Along with a tremendous amount of analysis on the three proposed student assignment plans, MIT released a technical appendix with details on how they created the demand model and used it to analyze the plans. Unless you're a pretty serious geek, I don't recommend trying to read this report. I did, however, find some tidbits that might be of more general interest.

One question parents often ask is whether their child is guaranteed to be able to stay in his or her current school if the family applies for another school and doesn't get in. In the past, the family resource centers have given conflicting answers to this question. According to MIT, there is a priority for students to stay in their current school. This means that if you apply to a new school, but list your current school second, you will get to stay in your current school if you don't get your first choice. But it turns out that this goes even further. Even if you don't list your current school, BPS will append it to your list of choices with the same priority. So if you apply to five new schools and get none of them, you will remain assigned to your current school. This confirms that parents should feel comfortable applying to a new school knowing that they won't lose their current slot.

The other interesting piece of information relates to twins. I've long suspected that twins have an advantage in the lottery. BPS has a stated policy of trying to keep twins together. Anecdotally, most of the families I know with twins have seen both kids get into their first choice. One exception was a family where one child got in and the other was first on the waiting list. So it did seem clear that BPS puts twins one right after the other when ranking them in the lottery.

Given the anecdotal evidence that twins seemed to be getting their first choice at a high rate, I suspected that BPS essentially gave both twins the lottery number of the child with the better number. The report confirms this. It does seem that this gives twins an advantage, but there are a few things to remember. One is that both twins could easily get very good or very bad lottery numbers. In those cases it won't make much difference. Another thing to think about is what would happen if BPS didn't do this. Let's take the case where one twin gets a great number and the other a poor one. One twin will get into his or her first choice, the other ends up on the waiting list. However, the child on the waiting list will get sibling preference because his or her twin is already assigned to the school. So it's fairly likely that the child will be first on the waiting list and will get in anyway. In the end, the impact of this policy might be minimal.