Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : Zeev's Turnips -- Ignore unavailable to you. Want to Upgrade?

To: Zakrosian who wrote (272)	12/5/2000 3:21:52 PM
From: SBHX	Respond to of 644

Warren Christopher had it right : if the Bush side had said : "Well, we think and hope we won, but the race is close and we'll let the process run its course before we jump to conclusions." I think that would not have poisoned the air so much. Right now, both sides think the other is the devil incarnate, and there can be no reasonable dialog. Then perhaps if they weren't so sure, the only way to be sure is that statewide recount thing, which may not be a bad thing if people actually knew what "count" really means. Perhaps if I can digress and express the manual recount/dimple thing in a grossly simplified mathematical concept, it would help explain why people have such a different perspective on this thing. Consider a scale of [0,1], where 0 = pristine unpunched chad 1 = clearly punched chad. If the above suddenly became an analog scale then suppose (grossly arbitrary simplification again) 0.90 = hanging chad (single corner stuck) 0.50 = sliding door chad (two corners stuck) 0.33 = chad with three corners stuck 0.25 = chad with hole in middle .... 0.01 = very faint barely discernable dimple. 0.001= smudge on chad from sweaty palms of human counters Team Bush initially wanted to use the count(chad)=floor(chad) such that count(1)=1 and count (1-epsilon)=0. Remember? They were screaming about chads on the floor. This is what they really meant : count(chad)=floor(chad). Team Gore eventually wanted to use the count(chad)=ceil(chad) such that count(0)=0 and count(0+epsilon)=1. This is the method used in Broward County and initially in Miami Dade during the recount. The "recovery" rate was 25% for them. Michigan seems to use count(chad)=round(chad) Judge Barton(sp?) count function is complex and hard to represent correctly. A rough approximation seems to be count(chad) = round(chad+0.25) but this fails to take note of the other chads in the other races. (Don't really want this to degenerate into a multi-variable eqn). His "recovery" rate is only 8%. Note that correlating PalmBeach's result with Broward's result could mean (with huge statistical errors) that 25%-8%= 17% of the 'uncounted'(never mind)-ballots have only a single dimple (no pinprick either) on the presidential race with no 2nd dimple in other races (ie: other races are punched out). This is quite an interesting conclusion out of a small sample, so it means nothing. :) So, what does count every ballot really mean? Well, depends on the counting function. But it seems that a state-wide recount of every county should be fair irrespective of which count function is used right? The errors will be the same and should be distributed out fairly. Not quite. The only thing that it would miss is : what if the support for Gore was much softer and every 1/5000 of his supporters really changed their minds? while 1/50000 of Bush supporters did the same? This whole thing really does need ESP... Maybe the seminole and martin county court cases will toss out the absentee ballots and save everybody all the grief of figuring this one out. SbH