| |
BGonline.org Forums
Perspective on measuring performance (long)
Posted By: scotty In Response To: Suggested Idea for Performance Grading (Rick Janowski)
Date: Sunday, 5 October 2014, at 10:49 p.m.
Great discussion! I've followed it with great interest, even though the outcome of it will probably not affect me in any way, as i am not a tournement player.
A few thoughts on measuring perfomance, or measuring anything for that matter...
We need to find 'indicators' of that we wish to measure. They don't need to be perfect, but we want them to have a strong correlation to that we wish to indicate. Whether we use PR as calculated by a bot, ELO, or include tourney results, or some combination of all three, we can find a useful indication for skill level.
As an example on measurement, from my work as a carpenter....when I was an apprentice, I was embarrassed that I had erected a wall that was slighty out of plumb (not perfectly vertical). After that experience, I decided to check the bubbles in my spirit levels, both of them of very high quality, and I found that the bubbles were not as consistent as I had previously thought them to be. I knew I had a problem. How could I correct these bubbles? Did I need to test my levels every day, before using them? How should I test them, to have confidence in them? Should I discard these expensive instruments and buy new ones? Could I build a level for myself that had adjustable bubbles so that I could calibrate my levels from time to time? If so, how would I would I know when the bubbles were accurately centered? It is known that the body of the level changes shape with time and changes in temperature. How do I deal with this?
After much thought, it was realized that a turntable could be constructed, with the level on top, such that the bubble would remain centered while the turntable was rotated, if and only if the turntable was perfectly level and the vials perfectly set in the instrument. Then came the aha moment!
I don't need a level with perfectly set vials. That setting will change daily with changes in temperature, and with any change in the shape of the body of the level. I simply need to rotate my level every time it is used and see the bubble in the same place in the vial to know that I am level or plumb. I don't need to see the bubble centered in the vial! I only need to see that the position of the bubble is unchanged within the vial to have confidence that a surface is true.
I'm retired now. I never met anyone who was familiar with this principle. Not my instructers in trade school, nor any of the carpenters I worked with. I met lots who thought an untrue level should be discarded. I met lots who really fudged a lot. I can walk into almost any modern building and within a minute find a few examples of something which is not plumb or true, do this by eye, and be absolutely certain about it. Yet these buildings are perfectly functional, and provide suitable environments in which to work or live in perfect contentment.
My point is that we don't have to have perfect tools for measuring 'skill'. Rather, we want to use ones which are good enough to represent that which we are describing. How is skill to be defined? I'll leave that to those who will be part of it. I feel this is an excellent initiative. I'm sure there will be long and serious discussion, many difficult decisions to be made, and at the end of a long, thoughtful process, a very effective solution will emerge.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.