jordin bottom 3? buy/sell?

[smbruin22]

dialidol has jordin as 3rd bottom, decent room to both 4th and 2nd.

if she is in bottom 3, do you think it's good buying or selling opportunity? assuming shares falls a few points.

or will AI producers massage the votes slightly or cut it off at bottom 2?... can't figure out if being in bottom 3 helps or hurts. i actually think it helps. EDIT: assuming you survive that incidence.

[thing85]

Man, you are putting way too much stock in the DI numbers right now. Voting has only been open for 45 minutes and isn't even open yet on the West coast. Furthermore, the MOE is such that everyone except Melinda could be in the B3.

Your post is entirely too panicky and "jumping the gun." I doubt Jordin will actually be B3, so don't worry about it.

[smbruin22]

i lost a bigger response i typed out...

everyone mentions MOE, but how many people have left in recent weeks that weren't near the bottom?.... i think DI's MOE are way too tight, partly because they don't want to look stupid. i'd just look at the scores and the earlier scores have been best recently...... i think the best returns (without tying up money for long) last year were for unexpected departures. no money in "coin-flipping" whether hayley will survive.

[revots33]

Last week Dial Idol had a total of 209,000 votes, out of over 30 million. I'm no statistics guy, but is this even enough to draw any meaningful conclusions?

And, according to an article posted on MJ's blog:

"About busy signals, Ken insists that there is no congestion on the national telephone grid. The source of problems, he says, are the local exchanges."

So what we have is a tiny percentage of votes, measuring busy signals that vary with local exchanges and so might not have any bearing on the the total voting volume for a particular contestant.

Given that, and the results so far this year, I see little reason to pay too much attention to the DI numbers. Maybe they will be more predictive when we have 2-3 people left, but at this point they seem pretty useless.

[StellarWind]

[ QUOTE ]
"About busy signals, Ken insists that there is no congestion on the national telephone grid. The source of problems, he says, are the local exchanges."

[/ QUOTE ]
Sometimes the best way to lie is to tell the truth.

The reason the "national telephone grid" isn't congested is because AI probably doesn't use it!

When you place a phone call, you send the dialed number to a local telephone switch which is actually a powerful computer. This switch will see 866-XXXXXXX and realize it's not a real area code or phone number. No problem, the switch sends an inquiry over a data network to a database server that specializes in decoding toll-free numbers.

The server provides a real phone number and the long-distance carrier information. Since toll-free calls are charged to the receiver, he gets to pick the long distance company. In this case AT&T is a show sponsor.

The trick here is toll-free routing can be very sophisticated these days. The server doesn't have to provide the same routing information for calls from Philadelphia as for calls from Boston. I don't know what their setup is, but probably the last thing you want to do is route 30 million calls from all over the U.S. to a single call center in Hollywood. My guess is that a call from Boston may never leave the city. A local AT&T server would accept the call, tally the vote, and play the recording. This avoids a fortune in long-distance costs, protects the network from overload, and avoids the pain of building a call center and local network that can take 30 million calls in one place but is only used a few times a year.

The obvious implication for DI is that congestion is a local phenomena. The local server in Boston may not be quite big enough while Atlanta happens to have excess capacity. Busy signals in Atlanta end after two minutes while Boston has problems for an hour. That can magnify regional differences between candidates into major distortions.

"Amy" may be hot among DI users in Atlanta while "Bob" gets lots of DI votes from Boston. Bob will get many more busy signals than Amy because of where his sample is being measured. There aren't that many different DI users in the count and some contestants aren't very popular with DI users. Some of the reported results may be based on only a few dozen dialers. Even if there weren't regional biases that's not exactly a big sample.

This theory also explains why DI has gone sour this season. AT&T probably made system improvements during the offseason that undermined DI's accuracy.

[smbruin22]

revots, interesting comments... not sure who ken is, although i'd seen some reference to that earlier.

if it's AI saying there are no national problems with busy signals, what do we expect them to say, "problems with busy signals because we don't have enough capacity??".......

i understood it wasn't votes, but busy signals... but doesn't it have most of the best people at the top most weeks.... doesn't that suggest it's capturing something worthwhile??

[smbruin22]

[ QUOTE ]
revots, interesting comments... not sure who ken is, although i'd seen some reference to that earlier.

if it's AI saying there are no national problems with busy signals, what do we expect them to say, "problems with busy signals because we're too cheap to buy enough capacity??" ......... of course, maybe i don't understand who's saying that.

i understood it wasn't votes, but busy signals... but doesn't it have most of the best people at the top most weeks.... doesn't that suggest it's capturing something worthwhile??

[/ QUOTE ]

[Brocktoon]

The numbers look pretty much in line with what I'd expect right now.

1 Melinda Doolittle -

2-5 Phil Stacey -

2-5 Chris Richardson -

2-9 (B3) Sanjaya Malakar -

2-9 (B3) Blake Lewis -

4-9 (B3) Lakisha Jones -

4-9 (B3) Jordin Sparks -

4-9 (B3) Haley Scarnato -

4-9 (B3) Gina Glocksen

The reason I think Phil is so high is because the judges, Simon in particular, were so over the top harsh on him. Standard sympathy vote.

Gina probably turned a lot of people off with her catty backtalk to Simon after he said she got outperformed. It came off looking bad.

[revots33]

[ QUOTE ]

i understood it wasn't votes, but busy signals... but doesn't it have most of the best people at the top most weeks.... doesn't that suggest it's capturing something worthwhile??


[/ QUOTE ]

I dunno, most people seem to agree that Melinda, Lakisha, Blake, and Jordin are the Top 4 (at least according to the betting lines). Last week DI had Melinda 1, Lakisha 3, Blake 7, and Jordin 8. Not saying this couldn't be close to correct - but it seems to me that you or I could make up our own Top 10, and have just as good a chance at being correct.

DI did have a pretty good record last season, but there were a couple of weeks where it was WAY off. So were the off weeks just variance, or were the correct weeks just a lucky coincidence? Again, could you or I have had just as good success as DI last season, if we had the same margins of error? Probably.

[thing85]

FWIW, Lakisha has fallen below Jordin into the "B3." Again, I don't think it means much, but considering Lakisha had the pimp spot, she isn't doing so hot on DI this time around. Assuming sympathy votes save Phil, my money is on Haley or Gina. Gina has never been B3 before, and the harsh comments on her may arouse her fan base as well.

Haley the odd one out? I'm expecting a "surprise" member of the B3 this week including one of Lakisha, Jordin, or dare I say, Sanjaya.