<p>If we use the Putnam Competition results as a measure of the strength of a college’s math department, I don’t believe it makes a case for Harvard’s superiority:</p>
<p>For the past ten years, there have been exactly 52 Putnam Fellows (generally the top 5 individual winners). These are the stars and are indicative of the STRENGTH of a math department. Probably, the most famous Putnam Fellow was Richard Feynman. </p>
<p>Harvard has won 13 or 25% of all Putnam Fellow Awards since 2000.
MIT has won 22 or 42.3% of all Putnam Fellow Awards since 2000.</p>
<p>The next category are the next 20 individuals who get a cash award. There have 198 recipients in the same period (2000-2009).
Harvard won 41 or 20.7%.
MIT won 52 or 26.3%</p>
<p>Finally are those students who get honorable mention (most recently the next 40-50 individuals). There have been 464 recipients in the past ten years. This is generally a good indication of DEPTH of a math department.
Harvard has won 56 or 12.1%
MIT has won 127 or 27.4%</p>
<p>Both in depth and strength of its math department, MIT out-competes Harvard at the Putnam by a nearly 2/1 ratio. </p>
<p>The Putnam Team award is a meaningless metric as measure of a math department strength. You just need to look at 2000 when Duke won without a single team member being a Putnam Fellow while Harvard and MIT had 2 Putnam Fellows each that same year. A college team score is just the combined ranks (not scores) of three individuals designated before the competition as “official” team members. It is NOT the combined score of the top three scorers of a particular college which would be much more meaningful. Teams like MIT and Harvard have just too many strong members to know in advance who will be the top scorers. In 2000, neither of the two Harvard Putnam Fellows were even on the Harvard "official’ team and only one of the two MIT Putnam Fellows was on the MIT team. Had both teams done a better job picking their “official” members, Harvard and MIT would have easily beaten Duke in the team award.</p>