I've always been told that alcohol will reduce reaction times. There are lots of sources for this online. For example:
https://www.sciencedaily.com/releases/2010/10/101019162149.htm http://www.alcoholismrehab.org/research/alcohol-affects-reaction-time/ http://www.alcoholismrehab.org/research/alcohol-increases-errors-and-reaction-time-during-decision-making-new-study-shows/
And countless others. It seems pretty solid.
However, anecdotally, I've noticed a trend where my performance in activities that benefit from a low reaction time improve after I've had a few drinks (namely online FPS games). I decided to investigate further and setup a small experiment.
Every 20 minutes I would record my BAC level using a breathalyzer, take an online reaction time test (http://www.humanbenchmark.com/tests/reactiontime) five times, and consume 1oz of 80 proof alcohol.
For reference, I also consulted an online chart estimating BAC levels (http://bloodalcoholcalculator.org/bac-charts/ - I'm male and ~160 pounds)
The results I saw are completely contradictory to everything I've ever been told. Over the span of three hours, my reaction time did not increase. In fact, it decreased! Each point in the chart below is 20 minutes apart - and the estimated BAC level from the website. And the test consisted of 10 measurements (and I dropped the 2 lowest, 2 highest and averaged the remaining).
The difference between ~270ms and ~230ms is pretty significant - according to the data collected by the Human Benchmark site I used in my experiment. At 270ms, my reaction time is right in the most common range. But as I drank, I ended up way ahead of the curve.
My question is - how can I explain this result? Everything I find online says alcohol increases reaction time. I've read studies and I can't disagree with them. It seems unlikely that I'm some rare abnormality though....so, what's the deal?