The AHA Data Task Force has been given access to NHC entry data from 2023 through the present, and we’re working toward a presentation at Homebrew Con 2026 titled “You’re Gonna Need a Bigger Spreadsheet: Analysis of NHC Data 2023 to Today”. Before we get too deep into the numbers, we want to hear from you.
As a small taste of what this data holds: the share of NHC entrants affiliated with a homebrew club has grown from 59.5% in 2023 to 83.5% in 2026. There’s a lot more where that came from.
But here’s where you come in. We need you to answer both of these:
What would you like to know from the data?
Why do you want to know it?
Both questions matter. The first tells us what to look at; the second tells us what you’re actually trying to understand or accomplish. Your answers will shape our analysis, the HBC presentation, and a potential future Zymurgy article. We’re looking forward to hearing what’s on your minds!
@freddthecat Thanks for laying those out so clearly, really useful framing for us.
On your first one, top styles by medals as a percentage of submissions, yes, we can absolutely do that. It’s one of the more revealing cuts of this data because raw medal counts are misleading on their own. A style that wins 15 medals but had 200 entries is a very different story from one that wins 8 medals out of 20 entries. Percentage of submissions gets at what you’re actually asking: where do your odds look best?
On the second and third, I want to be straight with you, judges’ comments are not in the dataset we have access to right now, so we can’t pull the top flaws or most common positive descriptors from scoresheets. I’m going to reach out to see if that data is something we can get our hands on. Fair warning though, even if we can get it, that kind of text analysis is a heavier lift than the numerical stuff, so no promises on timeline.
@cbeardsl That might be something for the AHA recipe data set analysis. Still worth doing in the long run but a different set of data than the BAP NHC data we’re working with.
Need to go back farther than that. Need to show Pre-Covid when entries were only $15 versus what they cost now. Also need to correlate low site numbers with the insane amount of first round sites. I would also say next time make this a presentation for the main stage instead of having it in a meeting room with a set number of chairs and also taking a speaking slot from someone that otherwise could have given an actual educational presentation.
As for the presentation, we’re pretty confident there’s a rich and useful story in this data for anyone interested in competitive homebrewing. Hope you’ll come check it out.
re: comments, sorting them and presenting them.
I actually heard someone explaining developments in “fuzzy” searching using AI the other day. In fact it’s doable to you now depending on what AI you have access to - even the right free online one could work pretty well.
You want it to be able to search for concepts rather than exact words a la a search engine. ie. its searching for words in the ballpark of idk - “tropical aroma” and would include anything related to pineapple, coconut, mango, passion fruit, various other strength values for related terms like fresh, juicy, sweet, ripe and other layers of strength for other supporting things and assigning negative values for terms that oppose the concept of tropical aroma.
the presentation of this data could be in a bunch of ways, but yeah - it would be a neat project.
as a start are the judges’ comments in handwritten or digital format?
@freddthecat They are typed into the software, so that’s encouraging for future analysis.
For this presentation though I’m keeping my focus on the data I actually have in hand right now, which includes entry identifiers, participant club affiliation, collaboration status, base style, special ingredients, ABV, IBU, category, scores, and award results. There’s plenty to work with there and only about a month to dig into it before Homebrew Con. The comments layer is a great conversation we can pick back up on the other side of that.
A trend analysis of the National Homebrew Competition (NHC) final round winners. Two main purposes would be to identify the BJCP styles which placed (1st, 2nd, 3rd) the most and to extrapolate which styles have the best potential for placing in the next competition.
Number of entries per style category, year over year. What I want to learn: how are style interests evolving over time? (Although the time window is short but still interesting.) And what style categories are crowded (hard to compete) vs. underrepresented (maybe a better shot at medalling)?
And another one: what are typical winning scores (per category)? What I want to know: when I score a 43 on something, is that about as good as it gets, or do I really need to be shooting for a 48?
@dougmandic The style category thing is definitely in the analysis and will be in the HBC presentation.
I’ll look into scoring details but scoring is so variable I’m not 100% sure what it might tell you other than judges feel differently about what a good beer should score. I might still include it just to show what the data says. I would rather analysis and present data that is inconclusive rather than just assume it is.
A quick update on the HBC presentation. If you didn’t see at the top of the forum page. I’m presenting “Analysis of NHC Data: 2023 to Today” on June 20th at 2 PM. If you’ve ever wondered why your score landed where it did, which styles are quietly climbing the rankings, or how to make smarter decisions about what you brew and enter, this one’s for you.
We’re talking 15,092 entries, 248,000 data points, and four years of NHC history distilled into actual insights you can use. View the full schedule here: https://bit.ly/hbc26
