Martini: two and a half drinks in one glass

The best tested fact about a martini is about the stomach you drink it on. In a crossover trial of 21 healthy adults given a measured dose of alcohol after an overnight fast, peak blood alcohol reached 0.064% with no food and 0.020% after a 635 calories meal, over 90 minutes of breath tests. The drink there was a standardized alcohol dose, and nobody in it drank a martini. Two more small crossover trials point the same way. In 20 adults drinking vodka, a meal or a food bar cut alcohol absorption by about 30% against fasting. In 9 men, peak blood alcohol after a meal was roughly half the fasting peak whether the meal was rich in fat, carbohydrate or protein. These short tests measured how fast the alcohol reached the blood, and none of them measured whether food makes drinking any safer. The glass itself is stronger than it looks. The USDA survey record for a martini holds 29 g of alcohol per 3.5 oz (100 g), so its 120 g drink portion carries 34.8 g, about 2.5 US standard drinks of 14 g each. That comes from one standard recipe, and a real bar pour may be bigger.

As people eat it, per 3.5 oz (100 g)

How much is actually in it

• A standard martini in the USDA survey database carries 201 calories per 3.5 oz (100 g), or about 241 calories in the 4.2 oz (120 g) portion coded as one drink. — US dietary survey food record, not a measured bar serving Evidence E source
  201 calories/100 g; 60 calories per fl oz (30 g); 241 calories per drink (120 g)
• The USDA survey martini holds 29 g of ethanol per 3.5 oz (100 g), so its 120 g drink portion contains 34.8 g of alcohol, about 2.5 US standard drinks. — US dietary survey food record, not a measured bar serving Evidence E source
  29 g ethanol/100 g; 8.7 g per fl oz; 34.8 g per 120 g drink (2.5 x 14 g)
• A classic martini is almost free of carbohydrate at 0.15 g per 3.5 oz (100 g), while the flavored martini record carries 6.36 g per 3.5 oz (100 g) and less alcohol at 23.6 g. — US dietary survey food records Martini (FDC 2710639) and Martini, flavored (FDC 2710640) Evidence E source
  Carbohydrate 0.15 vs 6.36 g/100 g; ethanol 29 vs 23.6 g/100 g; 201 vs 187 calories/100 g

What a real portion looks like

• A US standard drink is 14 g of pure alcohol, which is the yardstick for counting a martini. — US adults Evidence E source
  14 g or 0.6 fl oz pure alcohol per standard drink

What your body absorbs

• In a crossover RCT of 21 adults, eating a 635 calories meal before a moderate alcohol dose cut peak blood alcohol from 0.064% to 0.020% and alcohol bioavailability by 67.9%. — 21 overnight-fasted healthy adults (10 male, 11 female), alcohol 0.3-0.35 g/kg, breathalyzer over 90 minutes (n = 21) Evidence B source
  Peak BAC 0.064 (no food) vs 0.047 (210 calories snack mix) vs 0.020% (635 calories meal); 90-min bioavailability -22.0% to -67.9%
• In a four-way crossover RCT of 20 adults drinking vodka, a meal or a food bar before drinking reduced alcohol absorption by about 30% compared with fasting. — 20 healthy adults, vodka 0.35 g/kg (men) or 0.30 g/kg (women) in the morning, breathalyzer over 90 minutes (n = 20) Evidence B source
  Relative bioavailability (AUC0-t) about 30% lower after a meal or food bar than fasting
• In 9 men, drinking 0.30 g/kg ethanol after a meal roughly halved peak blood alcohol versus an empty stomach, whatever the meal's fat, carbohydrate or protein content. — 9 healthy men, ethanol 0.30 g/kg within 15 minutes of a standardized breakfast, five randomized sessions (n = 9) Evidence B source
  Peak BAC 13.3-17.7 mg/dl fed vs 30.8 mg/dl fasting; AUC 1270-1767 vs 3210 mg/dl min

What it does to your health

• In randomized comparisons with 12 volunteers per block, 3.4 fl oz (100 ml) of whisky after a solid meal slowed gastric emptying more than beer or wine, and only the distilled drink delayed orocecal transit. — three blocks of 12 healthy volunteers, 1485 kJ solid meal then 14 fl oz (400 ml) beer, 6.8 fl oz (200 ml) red wine or 3.4 fl oz (100 ml) whisky (43.5% vol) or control fluids (n = 36) Evidence B source
  Delay of gastric emptying grew with ethanol concentration; orocecal transit delayed by whisky, not by beer or wine
• In a randomized crossover study of 35 women, alcohol clamped at 50 mg% before a meal increased how much they ate and lowered ghrelin compared with saline. — 35 non-obese women, intravenous 6% v/v ethanol clamped at breath alcohol 50 mg% vs saline, then ad libitum pasta meal (n = 35) Evidence B source
  Food consumption significantly greater after alcohol; ghrelin reduced; size not given in abstract

Safety, limits and interactions

• In two double-blind crossover RCTs of 36 men, ranitidine 300 mg four times a day did not change peak blood alcohol after 0.5 g/kg, fed or fasting. — 36 healthy nonalcoholic men aged 25 to 48, ranitidine 300 mg four times daily or placebo for 8 days, alcohol 0.5 g/kg (n = 36) Evidence B source
  No significant difference in Cmax, AUC or Tmax of blood alcohol
• Three of four US-acquired lung fluke infections in one case series came from eating live crabs at sushi bars, including crabs served in martinis. — 4 patients (3 men, 1 woman) aged 20 to 66 with pleuropulmonary Paragonimus westermani infection (n = 4) Evidence D source
  Eosinophilic pneumonia in all 4; live-crab martini named as a previously unreported route

What people believe that the data does not support

• A 2.7 tbsp (40 ml) shot of 40% spirit after a 576 calories meal did not speed gastric emptying in 10 men (half-emptying 119-126 min vs 123 min with water), while a slow walk did (107 min). — 10 healthy men, 576 calories meal followed by 2.7 tbsp (40 ml) brandy, herbal liqueur, pear brandy or aquavit (40% v/v), espresso or water (n = 10) Evidence B source
  Gastric t1/2 123 +/- 5 min (water) vs 119-126 min (spirits), not significant; walking 107 +/- 5 min, p=0.02

Who this works differently for

• Under standardized conditions, the same amount of alcohol tended to produce a higher peak and larger blood alcohol curve in young women than in young men. — young healthy volunteers drinking diluted ethanol, 18 blood samples over 6 hours Evidence C source
  Higher peak and AUC in women at the same dose (trend); meal lowered both
• Twelve women tested while exclusively breastfeeding and again after weaning handled the same 0.4 g/kg alcohol dose differently, with lower breath alcohol on the descending curve during lactation. — 12 women breastfeeding 3-5-month-old infants, retested months after lactation ended, alcohol 0.4 g/kg fed and fasted (n = 12) Evidence C source
  Breath alcohol lower in the descending phase during lactation than after; pumping before drinking lowered levels
• CDC advises not drinking alcohol at all for people who are pregnant or might be pregnant. — people who are pregnant or might be pregnant Evidence E source
  No alcohol for this group

The bottom line

A classic martini is close to carbohydrate free at 0.15 g per 3.5 oz (100 g). The flavored martini record has 6.36 g of carbohydrate and a little less alcohol, 23.6 g per 3.5 oz (100 g). The energy barely moves, 201 against 187 calories per 3.5 oz (100 g), because almost all of it comes from the alcohol. One 120 g drink of the classic recipe is about 241 calories. Many people believe a shot of spirit after dinner helps the meal go down. This was tested in 10 men after a 576 calories meal, and 2.7 tbsp (40 ml) of 40% spirit did nothing measurable: half the meal left the stomach in 119 to 126 minutes against 123 minutes with water. A slow walk did better, at 107 minutes. That trial used an after dinner drink, so it tells you nothing about a martini taken before food. The idea that a drink before dinner makes you eat more has some support. In a crossover study of 35 non-obese women, alcohol given intravenously before a pasta meal raised how much they ate and lowered the hunger hormone ghrelin, and the abstract gives no size for the extra food. In 36 volunteers, 3.4 fl oz (100 ml) of whisky after a solid meal slowed stomach emptying, and the delay grew with the strength of the drink, again with no size reported. Women and men do not always get the same result from the same glass. In an older non-randomized study of young volunteers, the same amount of alcohol tended to give women a higher peak, and the abstract reports it as a trend without statistics. Twelve women tested while breastfeeding and again after weaning handled a 0.4 g/kg dose differently between the two stages, which is no evidence that drinking while breastfeeding is safe. The CDC advises no alcohol at all for anyone who is pregnant or might be pregnant. On medicines, the only drug tested here was ranitidine. In two double-blind crossover trials of 36 men taking 300 mg four times a day for 8 days, it did not change peak blood alcohol after 0.5 g/kg, and that result says nothing about any other drug. One case series of 4 lung fluke infections acquired in the United States traced 3 to live crabs eaten at sushi bars, including crabs served in martinis. The risk there came from the raw crab, and a plain martini carries none of it. We have no card here on daily limits or on the long-term harms of alcohol, and that gap is not a sign that there are none.

Sources

1. mtni-r1-01 · dataset
   Martini, FNDDS code 93301110, fdc_id 2710639: Energy 201 KCAL per 100 g; 1 fl oz 30 g; 1 drink 120 g
   USDA FNDDS 2021-2023, food code 93301110 (Martini), FDC 2710639 · checked 2026-09-30
2. mtni-r1-02 · dataset
   Martini, FNDDS code 93301110, fdc_id 2710639: Alcohol, ethyl 29 G per 100 g; 1 drink 120 g
   USDA FNDDS 2021-2023, food code 93301110 (Martini), FDC 2710639 · checked 2026-09-30
3. mtni-r1-03 · dataset
   Martini, fdc_id 2710639: Carbohydrate, by difference 0.15 G, Alcohol, ethyl 29 G, Energy 201 KCAL per 100 g; Martini, flavored, fdc_id 2710640: Carbohydrate, by difference 6.36 G, Alcohol, ethyl 23.6 G, Energy 187 KCAL per 100 g
   USDA FNDDS 2021-2023, FDC 2710639 and 2710640 · checked 2026-09-30
4. mtni-r1-13 · agency guidance
   In the United States, one standard drink contains about 14 grams, or about 0.6 fluid ounces, of pure alcohol.
   NIAAA, What Is A Standard Drink? · checked 2026-09-30
5. mtni-r1-04 · randomized controlled trial
   The pBAC of each group was different (P < .001) from all other groups (NF = 0.064 ± 0.003, SSM = 0.047 ± 0.002, RABB = 0.031 ± 0.002, MCM = 0.020 ± 0.002%; mean ± standard error of the mean).
   J Med Food, 2020 · checked 2026-09-30
6. mtni-r1-05 · randomized controlled trial
   Both OFBAR and the control meal reduced alcohol absorption by approximately 30% compared with fasting, as assessed by relative bioavailability based on area under curve (AUC0-t).
   An Acad Bras Cienc, 2026 · checked 2026-09-30
7. mtni-r1-06 · randomized controlled trial
   The peak blood-alcohol concentrations (BAC) were 16.6 +/- 4.0, 17.7 +/- 7.1, and 13.3 +/- 4.0 mg dl-1 (mean +/- s.d.) after fat, CHO, and protein-rich meals and 30.8 +/- 4.3 and 54.3 +/- 6.4 mg dl-1 after fasting and i.v. infusion, respectively.
   Br J Clin Pharmacol, 1997 · checked 2026-09-30
8. mtni-r1-07 · randomized controlled trial
   Alcoholic beverages (beer, red wine, whisky) caused a significant slowdown of the gastric evacuation of the solid meal, the delay being the more potent, the greater was the concentration of ethanol.
   J Gastroenterol, 2013 · checked 2026-09-30
9. mtni-r1-09 · randomized controlled trial
   Food consumption was significantly greater, and levels of ghrelin were reduced, following alcohol.
   Obesity (Silver Spring), 2015 · checked 2026-09-30
10. mtni-r1-12 · randomized controlled trial
    Maximum blood alcohol concentrations, area under the blood alcohol concentration--time curve, and time to maximum concentration were not significantly different during ranitidine coadministration compared with coadministration of placebo.
    Clin Pharmacol Ther, 1994 · checked 2026-09-30
11. mtni-r1-15 · case series
    Three patients ate live crabs at sushi bars (including crabs in martinis, a previously unreported mechanism for infection).
    Am J Surg Pathol, 2011 · checked 2026-09-30
12. mtni-r1-08 · randomized controlled trial
    Gastric half emptying time (t 1/2) of the meal with water was 123 +/- 5 min, while with brandy (119 +/- 9 min), herb flavored liqueur (123 +/- 10 min), aquavit (125 +/- 9 min), Williams pear brandy (126 +/- 6 min) or espresso (125 +/- 9 min) t(1/2) it was not significantly different.
    J Gastrointestin Liver Dis, 2008 · checked 2026-09-30
13. mtni-r1-10 · controlled clinical trial
    With the same amount of ethanol, peak concentration and area under the curve tended to be greater in women than in men.
    Alcohol Alcohol, 1988 · checked 2026-09-30
14. mtni-r1-11 · clinical experiment
    These findings add to the growing literature that lactating women metabolize alcohol differently, in part, due to the frequent breast stimulation during breastfeeding and the pronounced physiological changes that accompany one of the most energetically costly mammalian activities.
    Alcohol, 2010 · checked 2026-09-30
15. mtni-r1-14 · agency guidance
    Are pregnant or might be pregnant.
    CDC, About Moderate Alcohol Use · checked 2026-09-30