Nature Metabolism. Participants with excess weight consumed 40 grams a day of resistant starch for eight weeks. They lost an average of about six pounds, reduced fat mass and improved insulin resistance, even though researchers kept their calorie intake and general diet comparable. MRI scans also showed reductions in visceral fat—the particularly disagreeable kind that gathers around the internal organs and behaves as though it owns the property. The study also linked the benefits to changes in gut bacteria.

Other controlled research has found reductions in abdominal visceral and subcutaneous fat after 40 grams a day of resistant starch, although earlier studies in people with diabetes or metabolic syndrome have produced less consistent results. Science, like the average family, rarely reaches unanimity at the dinner table. One double-blind crossover trial found significant abdominal-fat reductions after four weeks.

Researchers strengthened the microbiome case by transferring fecal microbes from human participants into germ-free mice. The mice receiving microbes associated with the resistant-starch diet acquired some of the same metabolic benefits despite eating otherwise identical food. This was excellent news for microbiome science and a difficult day for the laboratory employee asked to explain his occupation at a cocktail party.

Not everyone responds equally. Some people possess more of the bacteria capable of turning resistant starch into useful metabolites. Others harbor competing microbes that regard the whole operation with bureaucratic suspicion. Recent research has associated stronger responses with particular Bifidobacterium species, while higher levels of Prevotella may help explain weaker results. A 2025 randomized trial traced substantial differences in liver-fat response to participants’ starting microbiomes.

Then there is sleep.

A small randomized trial using a blend of potato starch, green banana flour and apple fiber found that higher cumulative consumption was associated with less self-reported sleep disturbance after four weeks. This is intriguing, but it does not prove that a green banana at bedtime will produce eight uninterrupted hours and pleasant dreams about Ava Gardner. The finding was exploratory, based on questionnaires, and involved a blended supplement rather than bananas alone. The authors sensibly called for larger studies using objective sleep monitors.

The therapeutic dose used in the major metabolic trials was 40 grams of resistant starch daily. Obtaining that from whole green bananas could require roughly six bananas, depending on their size and ripeness. Six raw green bananas a day is less a diet than a hostage negotiation.

More practical sources include high-amylose maize starch, green banana flour and raw, unmodified potato starch stirred into cold yogurt or a shake. Heat destroys much of the crystalline structure of Type 2 resistant starch, so it should not be baked into cookies and declared medicine. Start with a small amount and increase gradually. Leaping directly to 40 grams may cause bloating, gas and a domestic situation in which improved sleep becomes largely academic.

I have therefore developed what may be the Holy Grail of modern nutrition: a dessert that might actually help rather than hinder the campaign against visceral fat.

The secret is green banana flour, a resistant starch that makes it through the upper digestive tract largely unmolested and feeds the beneficial bacteria farther south. In theory, everybody wins except the visceral fat.

You may want to give it a try. At the very least, it is one health experiment that tastes like chocolate.

No-Bake Resistant-Starch Brownie Bites

Makes about 16 small bites

• ¾ cup green banana flour
* ⅓ cup pecans, finely ground
* ¼ cup CocoaVia unsweetened cocoa powder
* 3 tablespoons dark chocolate chips, preferably 85% cacao or higher, or very low in sugar
* 3 tablespoons allulose, adjusted to taste
* Approximately ⅓ cup unsweetened macadamia milk
* 1 teaspoon vanilla extract
* ¼ teaspoon salt

Finely grind the pecans. In a bowl, combine the green banana flour, cocoa powder, allulose and salt. Stir in the ground pecans and chocolate chips.

Add the vanilla, followed by the macadamia milk a little at a time, stirring until the mixture is just moist enough to hold together. Do not heat it; the whole point is to preserve the resistant starch, which has already gone to considerable trouble to remain resistant.

Divide the mixture into 16 roughly one-inch balls and refrigerate. They will firm up considerably after several hours.

Begin with one. Your beneficial bacteria may be hungry, but there is no reason to seat the entire party at once.