<p>anyone else get a call?</p>
<p>
</p>
<p>Because I do not have everyone’s complete applications, I obviously will incorrectly predict many of the Finalists. New York tends to be an especially difficult state to predict.</p>
<p>
</p>
<p>Just because you didn’t get a call doesn’t mean it won’t happen! Hang in there. Either way, you’re a very accomplished science researcher with a bright future; don’t worry ;)</p>
<p>Wow, congrats earthsong! That’s an amazing achievement! Let your colleges know if you haven’t committed anywhere early decision1</p>
<p>^ Thank you! And yes, I plan on sending updates as soon as I figure out how each school wants additional info.</p>
<p>has anyone from california gotten a call yet?</p>
<p>Got a call at 6:30 EST, but they call your home phone so if no one’s home… also I wasn’t on the “predix” either haha</p>
<p>got the call from california about an hour ago 
wasn’t on either of the above prediction lists, haha. the best luck to you guys! can’t wait to meet everyone who made it</p>
<p>No calls, no messages… Oh well, it was a good run anyhow and it was an honor to be in the same pool as you finalists.</p>
<p>To those who got the calls, congratulations (!!!) and good luck in March!</p>
<p>Got my call yesterday around 4, and I was on the second prediction list. From talking to naum55, the prediction method is actually pretty cool, and isn’t just based on titles :P</p>
<p>congrats to everyone :)</p>
<p>^ Then what was it?</p>
<p>i’ll let him/her explain it, because i’m sure that explanation will be better than mine 
or just pm naum55 if you wanna</p>
<p>i’m curious as to what it was as well…though there were only 6 or 7 right :P</p>
<p>Actually, the first prediction list got 7 finalists right and the corrected one was significantly improved: 8 finalists right!</p>
<p>Being familiar with programming, I tried to devise a computer program that would predict the Intel Finalists accurately. After ten minutes of coding, I wrote an ingenious program that on average got 5-6 finalists correct. Here was my approach: I took a list of the semifinalist, and I used the Python function random.sample to extract a random sample of 40 finalists. I found that around 30% of the time, the program gets 7 or more finalists right and around 15% of the time, it gets 8 or more correct. So it seems that naum55’s list is still <strong>significantly</strong> more accurate than mine.</p>
<p>Note: the following sentences are free from sarcasm.</p>
<p>Any mathematician, computer scientist, statistician, physicist, psychologist, chemist, or a member of the general public who understands probability would see that the prediction list is <strong>not</strong> significantly better than random guesses. About 15% of random collection of 40 semifinalists will contain 8 or more finalists. Under most reasonable statistical-significance tests, the prediction list would probably pass off as a random list of names from the list of semifinalists. Given these facts, several conclusions are possible, some of which are:
- naum55 actually chose the names randomly.
- He/she is not very good at judging the applicants without reading their applications.
- The applicant pool is so strong that luck often plays a strong role, and so most predictions will inevitably fail.
I think 1) is false (correct me if I am wrong), and both 2) and 3) might be true to certain extents, despite naum55 being “confident in about half this list” (I quote directly from the post of his first list). And since we can assume that naum55 tried his best (again, correct me if I am wrong), the moral of this whole incident is that predictions are so hard to make that they are not significantly better than a random list of names, and that if they are effectively just a random list, why spend the time to make them? Besides, this sort of list could be hurtful to people not listed on the list despite their superior achievements and could make those people listed over-confident about being selected as a finalist.</p>
<p>Disclaimer: I found great humor in this list, as evidenced by the first half of my post.</p>
<p>I actually got 8 correct 
Unfortunately (for predictions purposes), this seemed to be a difficult year in predicting. I’m not going to delve too far into my prediction process (if you want to know more, just PM me, or I’ll explain more when I have time). </p>
<p>But I’ve analyzed previous years of Intel Finalists, and found trends; however, they fluctuate (of course) from year to year, and this year seemed especially random in terms of Finalists selections.</p>
<p>Two of the best indicators in recent history have been ISEF repeaters (students who have gone to ISEF multiple times) and USAMO qualifiers. Thus, I listed all 12 ISEF repeaters and all 16 USAMO qualifiers. </p>
<p>Last year, for example, featured 9 USAMO qualifiers among the Intel Finalists. This year, there is a whopping 2. Additionally, two years ago, there were at least 5-6 ISEF repeaters among the Intel Finalists; this year, there are 4.</p>
<p>This is late, but does anyone know when the semifinalists would get their $1000? Did anyone already get it?</p>