stupac2
Well-Known Member
I've been sitting on this for a little while because the numbers I was getting were different from before, and I couldn't figure out why. Turns out the data wizarded up for me was review-only. I thought for a second and realized that there were worse things that cutting out the hads, so I went with it. The one important thing to note is that this left some of the data fairly scant, so I combined Massachusetts and Vermont into a single state. Turns out most Hill Farmstead beers have 0-2 reviews from Vermont, so that was a way to keep them in the dataset, beers with 1 home review obviously aren't significant (and, it turns out, crash my stats library!).
If you want to see the whole damn thing of 240-some beers, it's right here.
Let's Get Significant
Since this dataset is so huge, I'm started out with only beers where the difference is statistically significant. For my definition of significant I went with P<0.05 because if it's good enough for medical research, it's good enough for this. The sheet of significant beers is here.
The Hype Civil Wars
I created a list off to the side of the states with the most significant beers, and you'll notice that California tops the list by far. But, at the same time, California has the most beers on the list, so I also divided by the total. In this light, it's actually Minnesota that's the worst, with all 6 of their top-250 beers having a statistically significant increase in scoring. The only other 100%-er with more than 1 beer on the list is North Carolina. The other 5 (Connecticut, New Mexico, Oklahoma, Utah, and Washington) only have 1 beer.
Moving past the 100%-ers, we see that Ohio, with 8 of its 9 beers being statistically overrated, is pretty bad. Illinois at 3/5 is too, but that's not all that many beers to be judging this off of, really. Oregon, Michigan, Mass, and Vermont/Mass are all around 40%, which is high but not ridiculous.
Florida is the shock to me on this sheet, with only 3/8 beers being significant. I think that would change with Hads included, since previously every single beer I looked at from Florida was not only significant, but among the most overhyped in the sample. That's not the case here, making Florida's reviewers seem pretty even-keeled, especially compared to their tickers.
Finally, California may have over double the next state in overhyped beers, but it's second-to-last on the sheet, trailing only Germany (Germany!) in hype-resistance. Go team!
In terms of average percentage overrating, most states are in the 2-5% range, with the only ones over 6% being Utah, Texas, and Wisconsin. But none of them have a lot of beers in the sample, so I don't know if we should read into that too much.
IPAs
As with last time, people were interested in this so I'm looking at it. (Just a note, I'm being pretty liberal with the term "IPA" and including Pale Ales, Pale Lagers, etc.) The IPA page is here. The top 10 overrated IPAs with significant differences are:
Citra Ninja- Pipeworks Brewing Company
Winter Ale (2012) / 5th Anniversary Ale- Kern River Brewing Company
Citra Pale Ale- Hill Farmstead Brewery
White Rajah- The Brew Kettle Taproom & Smokehouse / Production Works
Burton Baton- Dogfish Head Brewery
Jai Alai IPA- Cigar City Brewing
Gandhi-Bot Double IPA- New England Brewing Co.
Elevated IPA- La Cumbre Brewing Co.
Kiwi Rising - Double IPL- Jacks Abby Brewing
Alchemy Hour Double IPA- Great Lakes Brewing Company
Nothing on there seems wildly out of whack to me. Except Kiwi Rising, that **** is gold.
Anyway, nothing super interesting is jumping off of this page to me, except PsuedoSue, which is underrated by Iowans by over 4%. It's not significant, but it's interesting. None of the other underrated IPAs are anywhere near that magnitude.
Finally, the average difference in home vs away scores is only 2.34%. Really not all that big, and in line with what I got last time.
HEADY VS PLINY BECAUSE **** YOU THAT'S WHY
The home crowd definitely likes Pliny the Elder better than the away crowd, but the difference is tiny, just less than 1%. The enormous number of reviews makes that small difference significant. Compare this to Heady Topper, because everyone is always comparing these beers and no one ever gets tired of it, where the Vermont/Mass crowd rates it 2.4% higher than everyone else. Clearly, Pliny is the more consistent and therefore better beer. Clearly.
Stouts
Last time stouts were the more controversial style. Here's the data.
As before, here are the top 10 significantly overrated beers:
Fifteen- Central Waters Brewing Company
Chocolate Rain- The Bruery
Murdad Out Stout- Three Floyds Brewing Co. & Brewpub
Czar Jack Imperial Stout- Minneapolis Town Hall Brewery
Black Tuesday- The Bruery
Darkness- Surly Brewing Company
Big Bad Baptist- Epic Brewing Co.
Hunahpus Imperial Stout- Cigar City Brewing
Barrel Aged Sexual Chocolate- Foothills Brewing Company
Uncle Jacobs Stout- Avery Brewing Company
I'm not too surprised by anything on this list. Although you non-CA dudes are all clearly wrong about BT and CR, that **** is awesome.
The bottom of the list here is interesting, too, the following beers are all underrated by 2% or more:
Life Is Like...- Cigar City Brewing
Peche Mortel (Imperial Stout Au Cafe)- Brasserie Dieu Du Ciel
Assassin Imperial Stout- Toppling Goliath Brewing Company
Old Rasputin XV Anniversary Barrel Aged Stout- North Coast Brewing Co.
AleSmith Speedway Stout - Vietnamese Coffee- AleSmith Brewing Company
Of course, none of them are statistically significant, so it could just be by chance. But it's an interesting list to me.
Finally, the average disagreement is now only 2.43%, basically in line with the IPAs. Last time it was almost twice as high, so I guess I just picked controversial beers for that one.
Sours
Sours are funny. There aren't as many, and there are very few statistical differences. The presence of all the foreign ones compounds things in funny ways, too. Anyway, here's the data, and here are the ones with significant differences:
Cascade Sang Noir- Cascade Brewing / Raccoon Lodge & Brewpub
Atrial Rubicite- Jester King Brewery
Peche n Brett- Logsdon Farmhouse Ales
Oude Tart With Sour Cherries- The Bruery
Cuvee De Castleton- Captain Lawrence Brewing Co.
That's not the list I would've guessed, personally. Also kind of interesting to me is that all three 3F beers on the top 250 are in the top 7 biggest differences, just none are significant. I wouldn't have guessed that.
There are also a lot more underrated sours than the other ones (at least compared to the total number of beers), so the average only comes out to 1.4% overrating. Which is pretty small compared to IPAs and Sours. There are a couple of interesting underrated beers, first is Cable Car, which I guess just tastes better outside of the sticky, dirty confines of Toronado. The second one, 1v1willwreku and dontdrinkbeer will be interested to hear, is Fantome Saison, the most underrated sour, with Belgians liking it nearly 5% less than non-Belgians.
For Next Time
Since this data is only reviews, I think next time I'd like to look at only Hads, to see if those are different in any way. I'm out of things to examine with this dataset, but if you think of something, let me know!
If you want to see the whole damn thing of 240-some beers, it's right here.
Let's Get Significant
Since this dataset is so huge, I'm started out with only beers where the difference is statistically significant. For my definition of significant I went with P<0.05 because if it's good enough for medical research, it's good enough for this. The sheet of significant beers is here.
The Hype Civil Wars
I created a list off to the side of the states with the most significant beers, and you'll notice that California tops the list by far. But, at the same time, California has the most beers on the list, so I also divided by the total. In this light, it's actually Minnesota that's the worst, with all 6 of their top-250 beers having a statistically significant increase in scoring. The only other 100%-er with more than 1 beer on the list is North Carolina. The other 5 (Connecticut, New Mexico, Oklahoma, Utah, and Washington) only have 1 beer.
Moving past the 100%-ers, we see that Ohio, with 8 of its 9 beers being statistically overrated, is pretty bad. Illinois at 3/5 is too, but that's not all that many beers to be judging this off of, really. Oregon, Michigan, Mass, and Vermont/Mass are all around 40%, which is high but not ridiculous.
Florida is the shock to me on this sheet, with only 3/8 beers being significant. I think that would change with Hads included, since previously every single beer I looked at from Florida was not only significant, but among the most overhyped in the sample. That's not the case here, making Florida's reviewers seem pretty even-keeled, especially compared to their tickers.
Finally, California may have over double the next state in overhyped beers, but it's second-to-last on the sheet, trailing only Germany (Germany!) in hype-resistance. Go team!
In terms of average percentage overrating, most states are in the 2-5% range, with the only ones over 6% being Utah, Texas, and Wisconsin. But none of them have a lot of beers in the sample, so I don't know if we should read into that too much.
IPAs
As with last time, people were interested in this so I'm looking at it. (Just a note, I'm being pretty liberal with the term "IPA" and including Pale Ales, Pale Lagers, etc.) The IPA page is here. The top 10 overrated IPAs with significant differences are:
Citra Ninja- Pipeworks Brewing Company
Winter Ale (2012) / 5th Anniversary Ale- Kern River Brewing Company
Citra Pale Ale- Hill Farmstead Brewery
White Rajah- The Brew Kettle Taproom & Smokehouse / Production Works
Burton Baton- Dogfish Head Brewery
Jai Alai IPA- Cigar City Brewing
Gandhi-Bot Double IPA- New England Brewing Co.
Elevated IPA- La Cumbre Brewing Co.
Kiwi Rising - Double IPL- Jacks Abby Brewing
Alchemy Hour Double IPA- Great Lakes Brewing Company
Nothing on there seems wildly out of whack to me. Except Kiwi Rising, that **** is gold.
Anyway, nothing super interesting is jumping off of this page to me, except PsuedoSue, which is underrated by Iowans by over 4%. It's not significant, but it's interesting. None of the other underrated IPAs are anywhere near that magnitude.
Finally, the average difference in home vs away scores is only 2.34%. Really not all that big, and in line with what I got last time.
HEADY VS PLINY BECAUSE **** YOU THAT'S WHY
The home crowd definitely likes Pliny the Elder better than the away crowd, but the difference is tiny, just less than 1%. The enormous number of reviews makes that small difference significant. Compare this to Heady Topper, because everyone is always comparing these beers and no one ever gets tired of it, where the Vermont/Mass crowd rates it 2.4% higher than everyone else. Clearly, Pliny is the more consistent and therefore better beer. Clearly.
Stouts
Last time stouts were the more controversial style. Here's the data.
As before, here are the top 10 significantly overrated beers:
Fifteen- Central Waters Brewing Company
Chocolate Rain- The Bruery
Murdad Out Stout- Three Floyds Brewing Co. & Brewpub
Czar Jack Imperial Stout- Minneapolis Town Hall Brewery
Black Tuesday- The Bruery
Darkness- Surly Brewing Company
Big Bad Baptist- Epic Brewing Co.
Hunahpus Imperial Stout- Cigar City Brewing
Barrel Aged Sexual Chocolate- Foothills Brewing Company
Uncle Jacobs Stout- Avery Brewing Company
I'm not too surprised by anything on this list. Although you non-CA dudes are all clearly wrong about BT and CR, that **** is awesome.
The bottom of the list here is interesting, too, the following beers are all underrated by 2% or more:
Life Is Like...- Cigar City Brewing
Peche Mortel (Imperial Stout Au Cafe)- Brasserie Dieu Du Ciel
Assassin Imperial Stout- Toppling Goliath Brewing Company
Old Rasputin XV Anniversary Barrel Aged Stout- North Coast Brewing Co.
AleSmith Speedway Stout - Vietnamese Coffee- AleSmith Brewing Company
Of course, none of them are statistically significant, so it could just be by chance. But it's an interesting list to me.
Finally, the average disagreement is now only 2.43%, basically in line with the IPAs. Last time it was almost twice as high, so I guess I just picked controversial beers for that one.
Sours
Sours are funny. There aren't as many, and there are very few statistical differences. The presence of all the foreign ones compounds things in funny ways, too. Anyway, here's the data, and here are the ones with significant differences:
Cascade Sang Noir- Cascade Brewing / Raccoon Lodge & Brewpub
Atrial Rubicite- Jester King Brewery
Peche n Brett- Logsdon Farmhouse Ales
Oude Tart With Sour Cherries- The Bruery
Cuvee De Castleton- Captain Lawrence Brewing Co.
That's not the list I would've guessed, personally. Also kind of interesting to me is that all three 3F beers on the top 250 are in the top 7 biggest differences, just none are significant. I wouldn't have guessed that.
There are also a lot more underrated sours than the other ones (at least compared to the total number of beers), so the average only comes out to 1.4% overrating. Which is pretty small compared to IPAs and Sours. There are a couple of interesting underrated beers, first is Cable Car, which I guess just tastes better outside of the sticky, dirty confines of Toronado. The second one, 1v1willwreku and dontdrinkbeer will be interested to hear, is Fantome Saison, the most underrated sour, with Belgians liking it nearly 5% less than non-Belgians.
For Next Time
Since this data is only reviews, I think next time I'd like to look at only Hads, to see if those are different in any way. I'm out of things to examine with this dataset, but if you think of something, let me know!
Last edited by a moderator: