Many people describe matcha’s energy as different from coffee — alert but calm, focused without jitters. That description is consistent enough to take seriously as reported experience, but it has not been tested: no human trial compares matcha with coffee head to head on jitteriness, anxiety or crash. What has been measured is the two compounds behind the claim, mostly in isolation.
The Two Key Compounds
Caffeine
Matcha’s caffeine tracks the powder rather than the cup — roughly 19-44mg per gram, so a standard 2g serving (about one level teaspoon) is about 38-89mg, most often around 60-70mg. Caffeine works by blocking adenosine receptors in your brain. Adenosine normally accumulates during waking hours and makes you feel sleepy. Caffeine blocks this signal, keeping you alert and increasing dopamine release.
L-Theanine
L-theanine is an amino acid found almost exclusively in tea. It crosses the blood-brain barrier in a dose-dependent way, within roughly 30 to 60 minutes.
The mechanisms usually listed for it — increasing GABA, elevating serotonin and dopamine, dampening glutamate signaling — come from animal and cell studies rather than human ones. L-theanine binds the NMDA, kainate and AMPA receptors 80- to 30,000-fold more weakly than glutamate does, and a 2025 review of the field concludes that rigorous human mechanistic evidence remains limited. These are the proposed mechanism, not established human physiology.
Matcha carries more L-theanine per cup than steeped green tea for a structural reason rather than a chemical one: you drink the whole leaf as powder, so nothing stays behind in a discarded leaf. Per gram of dry leaf the picture is less dramatic — standard sencha measures 18.91mg/g and gyokuro 30.84mg/g, both inside or above matcha’s measured 2.48-44.65mg/g range. Shade-growing does raise L-theanine, and it raises caffeine at the same time while cutting catechins by more than half. Ceremonial grade runs higher in both caffeine and L-theanine per gram than culinary grade, with the larger gap in L-theanine, though no reliable per-gram figures by grade have been published.
The Synergy Effect
Controlled trials of the two compounds together — as isolated compounds, not as matcha — find better cognitive performance than either alone:
Owen et al. (2008): 100mg L-theanine with 50mg caffeine, 27 participants. Caffeine alone improved attention-switching accuracy at 90 minutes; the combination improved both speed and accuracy. The authors were affiliated with Unilever.
Giesbrecht et al. (2010): 97mg L-theanine with 40mg caffeine, 44 participants. The combination improved task-switching accuracy while increasing alertness and reducing tiredness. Also Unilever R&D authors.
Nobre et al. (2008): 50mg L-theanine on its own increased alpha brain wave activity between 45 and 105 minutes. A 2g matcha serving can reach 50mg, but only in the upper half of the measured product range — the two Japanese matchas characterized for 2024 clinical trials carried 48-50mg. This finding also needs its caveats attached: it is a two-page conference-supplement article with Unilever-affiliated authors, and a 2025 randomized placebo-controlled attempt at 200mg in 30 adults found a higher-alpha pattern that did not reach statistical significance.
Pooled, the picture is better evidenced than any single trial. A 2025 meta-analysis of 16 randomized trials (484 participants) found the combination improved attention-switching accuracy in the first hour and simple reaction time in the second. A separate 2026 meta-analysis of 31 trials (1,168 participants) found a single 200mg dose of L-theanine improved choice reaction time. Attention is the part of this literature that holds up.
Two mechanisms are proposed for why the pair might feel different from caffeine alone. L-theanine may counteract caffeine-induced jitteriness through GABA enhancement — animal and cell evidence, not human. And L-theanine cancels caffeine’s vasoconstrictive effect on brain blood flow, which was measured directly: at doses and ratios matched to one to two cups of tea, the combination “eliminated the vasoconstrictive effect and behavioural effects of caffeine” — and at those levels this did not lead to a positive impact on behaviour. The whole sentence matters, not the flattering first half.
Important Caveats
The research is promising but comes with limitations:
Dose Questions
Trials of the combination used a median of 82mg L-theanine with 88.74mg caffeine. Trials of L-theanine on its own used a median of 200mg.
A 2g matcha serving delivers about 38-89mg of caffeine, most often around 60-70mg, and roughly 5-90mg of L-theanine — the two matchas assayed for 2024 trials landed at 48-50mg.
Read plainly, the gap is on the L-theanine side only. Matcha reaches study-level caffeine comfortably; a typical serving sits inside the range the combination trials used. Its L-theanine lands below their median, and well below the 200mg behind the L-theanine-only attention results. Which end of that 5-90mg span your particular tin sits at is the single number that decides whether this research applies to your cup, and it is not printed on the tin.
The Ratio Problem
The dose gap is the well-known half. The sharper one is the ratio.
Measured matcha runs 1.63 to 3.43 parts caffeine to one part L-theanine by weight. Every trial named above ran the other way round: Giesbrecht at 0.41, Owen at 0.5, the pooled combination medians at 1.08, and Dodd’s deliberately tea-realistic arm at 1.5. Matcha is more caffeine-dominant than the trial explicitly designed to mimic tea — and that trial is the one that found no positive behavioural effect. The 2025 meta-analysis names this as the field’s own limitation: few studies used tea, or tea-equivalent doses and ratios of the bioactive compounds.
Drinking more matcha does not close either gap, because the ratio is a property of the powder.
Study Limitations
- The meta-analysed synergy trials covered 484 participants across 16 studies — an average of about 30 people each
- Most evidence is acute (1-2 hours post-dose), not long-term
- Industry funding is prevalent: both named attention studies and the alpha-wave study carry Unilever affiliations
- Individual variation is substantial and understudied
Matcha Itself, Versus Its Compounds
Nearly all of the above tests isolated compounds. The matcha-specific layer is thinner. A pooled analysis of 12 studies and 569 participants found no statistically significant cognitive benefit from matcha versus placebo — though it is a conference meeting abstract rather than a peer-reviewed paper, and its published confidence interval and p-value contradict each other, so it is worth reading for direction only. The one positive matcha trial, in 39 people, found lower anxiety with a theanine-rich matcha than with a theanine-poor placebo matcha. That is matcha against matcha, not matcha against coffee.
The “Slow Release” Myth
You’ll often hear that matcha caffeine releases slowly over hours. This lacks pharmacokinetic evidence. Caffeine from matcha absorbs as fast as caffeine from coffee — about 99% bioavailable within 45 minutes, peaking at 30-60 minutes regardless of source. The smoother experience comes from L-theanine modulating caffeine’s effects, not from slower absorption.
What This Means for You
High confidence: The caffeine + L-theanine combination improves attention in the first two hours. Two independent meta-analyses published in 2025 and 2026 agree on this.
Moderate confidence: Quality matcha carries enough L-theanine to produce measurable effects. Unno’s criterion has two halves — theanine above 17mg/g and a molar ratio of caffeine plus EGCG to theanine plus arginine below 2 — and it was set against a 3g daily intake. Quoting one half without the other overstates what matcha clears.
Low confidence: That the combination reduces jitteriness relative to caffeine on its own. The only pooled jitteriness result compares the combination against placebo, and the 2025 meta-analysis pools no comparison against caffeine alone. Subjective alertness and overall mood at the two-hour mark both have confidence intervals that cross zero, and relaxation intervals include no difference.
Low confidence: That a single serving of matcha consistently delivers the studied cognitive benefits. Individual responses vary significantly based on caffeine sensitivity, genetics, and tolerance.
Practical Recommendations
Choose quality matcha. A 2018 study judged matchas against theanine above 17mg/g plus the ratio criterion above: 32 of 76 samples bought in Japan met both, against one of 67 bought overseas. On the theanine threshold alone, 66% of Japanese samples met it versus 9% of overseas samples.
Quantity does not close the gap. Two servings (4g powder) carries 57-178mg of caffeine — past the pooled trial median, and potentially past a cup of coffee — while leaving the caffeine-to-theanine ratio exactly where it was.
Set realistic expectations. The effect is real but more subtle than coffee’s immediate kick. Think “focused calm” rather than “energized.”
Individual variation matters. Caffeine sensitivity varies up to 15-fold between people, driven by CYP1A2, which handles over 90% of caffeine clearance. What works for one person may not work for another.
The Bottom Line
Matcha’s “calm energy” has a genuine biochemical basis, and the attention half of it now rests on two independent meta-analyses. Both compounds are demonstrably present in the drink at measured levels.
What the evidence does not yet support is the leap from those compounds to this drink. The L-theanine dose in a serving sits below what the trials used, the caffeine-to-theanine ratio is inverted relative to all of them, the one trial run at tea-like ratios found no behavioural benefit, and the calm half — the part people arrive for — is where the measurements are thinnest.
Try it and see how your body responds. For some people, the difference from coffee is dramatic. For others, it’s subtle or unnoticeable.