So, I made some (not-so) educated guesses about the first few dailies of the ARG Tracking Poll for Dean and Kerry. Based on that, I figured out the dailies.
Please note that these are not official. They're just based off what I judged to be a reasonable assumption of initial numbers, and calculations based on those.
That said, here goes:
ARG (My Daily Prediction) (3 Day ARG)
DATE DEAN KERRY
12-26 (36)(-) (20)(-)
12-27 (38)(-) (19)(-)
12-28 (37)(37) (18)(19)
12-29 (36)(37) (16)(18)
12-30 (38)(37) (17)(17)
12-31 (37)(37) (15)(16)
01-01 (36)(37) (13)(15)
01-02 (41)(38) (14)(14)
01-03 (40)(39) (15)(14)
01-04 (36)(39) (13)(14)
01-05 (35)(37) (14)(14)
01-06 (37)(36) (12)(13)
01-07 (33)(35) (10)(12)
01-08 (35)(35) (11)(11)
01-09 (37)(35) (09)(10)
01-10 (33)(35) (10)(10)
01-11 (38)(36) (11)(10)
01-12 (31)(34) (12)(11)
01-13 (27)(32) (16)(13)
01-14 (29)(29) (17)(15)
01-15 (28)(28) (15)(16)
01-16 (27)(28) (22)(18)
01-17 (29)(28) (20)(19)
01-18 (28)(28) (15)(19)
01-19 (27)(28) (25)(20)
01-20 (23)(26) (32)(24)
01-21 (16)(22) (24)(27)
01-22 (15)(18) (37)(31)
01-23 (14)(15) (41)(34)
01-24 (19)(16) (36)(38)
01-25 (27)(25) (37)(38)
01-26 (29)(25) (32)(35)
So I predict Dean's final sample is 29%, a rebound of 15% from his low point only three days earlier. Kerry's predicted final sample is 32%, down 9% from his high point just three days ago. Interesting- Dean's lowest day was Kerry's highest. This suggests to me that Dean's support abandoned him for not only undecided, but Kerry as well. However, it appears a significant portion has returned to him.
As someone else mentioned, ARG has painted the bleakest picture for Dean as of late. And even they now recognize his incredible, late surge. Granted, these numbers are only predictions, but... they're close ball park figures. In addition, the final predicted samples (K 32 D 29) are with in the MoE of 4%.
Based on the trends of the last three days, I predict a nailbiter (big suprise), with the slimmest margin of victory for Howard Dean.
Discuss.