Caffeine in Matcha vs Coffee: The Numbers, Honestly Compared

Disclosure: This article contains affiliate links, including Amazon links. As an Amazon Associate I earn from qualifying purchases. It costs you nothing extra, and I only link to things I have used behind my own counter.

I run a small cafe in Kyoto called Fudemame. We open one day a week, Sundays from noon to 4:30pm, in a borrowed corner of a riverside cafe near Demachiyanagi. I roast our coffee myself on a sample roaster, in batches small enough that I know each one by sight.

I also serve tea. Which means that most Sundays, I stand at the same counter and make both: a hand-drip coffee for one person, a bowl of whisked matcha for the next.

So when customers ask me which one has more caffeine, and they ask constantly, usually in that slightly guilty voice people use around 3pm, I wanted a better answer than the one I kept finding online. I went and read the research on caffeine in matcha vs coffee properly, primary sources and all.

What I found is that the honest answer is more interesting than the marketing answer. Three things that almost every matcha-versus-coffee article tells you are either unsupported by evidence or contradicted by the studies those articles gesture at.

Let me show you.

The short answer

At the servings most people actually pour, drip coffee edges out matcha. But the ranges overlap heavily, and within each drink the spread is wider than the gap between them.

Two things jump out of that table, and neither of them is “matcha has more caffeine than coffee.”

The first is that a real espresso pulled in a real cafe ranged from 51 mg to 322 mg across twenty shops in a single city (Crozier et al., Food & Function, 2012). That is a six-fold spread inside one drink category. Any comparison that treats “espresso” as a fixed number is comparing a number to a fog.

The second is that matcha’s range is just as unstable, for reasons almost nobody explains. That is where we should start.

How much caffeine is in matcha?

Matcha whisked at the usual strength lands somewhere around 52–92 mg per bowl. The powder itself measures roughly 26–46 mg of caffeine per gram in the two studies that separated grades.

Those figures come from two independent peer-reviewed analyses. Both report mean ± standard error:

• Ceremonial grade 45.71 ± 3.01 mg/g, culinary grade 31.39 ± 2.27 mg/g (Journal of Food Measurement and Characterization, 2025)
• Ceremonial grade 31.45 ± 1.08 mg/g, culinary grade 26.29 ± 1.86 mg/g (Plant Foods for Human Nutrition, 2023)

A 2021 review of the literature to that point put the published range at 18.9–44.4 mg/g (Molecules, 2021), so the 2025 ceremonial figure sits a little above anything reported before it.

And one analysis of three European commercial matchas came in far lower than any of them, at 14.1–16.1 mg/g (Antioxidants, 2022). That is about a third of the top figure. I am using 26–46 mg/g below because it comes from the two studies that graded their samples, but you should know the floor may be lower than that if you bought a cheap supermarket tin.

Why the numbers swing so much

Notice that the two graded studies disagree about ceremonial matcha by a factor of 1.5, and that the widest published values differ by more than three times. That is not sloppy science. It is what matcha is.

Matcha is a ground leaf, not a standardized extract. The caffeine in your bowl depends on cultivar, how long the plants were shaded before harvest, whether it was first or second flush, and what grade you bought. Across both graded studies, ceremonial samples measured higher in caffeine than culinary ones, which is the opposite of what most people assume when they pay more for the good stuff to be gentler.

Then there is the variable that swings the number harder than any of them: how much powder you put in the bowl.

Working out your own number

Nobody has published an official standard serving for matcha. The 2 g figure everyone quotes, including me two paragraphs ago, is a Japanese tea-room convention for usucha, thin tea. It is not a regulatory or scientific standard. I went looking for one and it does not exist.

So the useful move is to stop looking up averages and measure your own:

At the counter I weigh matcha the same way I weigh coffee, on a 0.1 g scale. It takes two seconds and it is the only way to know what you are drinking. Most people who tell me matcha makes them jittery turn out to be using a heaped tablespoon, which is 156–276 mg. That is a double espresso, not a cup of tea.

People ask why I do not just use the chashaku, the bamboo scoop. I do use one — it is the nicest way to move powder from tin to sieve. But it is a serving tool, not a measuring one. Two scoops from mine can differ by close to half a gram, which at the top of the range above is the difference between a cup of tea and most of an espresso. The scoop is for the ritual. The scale is for the answer.

The one thing to buy

A 0.1 g pocket scale

If matcha is all you make, a basic pocket scale does the entire job. I used one for a year before I upgraded, and it told me everything this article took four thousand words to say.

Check on Amazon →

If you brew coffee too

Timemore Black Mirror Basic 2

This is the one on my counter. It weighs the coffee and it weighs the matcha, which is the whole point of this article in one object. The built-in timer is for pour-over, not for tea, but you stop noticing that.

Check on Amazon →

How much caffeine is in coffee?

About 95 mg in an 8 fl oz (237 ml) cup of drip coffee.

The USDA’s laboratory analysis lists brewed coffee at 40 mg per 100 g, which works out to roughly 95 mg per 8 oz cup (USDA FoodData Central, 171890). The Mayo Clinic independently lists 96 mg for the same serving.

But the FDA gives a range of 113–247 mg for a 12 fl oz cup (FDA), which is about 75–165 mg once you scale it back to 8 oz. Same drink, roughly a two-fold spread.

I can tell you where that spread comes from, because I create it every Sunday. Dose, grind, water temperature and contact time each move extraction. When I brew at a 1:15 ratio instead of 1:17, the same beans produce a noticeably stronger cup. Bean species matters too, since robusta is commonly reported to carry around twice the caffeine of arabica, but almost nothing sold as specialty coffee is robusta, so for most readers this is a smaller factor than the brewing.

Does matcha have more caffeine than coffee?

Usually no, at typical servings. A 2 g bowl of matcha lands around 52–92 mg. An 8 oz drip coffee lands around 95 mg. Coffee wins, but not by the margin people expect, and a generously scooped 3–4 g matcha latte can comfortably overtake a modest cup of coffee.

The claim you sometimes see, that matcha has more caffeine than coffee, is true only per gram of dry material. Dry coffee grounds and dry matcha powder are not a comparison anybody drinks. Compare what is in the cup.

How much caffeine is in matcha compared to coffee, in one line?

A standard bowl of matcha carries roughly 55% to 95% of the caffeine in a standard cup of drip coffee. Scoop heavier and you cross over.

Matcha vs espresso caffeine

This is the comparison where the internet’s numbers fall apart completely.

The tidy answer: espresso is about 63 mg per 1 fl oz shot (USDA FoodData Central, 171891). On that basis a 2 g matcha, at 52–92 mg, sits in the same neighborhood and may edge ahead.

The real answer: when researchers bought espressos from twenty different coffee shops in Glasgow and measured them, they found 51 mg to 322 mg per serving, with a median of 140 mg (Crozier et al., Food & Function, 2012;3(1):30–33). Cup volumes ranged from 23 ml to 70 ml. The strongest shot held more than six times the caffeine of the weakest.

So, is matcha stronger than espresso? It depends entirely on whose espresso. Against a 51 mg shot, a well-dosed matcha wins comfortably. Against a 322 mg one it is not close.

The practical takeaway is that the barista matters more than the botany. Two identical-looking drinks can differ threefold. The table at the top of this article is a map, not a measurement of your cup, and treating category averages as facts about your own morning will mislead you every time.

Is matcha stronger than coffee? Why the feeling doesn’t match the numbers

Here is where I have to disagree with nearly every article on this topic, including several I otherwise like.

The standard story goes: matcha contains L-theanine, which slows caffeine absorption, giving you calm sustained energy instead of a spike and a crash.

The first part is defensible. The middle is not.

The claim with no evidence behind it

There is no published pharmacokinetic study measuring how fast caffeine from matcha enters the bloodstream. I looked specifically for one. Studies exist on caffeine absorption in general, on coffee and green tea phenolic metabolites, on theanine and cognition. But nobody has given people matcha and tracked their plasma caffeine curve against coffee.

What we know about caffeine generally is that it is absorbed rapidly and essentially completely, reaching peak blood concentration in 30–120 minutes, with a plasma half-life of 2–8 hours in adults (EFSA). Nothing in that literature suggests the delivery vehicle changes it much.

So when you read that matcha releases caffeine slowly over four to six hours, that is not a finding. It is a sentence that got copied between blogs until it started to sound like one. I would rather tell you that than repeat it.

What the evidence does support, and where it gets awkward

The L-theanine story is real, but it is narrower and messier than the pitch.

A 2025 systematic review in Nutrition Reviews screened around fifty randomized controlled trials and found only sixteen poolable, covering 484 participants, at median doses of 82 mg theanine and 89 mg caffeine (Nutrition Reviews, 2025;83(10):1873). Combined theanine and caffeine improved:

• Simple reaction time at two hours, SMD −0.71 (95% CI −0.92 to −0.50)
• Attention-switching accuracy at two hours, SMD 0.33 (95% CI 0.13 to 0.54)
• Digit vigilance accuracy at two hours, SMD 0.20 (95% CI 0.02 to 0.38)

Note the timing. At the one-hour mark, the effect on reaction time was not statistically significant (SMD −0.16, CI −1.04 to 0.72, an interval that crosses zero).

Separately, a 2026 meta-analysis of 31 trials with 1,168 participants found L-theanine alone produced a reliable short-term improvement in attention, with choice reaction time at SMD 0.51 (CI 0.25 to 0.77) after a single 200 mg dose, and no dependable effect on fatigue or anxiety (Molecular Psychiatry, 2026).

Now the awkward part, which I have not seen quoted anywhere. In a trial of 24 people given 75 mg caffeine with 50 mg theanine, caffeine alone reduced oxygenated hemoglobin in the prefrontal cortex, a vasoconstrictive effect. Adding theanine removed it. But the same paper reports that theanine also eradicated caffeine’s effects on cognition and mood at those doses (Dodd et al., Psychopharmacology, 2015). Theanine did not add to caffeine there. It cancelled it.

That cuts against the tidy best-of-both-worlds story at least as hard as it supports the vasoconstriction one, and any honest summary has to hold both.

So what should you expect?

Matcha’s calmer quality is best explained by L-theanine acting alongside caffeine rather than by slower caffeine absorption. The measured effects are modest, land mainly in attention and subjective alertness rather than raw stimulation, and tend to appear around the two-hour mark rather than immediately. In at least one trial the combination performed no better than caffeine alone.

There is also a duller explanation that I think is doing real work. Most people take a smaller caffeine dose as matcha than as coffee, and drink it more slowly, from a bowl, sitting down. Sixty milligrams sipped over ten minutes feels different from a hundred and forty knocked back at a station kiosk. That is not the plant. That is the ritual.

I am not dismissing the ritual. I built a business on it. I just will not sell it to you as pharmacology.

One caveat about the theanine numbers

This undercuts a figure you will see quoted with great confidence elsewhere. Published L-theanine content in matcha runs from 4.22–9.85 mg/g in a direct measurement of three commercial products (Antioxidants, 2022) up to 44.65 mg/g in a review (Molecules, 2021).

That is a five- to ten-fold disagreement. At 2 g, your bowl might hold 8 mg of theanine or it might hold 89 mg. The trials above used 50–250 mg. A single serving of matcha may well sit below the doses at which effects were demonstrated, and anyone telling you precisely how much theanine is in your cup is guessing.

Benefits of matcha vs coffee: the finding nobody quotes

Matcha’s other selling point is antioxidants, specifically EGCG. Here the research contains a genuinely awkward result.

When researchers compared ceremonial matcha against ordinary bagged and gunpowder green tea, EGCG came in at 56.57 ± 2.19 mg/g for matcha versus 46.14 ± 4.24 mg/g for the regular green tea, a difference that was not statistically significant (Plant Foods for Human Nutrition, 2023).

Caffeine, in the same study, was significantly higher in matcha (31.45 vs 23.08 mg/g, p = 0.002).

Read that against the usual pitch. In this dataset, what matcha reliably gives you more of than ordinary green tea is caffeine, not EGCG.

The EGCG picture gets shakier the more papers you open. The 2025 study measured 21.76 mg/g for ceremonial and 26.10 mg/g for culinary, with culinary slightly higher and again not significantly so. The 2023 study measured 56.57 and 50.53 mg/g for the same two grades. That is a two- to three-fold disagreement between two recent papers, driven largely by extraction method. Anyone citing one number is citing one lab.

And the frequently repeated claim that matcha has 137 times the EGCG of regular green tea traces to a single 2003 chromatography paper comparing two specific products. It contradicts the 2023 analysis above. I would not build a decision on it.

Where coffee has the stronger case: a much larger body of long-term observational research, and its own well-studied polyphenols. “Matcha is healthier than coffee” is not a settled claim.

Where matcha has the stronger case: you consume the whole leaf rather than an infusion, and the theanine-plus-caffeine combination has real, if modest and inconsistent, trial evidence behind it.

Both are, by the standards of things people drink every day, unusually benign.

Matcha as a substitute for coffee: how to switch

If you want to swap, here is what I would tell you across the counter.

Match the dose, not the drink. Work out roughly what you get from coffee, then aim for the same in matcha. One 8 oz drip coffee at about 95 mg is roughly 2–3.5 g of matcha, depending on grade. Do not guess this. Weigh it once and you will know your scoop forever.

Expect it to be less immediate. Trial effects showed up at two hours, not one. If you use coffee as a switch you flip at 8:01am, matcha will disappoint you in week one.

Mind the iron question if it applies to you. Tea polyphenols are well documented to inhibit non-heme iron absorption (Hurrell, Reddy & Cook, British Journal of Nutrition, 1999). If you already know you run low on iron, leaving an hour between matcha and an iron-rich meal is a reasonable precaution, and worth raising with your doctor rather than a blog.

Keep the daily numbers in view. The FDA describes 400 mg a day as generally not associated with negative effects in healthy adults (FDA). For pregnancy, the American College of Obstetricians and Gynecologists has described moderate consumption of under 200 mg a day as not a major contributing factor in miscarriage or preterm birth (ACOG). Matcha counts toward these exactly as coffee does.

Buy the grade you will drink. Culinary grade whisked plain is bitter and will put you off the whole idea. Drinking it straight, buy ceremonial. Making lattes with milk, culinary is fine and cheaper.

But I should tell you what “ceremonial grade” actually is, because this is the fourth thing on this list that everybody repeats and nobody checks. It has no legal definition in Japan. There is no certification behind it, no governing body, no standard a tin has to meet to print those words. It is a marketing phrase invented for export.

You can watch this happen on Amazon. One analysis of the 84 best-selling matcha powders there found 52 of them calling themselves ceremonial grade — and 21 of those did not name the region the tea was grown in. Fifteen did not state a country of origin at all. Across the same 84 products the price ran from about one cent to over three dollars per gram, a 315-fold spread for a phrase that means nothing.

So ignore the grade and read the label. A named region and a named harvest is the whole test. A tin that says Uji, first harvest, is telling you something checkable. A tin that says ceremonial grade premium is telling you what it hopes you will assume.

Drinking it straight

First harvest, from Uji, on the label

This is the test applied: a named region, a named harvest, printed where you can read it before you buy. That is worth more than the word ceremonial in a serif font.

Check on Amazon →

Making it with milk

A barista or culinary blend

Milk flattens the delicate part you paid for. Blends built for lattes are chosen to stay legible through dairy, and they cost less per bowl, which means you will actually drink matcha instead of rationing it.

Check on Amazon →

Do not throw out the coffee. I make both every week and I do not think you have to choose. Matcha in the afternoon and coffee in the morning is a perfectly good answer that no listicle will give you, because it does not sell anything.

What you need to make decent matcha

You need less than the starter sets suggest. Four things do almost all the work, and one of them matters far more than the others.

Do not cheap out here

A chasen — 80 prongs, made in Nara

Cheap whisks are cut from old bamboo stripped with chemicals, and that smell goes into the tea. I have tasted a good matcha ruined by a bad whisk, which is a stupid way to lose a bowl. There are seventeen traditional whisk makers left in Japan. A milk frother makes foam, but not this texture — the tines do something a wire whisk cannot.

Check on Amazon →

Ten seconds, every time

A fine sieve

Matcha clumps, and most people who think they dislike matcha have been drinking lumps. Get one wide enough to sit over the bowl, so you are not holding it with one hand and tapping with the other.

Check on Amazon →

Optional, honestly

A chawan, or any wide bowl

A mug is too narrow for the whisk to move properly, but a cereal bowl genuinely works and I would not talk you out of one. If you do buy a chawan, buy one where the kiln is named — the flat floor is the part that matters, not the glaze.

Check on Amazon →

If you want one box

A starter set

A reasonable way to start tomorrow instead of next month. Treat the whisk in it as a consumable — when the tines start to go, replace it with a Japanese one. That path costs less than buying two mediocre whisks, which is what most people end up doing.

Check on Amazon →

Skip the electric frother, the ceremonial-grade-for-lattes upsell, and any starter set that includes a decorative tea caddy you will never fill.

FAQ

Does matcha have more caffeine than coffee?

Usually no. A standard 2 g bowl of matcha contains roughly 52–92 mg of caffeine, while an 8 oz cup of drip coffee contains about 95 mg. Matcha only exceeds coffee if you use a heavier scoop of 3 g or more, or if the coffee is unusually weak.

How much caffeine is in matcha compared to coffee?

A standard bowl of matcha carries roughly 55% to 95% of the caffeine in a standard 8 oz cup of drip coffee. Per gram of dry powder matcha is higher, but that is not a comparison anyone drinks.

How much caffeine is in a teaspoon of matcha?

A level teaspoon holds roughly 2 g of matcha powder, which is about 52–92 mg of caffeine based on lab values of 26–46 mg per gram. A heaped teaspoon can hold 3–4 g, pushing it to 78–184 mg.

Is matcha stronger than espresso?

It depends on the espresso. Nominally a 1 oz shot is about 63 mg, which a well-dosed matcha can match or exceed. But espressos measured across twenty real cafes ranged from 51 mg to 322 mg, so the comparison is not stable enough to answer in general.

Does matcha give you a slower caffeine release than coffee?

There is no published study measuring caffeine absorption from matcha, so this popular claim is unverified. What research does support is that L-theanine taken with caffeine improves attention-switching accuracy and subjective alertness, with effects appearing around two hours after intake.

How many cups of matcha equal one cup of coffee?

Roughly one to two, depending on your scoop. To match an 8 oz drip coffee at about 95 mg you would need about 2–3.5 g of matcha in total.

Will matcha keep me awake at night?

It can. Caffeine’s half-life in adults is 2–8 hours, so a late-afternoon matcha can still be in your system at bedtime. I stop serving both by mid-afternoon, which is partly why we close at 4:30.

Is matcha healthier than coffee?

Not clearly. Matcha delivers L-theanine and whole-leaf compounds, while coffee has a much larger long-term evidence base and its own antioxidant profile. One 2023 analysis found no statistically significant EGCG advantage for matcha over ordinary green tea, though matcha did contain significantly more caffeine.

Can I drink both matcha and coffee in the same day?

Yes. Just count the total against the FDA’s 400 mg per day guidance for healthy adults.

The bottom line

Drip coffee generally has more caffeine per serving than matcha, but both ranges are wide enough that preparation beats plant every time. The calm energy people describe in matcha has a partial basis in L-theanine research, with modest effects on attention that show up around two hours in, and at least one trial where the combination did no better than caffeine alone. The specific claim that matcha releases caffeine slowly has no published evidence behind it at all.

The most useful version of everything above: buy a scale. Weigh your matcha, weigh your coffee, and in one week you will learn more about your own caffeine than any comparison article can tell you, this one included.

And if you are ever in Kyoto on a Sunday afternoon, come to Demachiyanagi and I will make you both, and we can argue about it properly.

This article is general information and reflects my reading of the published research. It is not medical advice. If you are pregnant, taking medication, or managing a condition such as anemia or an arrhythmia, talk to your doctor about caffeine rather than a cafe owner.

Kohei Takeda runs Fudemame Coffee, a one-day-a-week cafe in Kyoto, where he roasts coffee in small batches and serves tea across the same counter.

Sources

• Caffeine and EGCG across matcha grades — Journal of Food Measurement and Characterization, 2025
• Matcha vs. green tea composition — Plant Foods for Human Nutrition, 2023
• Matcha composition review — Molecules, 2021;26(1):85
• European commercial matcha analysis — Antioxidants, 2022;11(5):889
• Brewed coffee — USDA FoodData Central 171890
• Espresso — USDA FoodData Central 171891
• Espresso caffeine in commercial outlets — Crozier et al., Food & Function, 2012;3(1):30–33
• Caffeine guidance — U.S. Food and Drug Administration
• Moderate caffeine consumption during pregnancy — ACOG
• Tea, theanine and caffeine RCT meta-analysis — Nutrition Reviews, 2025;83(10):1873
• L-theanine RCT meta-analysis — Molecular Psychiatry, 2026
• Caffeine, theanine and cerebral blood flow — Dodd et al., Psychopharmacology, 2015
• Caffeine pharmacokinetics — EFSA scientific opinion
• Inhibition of non-haem iron absorption by polyphenolic beverages — Hurrell, Reddy & Cook, British Journal of Nutrition, 1999
• Amazon best-selling matcha market analysis — One With Tea

As an Amazon Associate I earn from qualifying purchases.

喫茶ふでまめ

珈琲と手しごとの店

Instagram

京都市左京区・出町柳駅付近にある「リバーサイドカフェ」内にて日曜の昼だけ営業する、江戸っぽくてお茶目な自家焙煎珈琲の喫茶店です。