There is a myth saying that having a BAC in the 0.129% - 0.138% range can improve your cognitive abilities. This effect is called the Ballmer Peak (a reference to Steve Ballmer, CEO of Microsoft) and is pictured nicely in this xkcd. Is there any truth in this myth?

  • 6
    There may not be any studies, but championship darts players will sip on pints of lager during competition, and I absolutely think I am better at pool after a couple of beers, though that may just be the beer talking. It's not inconceivable that alcohol could help one relax and/or think in a slightly different manner, and that this may be of benefit to some very specific activities. However, until someone does some specific alcohol-and-programming studies, your question is unanswerable.
    – John Lyon
    Commented May 4, 2011 at 5:06
  • 17
    @jozzas: I have a friend who swears he is a master with women after a few drinks. I don't have the heart to tell him how pathetic he gets when he's hammered... The worst part is that he honestly believes he drives better after drinking, too. Commented May 4, 2011 at 5:45
  • 28
    @Blue: I hope you have the heart to give him a heads up on the latter, at least Commented May 4, 2011 at 6:13
  • 35
    The Ballmer peak is exactly at 0.1337% (as stated by Randall Munroe in a talk at Google). Another subtle hint that it's a joke :) Commented May 5, 2011 at 8:22
  • 1
    I know of at least one example where alcohol DOES improve one's driving. A friend of mine is a reckless driver when sober, but becomes careful after he's been drinking. :)
    – user13359
    Commented Apr 12, 2013 at 16:47

4 Answers 4


This article by Norlander specifically studies the relationship between moderate alcohol consumption (1.0ml/kg body weight) and creativity. According to my very rough calculations, this would correspond to a BAC in the range of 0.12–0.14 for a 73kg human. The paper concludes

...modest alcohol consumption inhibits aspects of creativity based mainly on the secondary process (preparation, certain parts of illumination, and verification), and disinhibits those based mainly on the primary process (incubation, certain parts of illumination, and restitution).

In other words, moderate alcohol consumption does improve certain types of creative thinking, while inhibiting other types of creative thinking. Since the skills required for computer programming are solely cognitive in nature (discounting the motor skills required to type, of course), and given that creativity is a large part of computer programming, it is at least plausible that one might gain some amount of improvement from alcohol consumption.

There have also been studies on the relationship between alcohol consumption and creative output. That study examined 34 well known, heavy drinking, 20th century writers, artists, and composers/performers. It concludes:

Analysis of this information yielded a number of interesting findings. Alcohol use proved detrimental to productivity in over 75% of the sample, especially in the latter phases of their drinking careers. However, it appeared to provide direct benefit for about 9% of the sample, indirect benefit for 50% and no appreciable effect for 40% at different times in their lives. Creative activity, conversely, can also affect drinking behavior, leading, for instance, to increased alcohol consumption in over 30% of the sample. Because of the complexities of this relationship, no simplistic conclusions are possible.

So for a small portion of people there was a notable increase in creative output as a result of alcohol intake. It does appear that the study did not control for the quantity of alcohol intake, though, so this may not be directly applicable to the Ballmer Peak.

The best study I was able to find on the subject was by Lapp, Collins, and Izzo. They gave subjects vodka tonics of varying strengths (by varying the ratio of tonic to vodka), some of which did not even contain any alcohol. The subjects believed that they were drinking a standard-strength vodka tonic. The subjects then were asked to perform a number of cognitively and creatively challenging tasks. Here is what they conclude:

The present results support the idea that creative people probably gain inspriation from consuming alcohol ..., but show that this effect may be due to the expected rather than the pharmacological effects of the drug. ... A convergence of evidence supported the idea that creativity is enhanced (at least in some aspects) by the expected effects of alcohol.

In other words, alcohol can improve certain aspects of one's cognitive ability, but this effect is not likely due to any pharmacological process (i.e., it is often sufficient to merely believe that one is drinking alcohol in order to achieve the same benefit).

And remember: The Ballmer Peak, as it is currently understood, is but a two dimensional projection of what in reality is a higher dimensional space, vi&.,

3D Ballmer Peak


  • 40
    +1 for esoteric joke no-one will get. -1 for original research. :-)
    – Oddthinking
    Commented Jun 7, 2011 at 2:04
  • 1
    moderate alcohol consumption (1.0ml/kg body weight) There is no time given. Per day? That would be 1.5l beer of 5Vol.% Alc, or 0.75l Wine of 10% for a person of 75kg weight, or 200 ml Whiskey of 40%. On a daily basis, this isn't so moderate any more. Commented Jul 27, 2011 at 2:47
  • 7
    i wonder if i can extrapolate the plot, as it currently only spans the range of esoteric languages on the safety-axis. i wonder how huge the peak becomes once you reach real programming languages… Commented Nov 18, 2011 at 10:04
  • 9
    @Oddthinking Unless they google Malbolge, of course. Commented Nov 20, 2011 at 9:21
  • 1
    @muntoo Currently working on some experimental code, sipping a whiskey and coke. I was feeling no effects of the booze until reading that article. Now I think I'm hammered. Commented Dec 16, 2012 at 6:29

New study at the University of Illinois at Chicago reported by Medical daily: Drinking Alcohol May Significantly Enhance Problem Solving Skills

Scientists found that men who drank two pints of beer or two glasses of wine before solving brain teasers were quicker in delivering correct answers.

Here is the related scientific publication


  • We examine the effects of alcohol intoxication on creative problem solving.
  • Sober and intoxicated (BAC = .075) individuals solved Remote Associates Test items.
  • Intoxicated individuals solved more items in a shorter time compared to sober.
  • Intoxicated individuals were more likely to rate their solutions as insightful.

Uncorking the muse: Alcohol intoxication facilitates creative problem solving by Andrew F. Jarosz, , Gregory J.H. Colflesh , Jennifer Wiley

  • 1
    A lot of programming work can be thought of as a series of interrelated problem-solving exercises. Commented Oct 31, 2013 at 14:23
  • 1
    As described, I find these results dubious. I can accept the results that after drinking they were quicker in delivering correct answers. I'd also expect them to be quicker in delivering wrong answers. What's dubious is that there is no mention of whether or not the quicker answers overall were more likely to be correct or not. If you give 7 solutions in 10 minutes while sober with 71% of them correct and 18 solutions in 9 minutes when intoxicated with 50% correct, then you technically solve more problems correctly in a shorter time, but I wouldn't say it improved creativity, just impulsivity.
    – Jim
    Commented Jul 7, 2017 at 13:33

If we think about driving a car, it is generally agreed that there is no alcohol level at which one is a better driver than when one is sober. This document lists some of the findings:

Behavioural studies suggest that driving related skills are significantly impaired at blood alcohol concentrations below 10-9 mmol/l and that little evidence exists for a threshold below which driving related skills are unimpaired.

(Alcohol does of course impair judgement, so it is likely that there is an alcohol level at which one believes one is a better driver, even though one isn't.)

Computer programming is a significantly more demanding task than driving a car, and it is extremely unlikely that there is any level of alcohol at which one is really a better programmer than when one is sober.

  • 25
    I'm guessing that the primary contributing factor to the impairment of driving skills is reduced reaction time and latency in motor coordination; higher level cognitive impairment is likely a much less contributing factor. If we discount the act of typing on a keyboard, those motor skills and visual impairments won't likely affect a programmer.
    – ESultanik
    Commented May 4, 2011 at 21:25
  • 3
    I agree with @ESultanik. Some of my most brilliant code was conceived and written while I was slightly impaired by lack of sleep, cold medicine, or alcohol.
    – oosterwal
    Commented Jul 23, 2011 at 23:41
  • 2
    @oosterwal Also, I recall reading an article about how sleepiness does not affect a person's ability to do math (think) - only the person's willingness to do math. And if you code while sleepy, you are obviously somewhat willing to code. Commented Nov 20, 2011 at 9:27
  • 7
    This is just a bad analogy. -1
    – TsSkTo
    Commented Feb 18, 2017 at 23:06
  • 1
    The consequences of mistakes while driving are usually faced rather immediately. With programming, there is always a chance to take a program that doesn't compile and go back over it. Furthermore, the claim is more along the lines that you try solutions in programming you'd not have considered when sober. This means one can implement a broad solution, which may have plenty of minor errors due to intoxication, but be able to fix the errors when sober again while maintaining the novel overall solution you otherwise wouldn't have
    – Jim
    Commented Jul 7, 2017 at 13:37

The Observer reports on a study showing it is real.

A recent study at the University of Illinois tested the creative problem solving ability of a group of men who were given vodka cranberry and snacks and asked to solve brain teasers. The results were starkly different for the tispy group, which had a blood alcohol concentration level of 0.075, versus the control group:

Astonishingly, those in the drinking group averaged nine correct questions to the six answers correct by the non-drinking group. It also took drunk men 11.5 seconds to answer a question, whereas non-drunk men needed 15.2 seconds to think. Both groups had comparable results on a similar exam before the alcohol consumption began.

  • 5
    Welcome to Skeptics! This is a summary of an Observer summary of a CBS News summary of a paper. It would be better to read the paper itself. Also, it isn't clear whether solving brain teasers is directly applicable to programming - particularly the short term memory aspects.
    – Oddthinking
    Commented Aug 23, 2016 at 16:59

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .