Yes, this graph made things far clearer, at least for me. (I think it could have been improved even more, but still, this was helpful.)
It's instantly clear that most of the absolute volume drop was in the first couple of years. You also instantly see the crash of '29. And it's quite easy to see that most of the later percentage drop caused by absolute volume remaining steady while "total ale & porter" increases.
For the percentage to make sense you need to graph that against the right-hand axis with a separate scale.
Thicker lines and bigger labels would have helped (or just a bigger graph). It's a bit tricky to see which line is which.
Hi Lars - I asked Ron about 1929 in his previous post and he said that there are only partial records for 1929 - 7 months worth - hence the lower figures for '29. Assuming an average monthly output of (for example) total porter and stout of 12,254 units for 1929, a full years record would be something in the region of 85779 + (5*12254) = 147,049 units. The total porter and stout would actually be greater in 1929 compared with 1928, and slightly less than 1930. So I'm not sure there is any impact of the Wall Street Crash of '29 in the data - and the crash happened in October, so quite late in the year. I took a look at the UK unemployment rate (although perhaps better to look at absolute numbers) against the output of total ale and porter and there appears to be an interesting relationship, particularly in the early thirties, on the assumption that a decline in output of ale and porter (or just beer) may be related to a rise in unemployment, although a daytime research-focused visit to a 'Spoons may provide evidence to the contrary.
I see 8 lines, but 10 legend entries? Still great to get the data as a graph.
Was it the Whitbread data that you have that covers 150 years? How about a plot that shows the OG of their 3 or 4 most important brews over that time span?
@Steve N: Thanks for clearing that up. Guess I was a bit too quick there. The crash happened pretty late in 1929, too. Anyway, the point is you see 1929 standing out immediately.
Not really 🙂
ReplyDeleteYes, this graph made things far clearer, at least for me. (I think it could have been improved even more, but still, this was helpful.)
ReplyDeleteIt's instantly clear that most of the absolute volume drop was in the first couple of years. You also instantly see the crash of '29. And it's quite easy to see that most of the later percentage drop caused by absolute volume remaining steady while "total ale & porter" increases.
For the percentage to make sense you need to graph that against the right-hand axis with a separate scale.
Thicker lines and bigger labels would have helped (or just a bigger graph). It's a bit tricky to see which line is which.
Yes! Of course there's many ways to make this plot yet better readable (physicist talking here) but for such long time series I'm all for plots. :)
ReplyDeleteHi Lars - I asked Ron about 1929 in his previous post and he said that there are only partial records for 1929 - 7 months worth - hence the lower figures for '29. Assuming an average monthly output of (for example) total porter and stout of 12,254 units for 1929, a full years record would be something in the region of 85779 + (5*12254) = 147,049 units. The total porter and stout would actually be greater in 1929 compared with 1928, and slightly less than 1930. So I'm not sure there is any impact of the Wall Street Crash of '29 in the data - and the crash happened in October, so quite late in the year. I took a look at the UK unemployment rate (although perhaps better to look at absolute numbers) against the output of total ale and porter and there appears to be an interesting relationship, particularly in the early thirties, on the assumption that a decline in output of ale and porter (or just beer) may be related to a rise in unemployment, although a daytime research-focused visit to a 'Spoons may provide evidence to the contrary.
ReplyDeleteI see 8 lines, but 10 legend entries? Still great to get the data as a graph.
ReplyDeleteWas it the Whitbread data that you have that covers 150 years? How about a plot that shows the OG of their 3 or 4 most important brews over that time span?
Cheers!
@Steve N: Thanks for clearing that up. Guess I was a bit too quick there. The crash happened pretty late in 1929, too. Anyway, the point is you see 1929 standing out immediately.
ReplyDelete