Evidence desk · Supplement
L-theanine
The verdict
An amino acid from tea that produces a mild, non-sedating "relaxed alertness." The effects are modest and partly subjective — you feel a little calmer, maybe a little more focused. The best-studied real effect isn't theanine alone; it's theanine + caffeine, which slightly improves sustained attention and accuracy. Standalone benefits for anxiety and sleep are small and mixed, and there are no hard outcomes.
It's very safe, cheap, and acute, so the bar to try it is low — because the cost and risk are low, not because the evidence is strong. Treat it as a pleasant as-needed tool, not a foundation or a performance lever.
Before you read on
You take 200 mg of L-theanine with your morning coffee and the day feels smoother. Quick: how would you tell whether the theanine did that — versus the coffee, versus a good night's sleep, versus simply expecting to feel calmer because you swallowed a capsule labelled "calm"? Hold your answer. This is exactly the kind of acute, "feel"-based effect where that question is hardest to answer honestly.
The claims, sorted
Theanine is sold for a handful of different things, and they don't share a verdict. Separate them before judging any one study.
| Claim | Best evidence | Holds up? |
|---|---|---|
| Attention / accuracy (with caffeine) | Several acute RCTs | Yes — small but the cleanest |
| Reduced stress reactivity | Small acute RCTs, reviews | Partly — modest, subjective |
| Anxiety relief | Small, short RCTs | Weak / mixed |
| Better sleep | A few small RCTs | Weak / mixed |
| Focus / cognition (alone) | Underwhelming RCTs | No clear standalone effect |
What it actually does
L-theanine (γ-glutamylethylamide) is the amino acid that gives green tea its savory, calming character. Taken at a typical 100–200 mg, it produces a subjective, non-drowsy ease — it takes the edge off without slowing you down. That "relaxed alertness" is the whole pitch, and it's plausible: theanine modestly nudges relaxation-associated brain activity and may temper the stress response.
But "plausible and pleasant" is not the same as "proven and large." The standalone effects on anxiety, mood, and sleep come from small, short trials that don't agree with each other. The one place the signal firms up is in combination — which is why the figure below is worth a look.
The honest size of the effect
The cleanest finding is the caffeine + theanine combo: in several acute studies, the pair improved sustained attention and accuracy on demanding tasks more than caffeine alone, and seemed to blunt some of caffeine's jitter. That's a real, replicated effect — but it's small, and it's about a combination, not theanine on its own. Theanine here is best understood as something that smooths caffeine, not a standalone nootropic.
The placebo trap
An acute, subjective "I feel calmer" effect is the single hardest kind of claim to trust, because it's exactly where expectancy dominates. You took a pill labelled "calm," you noticed calm — the pill may have done little or nothing. There are no hard outcomes here (no mortality, no disease endpoints); the outcomes are surrogate and self-reported. So apply the "is it strong enough to act?" question plainly: the evidence is weak, but the cost and risk are also near-zero, which is the only reason "sure, try it" is a defensible answer. Compare that to a supplement like creatine — large, replicated, objective effects — to feel the difference between "strong enough" and "harmless enough."
By phase
Unlike most entries here, theanine is essentially phase-invariant — it isn't tied to the cut, the build, or maintenance. It's an optional, low-stakes tool in all three. The general reasoning below; your specific call is in the For you section.
Phase 1 · Aggressive cut (Zepbound)
Optional, low stakes. The cut doesn't change theanine's case either way — it doesn't affect fat loss, muscle, or appetite. The only relevant angle is comfort: if a dieting day plus caffeine feels jittery, theanine is a cheap, safe thing to take the edge off. Don't expect it to move any number on the scale or in a lab.
Phase 2 · Building muscle
Optional, low stakes. Not a performance or recovery lever — it won't help you build. Some people add it to a pre-workout to soften the caffeine, which is fine, but it's a comfort tweak, not an ergogenic.
Phase 3 · Maintenance
Optional, low stakes. Same as ever: a pleasant as-needed tool for calm/focus. Keep it if you genuinely notice and value the effect, drop it if you don't — and watch the placebo and regression-to-the-mean traps when you judge.
For you
You're not running this one as a fixed part of the stack, and it doesn't need to be. Where it's actually interesting for you is the stimulant angle: you take Vyvanse 30 mg (lisdexamfetamine, a stimulant) and presumably some caffeine. People on stimulants and caffeine often use L-theanine to soften the jitter, edge, or wired feeling — that's plausible and low-risk, and it lines up with the one effect theanine has the best evidence for.
My call — a fine as-needed tool, with two caveats
Treat theanine as optional and as-needed, not a daily fixture. Reasonable uses: 100–200 mg with coffee, or to take the edge off Vyvanse/caffeine jitter on a wired day. It's cheap and very safe, so there's little downside to trying.
Two caveats. First: run any stimulant-stacking past your prescriber. You're combining it with a prescribed stimulant and lisinopril — theanine is benign, but "what I take alongside my meds" is exactly the conversation to have with the person managing them. Second: manage expectations. This is a "feel slightly calmer / a touch more focused" tool, not a needle-mover. It won't do for you what creatine, protein, or your training do.
On testing it: the effect is acute, so an n-of-1 is easy — take it some days, skip it others, notice the difference. But be honest that acute "feel" effects are precisely where placebo is strongest, so your read will be noisy. That's fine for a harmless, cheap tool: you don't need clean proof to justify something this low-stakes. Just don't talk yourself into it being more than it is.
Profile used: 41, 5′9″, Phase 1 cut on Zepbound, on Vyvanse 30 mg (a stimulant) + lisinopril + tretinoin, biomarkers unremarkable for this one. Tell me if any of that moves and I'll update.
Recall check
- What is L-theanine's single best-evidenced effect — and what is the catch about it?
- Why does "it's safe and cheap" lower the bar to act, even when the evidence is weak?
- Why is an acute "I feel calmer" claim especially vulnerable to the placebo trap?
- Why might someone on Vyvanse or caffeine reach for theanine — and what's the one step to take first?
Explain it back
In one sentence: why is "harmless and cheap enough to try" a different (and lower) bar than "strong enough to count on"?
Check it yourself
This verdict is just the toolkit applied. Re-derive it with the evidence ladder, surrogate vs hard outcomes, and is it strong enough to act? See it in the wider stack via the protocol and operating filters; for the contrast of what strong evidence looks like, compare creatine. And since theanine is a classic behavioral/supplement-lever pick, weigh it against the guru tone in Huberman.
Sources · Owen et al., 2008 — caffeine + L-theanine improves attention (Nutr Neurosci, PubMed) · Giesbrecht et al., 2010 — L-theanine + caffeine and alertness (Nutr Neurosci, PubMed) · Williams et al., 2020 — review of L-theanine on stress & cognition (Plant Foods Hum Nutr, PubMed). Evidence summaries, not medical advice — confirm anything that touches your medication, especially stimulant stacking, with your prescriber.