The Edible Dose Chart by Weight Is Fiction. Here Is What Actually Predicts Your Dose.

Every edible dosage chart by weight on the internet was invented by whoever published it. There is no study behind any of them, and the three that rank for the phrase this week cannot even agree with each other: a 250-pound beginner is told 10 to 15mg by one, 10 to 20mg by the second, and 20 to 30mg by the third, from tables that cite nothing. The idea is borrowed from laboratory protocols that dose THC per kilogram of body weight, because trials standardize exposure that way, and somewhere a content writer turned a study convention into consumer advice. The evidence points the other direction. This page walks through it, and then publishes the chart that should exist, which has experience and goal on the axis where weight used to be.

Does body weight change how much THC you need?

Not in any measurable, consistent way. The one controlled trial to test the question, Ewell and colleagues (2021, Pharmaceuticals), gave adults five commercial 10mg edibles in a crossover design and measured body mass index, fat mass, lean mass, and body fat percentage against the blood THC curve for each product. No body-composition variable predicted exposure consistently across the five products, and the authors concluded fat tissue contributes little to short-term edible pharmacokinetics.

Two honest caveats about that study, both of which cut against the charts rather than for them. It was small, with seven participants completing every arm, which is thin evidence for anything. It is also more evidence than any weight chart has ever cited, because the charts cite nothing. And three of the five products were fast-acting or water-soluble formats (Ripple's dissolvable powders and quick sticks, Wana's fast-acting gummies) alongside two conventional gummies, so the trial covered the modern shelf rather than a single brownie. Time to peak blood THC ran from 35 to 90 minutes across products, a spread driven by the format, not the participant.

The intuition behind the charts is real physiology applied to the wrong hour. THC is fat-soluble and does end up in fat tissue, and a larger body does dilute a fixed dose across more tissue. Both of those are slow processes measured in hours and days, which is why body composition shows up in how long THC lingers (the drug-testing essay covers that tail) and not in how hard the first night lands. The first night is set by absorption and by the liver's first pass, and in Ewell's data the thing that moved the peak was the product format, with time to peak running 35 to 90 minutes depending on whether the edible was a conventional gummy or a water-soluble one. Weight had no consistent say in it.

The per-kilogram origin story explains why the charts feel scientific. Laboratory studies do dose by weight: Ramaekers and colleagues (2016, Scientific Reports) gave 300 micrograms of THC per kilogram to 122 volunteers, which puts an 80-kilogram participant at 24mg. Researchers do that to standardize exposure across a cohort, the way an anesthesiologist doses by weight, and it says nothing about whether a heavier person needs more to sleep. The chart-makers kept the arithmetic and dropped the reason for it.

What actually predicts your edible dose?

Four things with measured effect sizes, none of which appear on a weight chart: which version of one liver enzyme you carry, your sex, whether the label on the gummy is right, and what you ate. The largest of them, a single CYP2C9 gene variant, triples THC exposure, which is bigger than the entire spread a weight chart claims from its lightest row to its heaviest.

The enzyme. CYP2C9 is the liver enzyme that does most of the work of clearing THC, and it comes in variants. Sachse-Seeboth and colleagues (2009, Clinical Pharmacology and Therapeutics) gave oral THC to 43 volunteers of known genotype and found the median THC exposure in CYP2C9*3/*3 carriers was three times that of normal *1/*1 metabolizers, with a trend toward more sedation. The NCBI Medical Genetics Summaries chapter on dronabinol and CYP2C9 puts the *3 allele below 10% in most populations and very rare in African populations, which by the arithmetic of a single allele makes the *3/*3 group a small fraction of one percent of any crowd. Small and real: somewhere in every hundred first-time buyers is someone for whom the 5mg gummy is a 15mg gummy, and no scale in the world would have warned them. The same enzyme metabolizes warfarin, which is why the disclaimer at the bottom of this page names that drug.

Sex. Sholler and colleagues (2021, Addiction Biology) pooled four placebo-controlled studies, 27 men and 23 women who used cannabis infrequently, and found women reached higher peak blood concentrations of 11-hydroxy-THC after oral cannabis and rated the drug effect higher, and the difference held after controlling for body weight. Aghaei and colleagues (2024, Psychopharmacology) ran a dedicated crossover trial and found the physiological effects similar across sexes with a more pronounced subjective high in women. Neither study supports a "women take half" rule, and this page does not propose one. They do show that the variable the charts ignore moves the needle in a direction the charts cannot predict.

The label. This is the one that makes the whole exercise absurd. A weight chart presumes a 10mg gummy contains 10mg. Vandrey and colleagues (2015, JAMA) bought 75 edibles from 47 brands at medical dispensaries in three cities and found 17% accurately labeled, with the rest over or under. That was 2015 and the medical market. The 2026 follow-up from Colorado, Limbacher and colleagues in PLOS One, tested 29 edibles from licensed retail and found a median labeled claim of 10mg against a median observed 9.3mg, a statistically significant shortfall that still sat inside the state's legal 15% tolerance. So the modern gummy is better than the 2015 gummy and still, by law, anywhere from 8.5 to 11.5mg. A chart that recommends 7.5mg for your weight class is asking you to measure something the product itself does not.

Fed state, tolerance, and format each move the number too, and the site's dosing guide and duration guide cover them in detail. The point of listing them here is only that every one of them has a measured effect and weight, on the best available evidence, does not. It is also why the site's dosing calculator never asks what you weigh.

How much of a "10mg" gummy is actually 10mg?

Somewhere between 8.5 and 11.5mg if the product is legal in Colorado, and probably on the low side. The state permits a 15% deviation either way; the 2026 sample ran a median 0.7mg under the label, with labels overstating more often than understating. That range is wider than the gap between adjacent rows on most weight charts.

Put the two numbers side by side. Neurogan's chart, updated in August, separates a 150-pound beginner (2.5 to 5mg) from a 200-pound beginner (5 to 7.5mg) by 2.5mg. The legal variance on a single 10mg gummy is 3mg. The chart's precision is smaller than the product's error bar, which is a polite way of saying the chart is measuring nothing. Hollyweed's version goes further and publishes weight tables for delta-8, THCP, HHC, THCV, and CBD as well, five molecules with five different potencies, all somehow following the same weight logic, none with a citation. That is a template with the word dosing typed into it.

The chart that should exist

Here is the table the evidence supports, with experience and goal where weight used to be. It reports what titration protocols in the clinical literature and patient accounts describe, and it is not an instruction. MacCallum and Russo (2018, European Journal of Internal Medicine) set the convention most cannabis clinicians use: start at the lowest dose that produces benefit, move slowly, and expect that doses above 20 to 30mg a day before tolerance is established bring adverse effects. Everything below follows from that.

Notice what is missing. There is no row for the 250-pound reader, because the evidence gives that reader the same first dose as everyone else. The chart-makers would send that person to 15 or 20mg on night one. That number is a referral to the site's too-much-edible page, and the traffic that page gets is partly the weight charts' doing. The 2.5mg step is possible because the products in the right-hand column exist; the beginner ranking is built around them, and the microdosing guide covers the bottom row in depth. The long tail after the dose, the reason THC lingers in fat tissue for days even though fat does not change the first night's peak, is explained in the drug-testing essay, and it is the one place body composition genuinely matters.

The Grasscity forums hold a near-perfect case study, posted in May 2020 by a first-ever THC user trying edibles for insomnia: 2.5mg of a THC mint on night one and the poster slept like a baby, 7.5mg on night two barely made them tired, and 10mg on night three brought ninety minutes of nausea and body aches with no high to show for it. The dose that worked was the first one. Nobody in the thread asked what the poster weighed, because nobody needed to.

What the weight charts get right

One thing, and it is the thing that matters most: every one of them tells a beginner to start low. Neurogan's bottom row starts at 1mg. Fyre Ants labels 1 to 2.5mg a microdose for new users. If the charts stopped there, they would be fine. The problem is the second row, and the third, where a made-up table hands a heavier first-timer four times the dose with the same confidence it used for the first. Start low is evidence. The rest is a spreadsheet.