FatPhil's Brobdingnagian Challenge

What?

The plan was to reach 12 countries from which I've had 12 different beers by 12 different brewers.

Why?

Some people want to have at least one beer from every country in the world. Others want to have all the beers from their country/state/region. Those ticker/trainspotter targets didn't interest me so much.

Instead I wanted to take a more balanced approach - a large number of beers from a large number of breweries from a wide range of countries. I wanted to be able to say not just that I knew certain breweries well (by having many beers from from them) but also that I knew a country's beer scene very well (by having a large number of breweries that I knew well). I also wanted to extend this concept to the world - I wanted to be able to say that I knew the beer scene in a lot of different countries.

Where Did The Idea Come From?

The "X countries with X breweries with X beers" idea is based on the "h index" in academia. If you've written a hundred papers that have only been cited once each, you're not having much influence. If you've written one paper that has been extremely widely cited, then you're not very productive. The h-index balances these two extremes - what's the highest X such that you've published X papers that have been cited at least X times. If you've written 10 papers that have been cited 10 times each, then you're reasonably productive, and reasonably influential, better than either of the extremes.

It's an easy calculation, list your pubs in citation count order, and then just count down the list, as soon as your list position exceeds the citation count, you've gone too far. E.g., with 100 × 1 cite, you can't count to 2, and with 1 × 100 cites, you can't either. With 10 × 10 cites, you can't count to 11. So the h-indices are 1, 1, and 10.

My idea is just a 3-dimensional equivalent.

When Did I Come Up With This Idea?

Some time around 2015, I guess. Some people on the RateBeer fora were comparing their beer-per-country h-index. I think at the time, my h-index by this metric was 29, it's now 32, progress is slow.

It struck me that not all countries are equal. Some countries don't have much going on apart from a few brewers that just churn out a new beer every couple of days, all much of a muchness, and you can have had literally hundreds of beers from that country and still only have experienced a narrow range of beers.

Addressing this issue, someone, perhaps me, proposed the idea of counting not the number of beers from each country but the number of different breweries instead. I have no idea what my score was back then, but it's now 23.

This can be hacked by someone just trying one thing from each brewery, which again doesn't imply any real knowledge of that brewery's oeuvre.

So it needed fixing. It had to be made 3-dimensional: Countries × Breweries × Beers. I did a quick scan down my beer ratings, and noticed that I had already rated 11 beers from each of 11 different breweries in 11 different countries. The next target was clear: 12 × 12 × 12!

Why The Name?

The concept's from Jonathan Swift's Gulliver's Travels. The giants on Brobdingnag were twelve times larger in each dimension, so 12×12×12=1728 times larger in total.

Why Did It Take So Long?

Because it's damn hard. You don't just need another new beer from another new brewery from another country. 11×11×11 has a "volume" of 1331, so 1728 is ~30% more work. Add to that the fact that just finding countries with lively beer scenes gets harder and harder - even some of the most well-known beer countries, such as Germany and Czechia, are full of breweries that only make 3-5 different beers - finding 12 beers from most of those is practically impossible. So you can probably double that work factor.

I'd already been rating beers for well over 10 years, so it's not weird that it would take another 5+ years. That, and in 2019 I stopped rating on RateBeer, and started to build up a backlog on paper, and on my phone. This meant that there might be stealth breweries that would contribute to my total, but I didn't yet realise it. As the target got closer, I specifically hunted out ratings in my backlog that were most likely to assist me in making progress towards my goal. There are still 2-3 thousand in the backlog, who knows what's within...

The nice thing is that I managed to hit this target before my 50th birthday. 3 days before it, to be precise. I entered the final ratings on Wednesday 10th Feb 2021, I grabbed a backup of my ratings on Thursday 11th, and ran my checker scripts then, and finally I put this webpage together on Friday 12th.

And the cock's just crowed - just in time!

Prove You've Done It!

These are the numbers of breweries that I've had at least 12 beers from in each of my top countries - and all these are now entered on RateBeer:

$ ./rbsum.pl --template='$country:$brewer' < 51287.raw | sort | uniq -c | perl -ne 'print("$3\n") if(m/^\s+(\d+)\s+(.*?, )?(.*?):/ and $1>=12)' | sort | uniq -c | sort -rn | cat -n | head -14
1       82 England
2       52 Finland
3       28 Belgium
4       26 United States
5       26 Latvia
6       26 Estonia
7       19 Denmark
8       17 Sweden
9       16 Lithuania
10      14 Poland
11      13 Russia
12      13 Netherlands
13      12 Germany
14     8 Italy

So I've actually done 13 countries at 12 breweries with 12 beers. Or 12 countries at 13 breweries with 12 beers. But I did just do a specific push to get Polish, Russian, Dutch, and German beers from my backlog, so it's not surprising more went over the line together than strictly needed.

What's Next?

Obviously 13×13×13. When I restrict the above search to breweries that I've had 13 beers from, rather than 12, the result is as follows:

$ ./rbsum.pl --template='$country:$brewer' < 51287.raw | sort | uniq -c | perl -ne 'print("$3\n") if(m/^\s+(\d+)\s+(.*?, )?(.*?):/ and $1>=13)' | sort | uniq -c | sort -rn | cat -n | head -13
1      75 England
2      50 Finland
3      27 Belgium
4      26 Estonia
5      23 United States
6      22 Latvia
7      17 Denmark
8      15 Lithuania
9      14 Sweden
10     14 Poland
11     13 Russia
12     12 Netherlands
13     11 Germany

So I have 11 countries at 13 breweries with 13 beers, both the Netherlands and Germany need a push. I reckon this might be possible this year, if I can travel. However, the next target after that looks mindblowingly hard.

Are You The First?

Almost certainly not, I'm sure the really big raters (with 30, 40, 50+ thousand ratings, compared to my ~24000 (of which ~2500 in the backlog)) got there ages ago, and have done it to the 13th, 14th, perhaps even 15th level.

I have run the stats for a few friends, some of whom are still rating, while others aren't, and they are as follows:

Marduk (retiring)

$ ./rbsum.pl --template='$country:$brewer' < 136546.raw | sort | uniq -c | perl -ne 'print("$3\n") if(m/^\s+(\d+)\s+(.*?, )?(.*?):/ and $1>=11)' | sort | uniq -c | sort -rn | head -12 | cat -n
1	38 Estonia
2	25 Finland
3	25 England
4	24 Latvia
5	23 United States
6	20 Belgium
7	15 Lithuania
8	14 Russia
9	14 Poland
10	13 Sweden
11	11 Netherlands
12      10 Germany

So 11×11×11 already reached, and only Germany and the Netherlands needed for 12×12×12.

martjoobolut (retired)

$ ./rbsum.pl --template='$country:$brewer' < 286327.raw | sort | uniq -c | perl -ne 'print("$3\n") if(m/^\s+(\d+)\s+(.*?, )?(.*?):/ and $1>=9)' | sort | uniq -c | sort -rn | head -11 | cat -n
1       34 Estonia
2       32 United States
3       32 Belgium
4       22 England
5       14 Netherlands
6       12 Germany
7       11 Sweden
8        9 Russia
9        9 Poland
10       8 Italy
11       7 Latvia

So 9×9×9 already reached, however, only 7 countries are at 10×10 level, so not that close.

Tbone

$ ./rbsum.pl --template='$country:$brewer' < 10233.raw | sort | uniq -c | perl -ne 'print("$3\n") if(m/^\s+(\d+)\s+(.*?, )?(.*?):/ and $1>=10)' | sort | uniq -c | sort -rn | head -11 | cat -n
1       78 Finland
2       73 United States
3       51 Belgium
4       50 England
5       46 Sweden
6       38 Denmark
7       27 Germany
8       26 Norway
9       25 Netherlands
10      18 Estonia
11       7 Scotland

Surprisingly (Tbone has more ratings than me, and way more than Marduk), only the 10×10×10 level has been reached so far.


Appendices

My Beers/Country H-Index

32, but 33 is in my sights:

$ ./rbsum.pl --template='$country' < 51287.raw | sed -e 's/.*, //' | sort | uniq -c | sort -rn | cat -n | head -33
1     4499 England
2     2578 Finland
3     1667 Estonia
4     1262 Belgium
5     1220 Germany
6     1219 United States
7     1170 Latvia
8     1005 Denmark
9      776 Poland
10      689 Netherlands
11      668 Lithuania
12      616 Sweden
13      519 Scotland
14      450 Czech Republic
15      429 Russia
16      397 Italy
17      284 Norway
18      262 Spain
19      241 France
20      176 Wales
21      159 Japan
22       94 Austria
23       83 Belarus
24       68 Iceland
25       63 Hungary
26       60 Slovakia
27       57 Canada
28       54 Ireland
29       50 Switzerland
30       47 Australia
31       45 New Zealand
32       41 Ukraine
33       29 Greece

My Breweries/Country H-Index

23, I can't quite sneak a 24th:

$ ./rbsum.pl --template='$country:$brewer' < 51287.raw | sort | uniq | sed -e 's/.*, //' -e 's/:.*//' | sort | uniq -c | sort -rn | cat -n | head -24
1      892 England
2      389 Germany
3      300 United States
4      208 Belgium
5      140 Czech Republic
6      133 Finland
7      123 Poland
8      105 Estonia
9       99 Netherlands
10       92 Latvia
11       86 Italy
12       82 France
13       77 Denmark
14       76 Spain
15       72 Sweden
16       69 Lithuania
17       61 Russia
18       56 Scotland
19       50 Wales
20       39 Japan
21       34 Austria
22       31 Slovakia
23       25 Norway
24       23 Hungary


Another hastily constructed page by Phil Carmody
Home / Beer / brobdingnag.html