Not to be rude, but thats your fault @SillyJillie423. I would’ve solved that the next day.
I feel that I am not in the top 6% and I was never asked to write a LOR, does that mean I can assume I was rejected?
@Stonyfire If you don’t feel like you’re in the top 6% in terms of stats + you didn’t receive an LOR request, then unfortunately I wouldn’t get my hopes up for an acceptance.
On the other hand, though, a lot of this is still pretty ambiguous; we can only infer so much from the numbers. If you’re able to post your stats & result once decisions come out, that would be helpful for seeing whether what we’re hypothesizing here is correct.
Maybe it’s harder to analyze someone OOS vs in state?
Any undeclared engg who got lor request
what would top 6% be considered in terms of stats? I have a 1520 SAT, 33 ACT, UC GPA 3.73 and a 3.83 UW GPA
@ALLCAPS24 How come your UC GPA is lower than unweighted?
@ALLCAPS24 I’m sorry but I heavily doubt you are top 5-6%.
My child is definitely not in the top 6% and received a request for LOR. In state student. Just to add to the mystery of LOR requests. 
LOR requests were sent after a preliminary review of applicants, too. So, not all were stats-based. My OOS daugher received a request for LOR in December and she submitted both. We will know soon enough if she was borderline and it made a difference.
So basically Ima get rejected? @Bdaves12
@Bdaves12 You don’t have to be top 5-6% to get admitted, even if you don’t get the LOR request. In the video that everyone keeps referencing, the guy explains that letter requests went out to borderline applicants. If you didn’t get one, you were probably either Yes or No. But not getting one does not mean No. That just wouldn’t make sense, since they accept so many more students than they sent out LOR requests for.
D admitted after LOR request. She submitted 2.
For future reference, my son did not get a LOR request, was not a top applicant, yet was waitlisted. We thought for sure he would be rejected since he did not get a LOR request.
I’m thinking two groups of people get requested for LORs. The one group are the borderlines which may or may not help your application depending where you are among the group of the borderlines. I have three classmates that were all asked for LORs, their sophomore-junior year grades were good but their test scores were not competitive for Berkeley. Two sent in letters and the other one did not. They all got in. I suspect the one person that did not submit a letter was desirable enough without a letter while one of the two that did submit, I believe may have not gotten in without one resulting in her acceptance after submitting the LOR.
I believe the other group of people that were requested LORs with high stats that they most likely wouldn’t be considered borderline are probably people who are considered for regents. Depending on that letter, they may or may not be formally asked to be a regents scholar nominee.
This is just what I think based on my observations
I have watched the video by Prof. Jesse Rothstein three times. Here is the summary of my understanding.
Key points:
Only California applicants were considered for the analysis (the process may have been similar for OOS and International students)
There were 45570 California applicants in 2015, 8570 were admitted.
There were 45626 California applicants in 2016 and 9610 were admitted.
Quantitative information from the applicants (SAT, ACT, AP etc.) were run through a statistical analysis package to predict the admission score for the applicants.
Admission score roughly translates to the probability of being admitted given the raw quantitative information.
For example, somebody with a SAT score better than 2350 and perfect AP scores in 5 APs may have an admission score of 0.98 because 98 out of 100 applicants in 2015 with those stats were admitted.
At the other end of the spectrum, somebody with a SAT score of 1100 and no APs may have an admission score of 0.02 because only 2 out of 100 students with that SAT score were admitted.
Around 70% of applicants had admission score less than 0.2.
40% of applicants had admission score less than 0.02.
Fraction of applicants admitted was more or less uniform across admission scores. This means 70% of admits had admission score greater than 0.3 and 80% greater than 0.2 and so on (please read this line twice).
Those with admission scores between 0.3 and 0.9 had greater than 50% chance of being asked for LORs
Admission score of > 0.9, only 33% were asked for LORs
Admission score of < 0.2, only 10% were asked for LORs.
Total of 14406 California applicants (32%) were asked for LORs.
46% of those asked for LORs got in.
In reality, the admission score above was not used to ask for LORs. A different parameter called “Predicted Read Score” was used to ask for LORs.
Predicted read score is a number used to classify applicants into “Yes”, “May be” and “No”. A machine could arrive at this predicted read score based on quantitative information.
A human reader can go over the application and decide “Yes”, “May be” and “No”.
The machine gives a predicted read score between 1.5 and 4.5. The lower the number, the better is the applicant. A number generated by the machine between 2.4 and 3.25 is classified as a “May be”.
All the “May be”s were sent LOR requests.
If Predicted Read Score was > 3.25, there was less than 25% chance of being asked for LORs
If Predicted Read Score was < 2.25 (top applicants), there was less than 50% chance of being asked for LORs.
80% of LOR requests came from machine recommendations. 20% of LOR requests came from humans. (There was a 40% overlap).
Regression discontinuity at the boundaries:
Compared to no LOR request, being asked for LORs reduced the probability of being admitted (from about 85% to about 75%) for the top end of the applicants.
Compared to no LOR request, being asked for LORs increased the probability of being admitted by 2 to 4 percent for applicants with more than 3.25 predicted read score.