NBS BIF #17 Hints and Hauls: Subaru Drivers Won't Keep Us From Our Magic Water Breakfast

As we all should

 

I cannot wait for your write up with this review style

 

My roof got shredded the same day Little Rock got hammered. I might need you to take my insurance company to the mat.

 

I am sure I am the only one propping up Arkansas & Okie, for the most part

 

Brief

Brief & profane, most likely

 

It's a problem a lot of us have.

 

I've had two Prairie beers, and even one from American Solera (checked it in on Untapped a while ago, probably should update BA), so I slightly affect OK statistically, but I think you're the only one here who's got a body of work with Arkansas beers.

 

That will change

 

Now I have to figure out if that is a threat, a promise, a threat of a promise, or a promise of a threat...

 

maybe

 

Well, 2190 is the distance between your city and mine, but I'm not sure how the 230 fits in.

Still . . . are you my sender?

 

Have done. Nothing remotely pertinent.

 

Dangit Mike, those were supposed to come to my house!
I mean, Wow, now that is some seriously heavy artillery. Boggles the mind! Mega hit @ovaltine

 

OK, time to geek out
I couldn't compare two curves directly, because BA1 and BA2 may have significantly different quantities of beers. So, I converted the curves to percentage-based for comparison purposes -- "x% of my ratings are in the 4.0-4.2499999 bucket..." sort of thing. Then, for every pair of users in this BIF, I added the sum of the absolute value of the difference between their percentages in each bucket. The lowest sum means there was the least overall difference, the highest sum means the largest overall difference, ignoring total quantity of beers.

I did, unfortunately, reject JayORear, he doesn't have enough beers recorded in BA for the curve to be very meaningful.

Here are the most similar (@WunderLlama and @ovaltine):


Here are the most different (@WunderLlama and @2beerdogs):


Frank, if you're curious, your most similar match is @Roguer, and your least similar is @2beerdogs. I put all three of you on one chart for the sake of brevity:


For the record, getting this all put together was a pretty big pain in the butt...lol

 

Although it was a pain in the butt, I did it in a way that makes it easy to repeat with different data.

I think I might be able to do a comparison of the same person before and after a certain point in time, which might be interesting for someone who hit a point in time where they decided to change their internal scoring scheme.

 

Thought it would be interesting to look at a single user's score distributions based on style group. I used @Roguer as a test subject: