D'oh.

Rieux — Sun, 02 Mar 2008 09:12:56 -0800

My paragraph breaks appear to have been eaten by the software (mine or yours, I can't tell). Sorry about that.

A Couple of Tweaks

Rieux — Sun, 02 Mar 2008 09:09:17 -0800

Hi, Jerid. Nice applet. There are a few things I think deserve changes. First, none of the percentage windows allow the entry of tenths or hundredths. This could be a real problem: an extra 0.1% or two can make a huge difference in the delegate count. (For example, in a 6-delegate district like Ohio's Districts 9, 10, 13, and 14, the difference between Candidate X getting 58.2% of the vote and Candidate X getting 58.4% of the vote is the difference between a 3-3 delegate split and a 4-2 delegate win. The calculator here can't illustrate that.) Second, though I haven't looked "under the hood" at the Java, it appears to me that your applet counts the 49 statewide delegates as if they were a single pool of 49 delegates, to be given out according to the candidates' shares of the vote. That's not quite correct. As the window text testifies, 18 of the 49 statewide delegates are PLEOs; the other 37 are ordinary at-large delegates. The important mathematical point is that those pools are calculated separately, as if they were separate districts (that happen to have equal vote totals). As it stands now, the calculator doesn't appear to understand this. Try inputting 36% of the statewide vote for Obama, and 64% for Clinton. (Yes, this is an implausible finish--but the several more realistic finishes that demonstrate this problem require the entry of 0.1% units, which the applet won't allow.) Given 36% Obama-64% Clinton, the calculator yields 18 statewide pledged delegates for Obama and 31 for Clinton. This is reasonably simple math: 0.36 x 49 = 17.64, which rounds up to 18; and 0.64 x 49 = 31.36, which rounds down to 31. Unfortunately, that's wrong. The at-large and PLEO pools need to be calculated separately: 31 at-large delegates: 0.36 x 31 = 11.16, which rounds down to 11 delegates; 0.64 x 31 = 19.84, which rounds up to 20. 18 PLEOs: 0.36 x 18 = 6.48, which rounds down to 6 delegates; 0.64 x 18 = 11.52, which rounds up to 12. That makes 11 + 6 = 17 delegates for Obama and 20+12 = 32 for Clinton. So from 36%-64%, the output should be 17-32 in delegates, not 18-31. All of this, I'm sure, is fixable.

I'll be adding a button for it to the sidebar tomorro

Administrator — Sat, 01 Mar 2008 22:00:49 -0800

...after I get home from Cbus.

Thanks, Jerid

Tom Smith — Sat, 01 Mar 2008 21:56:46 -0800

Is there any way you can keep this on the top of the front page? Perhaps even have a few folks like Chuck Todd at MSNBC or someone at CNN take a peak.

Btw...help

Administrator — Sat, 01 Mar 2008 20:59:11 -0800

If anyone is good at javascript, I'd really like to add a variable that tallies delegates statewide based off entered percentages.

The calculator file is here if anyone feels like tweaking it for me to add that (I had to teach myself a sparing amount of java just to do that today).

http://filer.case.edu/jkk3/working1.html

Give me a heads up if you tweak the file, and I'll post the update (I literally knew nothing about java till today).

Ohio Delegate Calculator