Measurements of central tendency
Jul. 20th, 2019 08:47 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
offcntrPeople are always amazed when they order a set, just how well matched in size the pieces are. They imagine me with calipers in hand, carefully measuring each piece as it comes off the wheel to get the perfect set.
It's not like that at all.
Frankly, I hate calipers, hate measuring. I only do it when I absolutely have to, say, when making pots with lids. Any other time, I a) weigh the clay out ahead of time, b) throw them the same way as much as possible, and c) make lots of extras.
Most of the time, I'm making a production item anyway: plates, soup bowls, mugs. I'll have use for all of them; I might as well make a bunch. (When I'm not making standard ware--I just took an order for four dinner salad bowls, which are no longer a regular thing--I'll grit my teeth and get out the measuring devices, but otherwise, it's free-throwing. Stuff I can do in my sleep.)
Then I use a trick one of my professors shared with us years ago: Take all of the pots and line them up on the table, smallest to largest. Pick the number you need from the middle of the line, including a few extras in case of misfiring. Set the rest aside and glaze with standard patterns, put 'em in the booth to sell.
Here's a plate order for this last firing, five matching octopus plates (plus two or three extra). Just for funsies, I decided to do a statistical analysis of my data.*
It's a total of 20 plates. The stacks run from smallest diameter on the left to largest on right, in 1/8-inch intervals. Right away, you can see the 11.25" stack has enough for the whole order; if I'd been short, I would have added in some from the next bigger or smaller set.
All in all, they're surprisingly uniform, less than 3/4 difference between largest and smallest. I was expecting a normal (bell-shaped) distribution, but I have more of a bi-modal curve with a long left tail and maxima at 11 and 11.25. Even so, the mean diameter is 11.12", mode is 11.25" and median falls right between 11.125 and 11.25".
That's pretty darned close.
*What can I say? I took a double major in Art and Mathematics as an undergrad; the Math classes were for fun.
It's not like that at all.
Frankly, I hate calipers, hate measuring. I only do it when I absolutely have to, say, when making pots with lids. Any other time, I a) weigh the clay out ahead of time, b) throw them the same way as much as possible, and c) make lots of extras.
Most of the time, I'm making a production item anyway: plates, soup bowls, mugs. I'll have use for all of them; I might as well make a bunch. (When I'm not making standard ware--I just took an order for four dinner salad bowls, which are no longer a regular thing--I'll grit my teeth and get out the measuring devices, but otherwise, it's free-throwing. Stuff I can do in my sleep.)
Then I use a trick one of my professors shared with us years ago: Take all of the pots and line them up on the table, smallest to largest. Pick the number you need from the middle of the line, including a few extras in case of misfiring. Set the rest aside and glaze with standard patterns, put 'em in the booth to sell.
Here's a plate order for this last firing, five matching octopus plates (plus two or three extra). Just for funsies, I decided to do a statistical analysis of my data.*
It's a total of 20 plates. The stacks run from smallest diameter on the left to largest on right, in 1/8-inch intervals. Right away, you can see the 11.25" stack has enough for the whole order; if I'd been short, I would have added in some from the next bigger or smaller set.
All in all, they're surprisingly uniform, less than 3/4 difference between largest and smallest. I was expecting a normal (bell-shaped) distribution, but I have more of a bi-modal curve with a long left tail and maxima at 11 and 11.25. Even so, the mean diameter is 11.12", mode is 11.25" and median falls right between 11.125 and 11.25".
That's pretty darned close.
*What can I say? I took a double major in Art and Mathematics as an undergrad; the Math classes were for fun.