View Single Post
      03-14-2018, 02:07 PM   #15
bimmer456
Major General
3008
Rep
6,069
Posts

Drives: 340i
Join Date: Nov 2016
Location: Pasadena, CA

iTrader: (0)

Quote:
Originally Posted by DBACE View Post
OP, The bar is showing actual standard deviations (SD). Light Blue if you are <1 SD, Dark Blue if you are <2 (SD), Red if you are <3 (SD). Obviously if the bar is full with all three colors than you get the >3 standard deviations note.

In statistics, the standard deviation (SD, also represented by the Greek letter sigma σ or the Latin letter s) is a measure that is used to quantify the amount of variation or dispersion of a set of data values.

Think of a bell curve used on a chart to represent data, the target is the peak of the bell. In some cases, like a high volume production line, you want to hit the peak or target every time. Anything higher or lower means a defect in your product. "All variance is evil". Companies would react to the quality indicator and take action to get back on target by changing out wearable items in the production line equipment, or change a tool, process, etc. So, if your production line of animal crackers is running on a six sigma level (meaning you can go -3/+3 (SD) from target, you would have some absurdly low error rate for every 1 million crackers made.

In Bimmerpost's usage they are taking the whole population's average and setting the peak at the start of the curve to score how many standard deviations away in a positive direction from the average you are. (using a half bell curve, since you can't go negative with rep points).

It would be interesting to see the shape of the Bimmerpost bell, which I presume changes with membership numbers, and I would imagine changes the area value (or percentage) of an actual standard deviation. I think....

Here is an example of a bell curve used with standard deviations, to give you an idea of what I am on about. I hope it helped you along with what others are saying.
Lol is it really that complicated?
Appreciate 0