While making preparations for this year’s MOP, I imagined to myself what I would say on orientation night if I was director of the camp, and came up with the following speech. I thought it might be nice to share on this blog. Of course, it represents my own views, not the actual views of MOP or MAA. And since I am not actually director of MOP, the speech was never given.
People sometimes ask me, why do we have international students at MOP? Doesn’t that mean we’re training teams from other countries? So I want to make this clear now: the purpose of MOP is not to train and select future IMO teams.
I know it might seem that way, because we invite by score and grade. But I really think the purpose of MOP is to give each one of you the experience of working hard and meeting new people, among other things. Learn math, face challenges, make friends, the usual good stuff, right? And that’s something you can get no matter what your final rank is, or whether you make IMO or EGMO or even next year’s MOP. The MOP community is an extended family, and you are all part of it now.
What I mean to say is, the camp is designed with all 80 of you in mind. It made me sad back in 2012 when one of my friends realized he had little chance of making it back next year, and told me that MAA shouldn’t have invited him to begin with. Even if I can only take six students to the IMO each year, I never forget the other 74 of you are part of MOP too.
This means one important thing: everyone who puts in their best shot deserves to be here. (And unfortunately this also means there are many other people who deserve to be here tonight too, and are not. Maybe they solved one or two fewer problems than you did; or maybe they even solved the same number of problems, but they are in 11th grade and you are in 10th grade.)
Therefore, I hope to see all of you put in your best effort. And I should say this is not easy to do, because MOP is brutal in many ways. The classes are mandatory, we have a 4.5-hour test every two days, and you will be constantly graded. You will likely miss problems that others claim are easy. You might find out you know less than you thought you did, and this can be discouraging. Especially in the last week, when we run the TSTST, many of you will suddenly realize just how strong Team USA is.
So I want to tell you now, stay determined in the face of adversity. This struggle is your own, and we promise it’s worth it, no matter the outcome. We are rooting for you, and your friends sitting around you are too. (And if the people around you aren’t your friends yet, change that asap.)
Power Overwhelming 
>>>”People sometimes ask me, why do we have international students at MOP?”
I think the real answer is that the US never invites peer-level competitor countries to train together at MOP (or does it?). As long as the internationals are from countries with few gold/silver medalists, the marginal effect on medal cutoffs is close to imperceptible for the US team members, and there is no effect whatsoever on the US team ranking.
>>> “So I want to make this clear now: the purpose of MOP is not to train and select future IMO teams.”
The US results of the last few years show that the training has gone into overdrive. That’s after cranking things up to 11 about a decade ago (2010?) by extending the training by a year (“TSTST”). So if the amp goes up to 12 now, it certainly looks as though the sole purpose of MOP is to train and select IMO teams and anything else is (even if beneficial and intentional) a side effect.
I’m curious as to what has changed lately to allow the USA to outperform China (China!) at the IMO. Is the training even more IMO-optimized than it was in, say, 2012?
LikeLike
Re invitations: we invite the *top* N countries of the previous year’s IMO to MOP, which in particular includes China/Korea/Russia every year. So I think those count as peer level :)
> Is the training even more IMO-optimized than it was in, say, 2012?
I could make a whole blog post on why the USA has been doing so well recently, but the short story is that it’s not the summer camp, which has gotten much more relaxed compared to early 2010. Zuming Feng would always say “MOP is a three-year program, not a three-week program”, and I basically agree with this: the meaning is that the preparation you do the other 49 weeks of the year matters much more than just the 3 weeks at camp.
(I think it’s a little misleading to say that TSTST extended the training by a full year; it extended the team selection process by a full year, but it’s not as if there is any sort of instruction or even preparation material distributed by the program outside of the three weeks. I’m actually planning to complain about that in a post or two.)
LikeLike
Wow. That’s a huge change from the olden times. What does it mean, logistically? Have other top (small N) teams come and stayed for the duration of MOP? Are we talking about their actual IMO teams for that year, or a sort of summer “student exchange” where their future potential IMO-ers train at MOP instead of at home?
I vaguely recall hearing or reading that in the ultra-olden days of Klamkin and Greitzer as leaders, the US combined the MOP training with Canada in some years, and even that might have been only for the last week or so when only the IMO selectees were there. Later on, some teams from countries whose training systems were much less developed than MOP were invited. To go from that to inviting the top N countries is pretty amazing. I’d love to know more about this MOP-sharing arrangement.
The TSTST approach closes the US selection pool for the IMO one year before the contest. (At least that was the policy when the system changed around 2010, I don’t know if things have changed since then.) Officially negating the possibility of late bloomers making the IMO team makes only a minor and occasional change to the composition of the team, but a large change to the nature, philosophy and purpose of the whole process. Before 2010 the IMO selection was a pyramid of open individual competitions, since then its purpose has been to maximize the expected results of Team USA versus Team China.
LikeLike
The top N countries (N=8ish in a typical year, after declines) can send two students to MOP. It’s up to the countries who they want to send; some like to send their team members, others will send their 7th/8th rank instead. It sounds like not a lot, but it is 25% of the camp being internationals. USA pays for all expenses. There’s a few reasons for having this program, but int’l exposure and building worldwide connections between the students is a big one. We think the USA students learn a lot from the intl’s and vice versa.
LikeLike
A footnote to the comment about inviting Canada to the training in the very old days. This comment by David Ash (Canadian IMO/Putnam god from back in the day) over at Andrew Gelman’s blog indicates that this arrangement was at least proposed around 1980, and he himself attended MOP by special arrangement. At my age it’s reassuring to learn I wasn’t confabulating the memory! But it’s still not clear whether a larger Canadian contingent at MOP ever happened. Also, quite a few Canadian IMO team members are US residents with Canadian citizenship, and some of those trained at MOP individually regardless of any binational arrangement.
MOP aside, the MAA has gone back and forth on participation of Canadians in the USAMO over the years. The Canadians used the exam as one of their selectors for their IMO team. Making the CMO that difficult was infeasible and the USAMO provided valuable information earlier in the year about potential for the IMO. I never knew what the reason was for excluding (or including) Canadians, and now that I think about it I also don’t know if ineligibility ever extended to the AHSME (AMC 12 in current parlance) or the AIME. Maybe you know more about this history.
LikeLike
( Ash’s comment : https://statmodeling.stat.columbia.edu/2015/03/17/1980-math-olympiad-program-now/#comment-936977 )
LikeLike
I’m too young to know the history of USA-Canada relations with respect to the AMC’s, alas. (Actually, I admittedly don’t even really understand the current situation.)
LikeLike
older IMO veteran, yes, I am the person who posted earlier at Andrew Gelman’s blog. The thing is that I may not have a complete picture even of what was going on in 1980. I was invited to, and attended, the 1980 USA MOP at the US Naval Academy in Annapolis, MD. I was told this was in anticipation of the two teams possibly training together in 1981. There was no IMO at all in 1980. I had not been aware of any Canadian MOP in 1980, but there was one in 1981. However fast forward to 2016, when I attended an event at the Canadian MOP in Banff–my employer at the time was helping to fund the team. At that event, Robert Morewood (perhaps you are he?) told me that he had attended a Canadian MOP in 1980 at the University of Toronto.
It has been over 40 years, so I might be forgetting something, but I don’t even remember being invited to the Canadian MOP in 1980–which Robert seemed to recall pretty clearly. Presumably I would have been finishing second in the Canadian Olympiad that year. Maybe I was an early example of an international attendee at the US MOP, and a coordinated decision was made to invite me to the US MOP rather than the Canadian equivalent. Even after more than 40 years, it would be good to know what was going with planning in something where I seem to have played a key role, but without fully understanding the role I was being asked to play.
Ed Barbeau has an article from years ago about collaborating with Geoff Butler in those early years training the team (https://www.projecteuclid.org/journals/rocky-mountain-journal-of-mathematics/volume-20/issue-4/G-J-Butler-Summers-on-Olympus/10.1216/rmjm/1181073043.full). From that article, it appears that the Canadian IMO team trained at the University of Calgary in 1982 and Queen’s University in 1983, so at least those years a joint MOP with the US didn’t happen.
At least some years–much more recently–I believe the Canadian team has trained with the Mexican team.
LikeLike
As someone who was part of the 2015 USA team, I also do not know why we outperform China now. It could be that China is weaker. It could also be that we are far better at geometry, due to Evan’s cultural influence or for some other reasons (though I am certainly not). As with anything in social science, you would need some careful statistical tests.
Anecdotally, the demographic distribution of people doing math olympiads has also changed in the last ~5 years. My impression is that the geographic, racial, and gender diversity of strong math olympians is even lower than ~5 years ago when I was in high school, which could be consistent Olympiads are now driven more by training. Anecdotally it also seems that fewer math olympians are going into academia now, and more to finance and industry, which again is consistent with math Olympiads becoming more about training and less about math. So perhaps our system has become more like the Chinese system.
Maybe Evan and his cultural attitudes are to blame, what with writing EGMO, his extensive tutoring, his belief that contest math has value independent of research, and so on. In any case though I think some of these changes I am describing came *after* we started winning. Anyway, as always in social science, you would need to look at a lot of data. It is easy to fall into the trip of believing some plausible sounding hypothesis that confirms some pre-existing bias.
LikeLiked by 1 person
[…] now going to say something explicitly that I hinted at in June: I don’t think a student deserves to make MOP more because they had a higher score than […]
LikeLike