Time to extrapolate

I really hated to do it, but I had two numbers in my tracking spreadsheet: one was actual and one was absolute.

Sorry, I'll back up a bit. I use (obviously) Google Blogger (formerly Blogspot, I think) to publish this blog. Google has made blogging very simple... so simple, even a caveman could do it. After all, even I can do it! However, being simple has its drawbacks. As an example, there's this blog. Although it mostly is not about data and data analysis, that is my tagline and the underlying theme. And apart from a few recent blogs, data and data analysis have pretty much been a feature of most of my blogs.

However, simplicity has further trade-offs, one of which is accuracy. When I first started this blog, I didn't have a vision of what it would be, what message it would communicate, or even if it was important to me- this is quite evident in the sporadic posting of the first several months. I'm still not 100% certain about the vision... today it looks like life and data. The message, though... the message is my voice. My dream.

Ever since I was a boy, a young boy, I had a dream. I had lots of dreams. I still have them. One of them was to join the military, which I did. Another was to visit Germany, which I did. Another was to write a book, which I have not done- and may never do. I can't say how many books- stories- I've started.  Some were good by my standards, and held promise. Others might have been no more than treatments, basic sketches that required fleshing out. Some were a waste of paper, or storage space. After all of this, the blog has become my preferred vehicle for creative writing. In other words, I've become a content creator.

So, here I sit creating web content, hoping that anonymous internet denizens will find this content via tags, labels and catchy titles and consume it... maybe even pass it along to other internet content consumers. Here's the problem I face... well, I regard it as a problem. I'm all about numbers when it comes to certain things: Legos, blog views, etc. Google is really good at giving me realtime stats- I absolutely love this. However, they only give Top 10 historical stats, which truly stinks. So, Monday night I made a somewhat painful decision: I extrapolated data. This brings out the Dr. Sheldon Cooper in me, wanting to do unmentionable things to Google data scientists.

I extrapolated. I took the data which I had manually copied into a spreadsheet and essentially created a new reality for my data, in which numbers are essentially correct, but variances are quite possibly not. In my data, I had 1261 confirmed hits.  Google reported 2566 hits, so I had a roughly 2:1 spread:  I took the known good numbers as a percentage, and multilplied them against the absolute number to get an adjusted number.

And here I sit. Going forward, I will pay closer attention to the numbers.

As always, I am hochspeyer, blogging data analysis and management so you don't have to,