Countdown to Match 2012

Webinar on May 21: Transitioning to Residency
This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
angleoflouise said:
OH! Hahahahahha clever.:thumbup: I guess she could go for 19 too.
Click to expand...

What's sad is many people waiting for Match Day weren't even in high school yet when that movie came out. :scared: I feel old.
Members don't see this ad.
 
freshman_fifteen_feet_1.jpg
 
ferrari-f12-berlinetta-1.jpg
search
search
imgres
http://data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/2wCEAAkGBhQSEBUUEhQWFRQUFRQVFBQVFBUUFBQUFBQVFBUUFBQXHCYeFxkjGRQUHy8gIycpLCwsFR4xNTAqNSYrLCkBCQoKDgwOFw8PGiwkHCQsLCksLiksLCkpLCksKSkpLCkpKSwsLCosLCkpLCkpKSkpKSwpKSopKSkpKSkpLCksKf/AABEIAKYBLwMBIgACEQEDEQH/xAAcAAAABwEBAAAAAAAAAAAAAAAAAQIDBAUGBwj/xABJEAACAQIDAwgGBwYDBgcAAAABAgADEQQSIQUxUQYTQWFxgZGhIjJCUrHRBxQjYpLB8BVTcoKi4YOT0hYzQ0RUsiQ0Y8LD4vL/xAAaAQEBAQEBAQEAAAAAAAAAAAAAAQIEAwUG/8QAMREAAgECBAMGBgEFAAAAAAAAAAECAxEEEiExQVFhBRMUkdHwIkJxgcHhMhUjkqGx/9oADAMBAAIRAxEAPwDoMKOZIWSeZ7iLQWi8sGWANwWjmWDJAEBYeWLCw8sARlgyxdoMsATlgtF5YMsARaERHMkPJAI7KYESSObh5IAhRF2h5YJCiHEbCR4rE2ggjLAVjmSHkgDGWDLH8kGSUDISApHckSVgDREQxjpWIKQBhmMaYGSWSINOUEcpE5ZINOJNOCDBELLH8kTlgDOWJKx4rE5IA1licseKQskpDQlIXNxyCYFxvmoRpx60KUXGubhc3HrQssC41kgyR3LDymAMc3AKckBIrm5BcjilFClH+bh83AuMc2IMkfyQ+bgXGMkHNx/JCywW4zkg5uPZYMsC4xzcHNx7LBlgDWSDJHcsGWANZIMkdyQ8kAZyRJSP5IRUQCOUiDTkqwiTaAQ2pxBSTTEGW4IRSJKSaREFYuUiGnEmlJZSFkgETmoXMyXkhZJQReZhczJWSFkgFjeCKgkMBQWiocARaHli4LQBOWHaKtKPbPLCjQY01DV6676NGxKcOeqE5KQ/iN+AMEuXYWAuBvI8ZzPafKvH1b/aUsMnQtJOdfvq1PRv2JM1iNrYgG/1quT1utvw5cvlN5GTOjtr41Bva3jGKu38OvrVkXtZR8TOc7J221bDVAxuQrHfudMt8vAMrXsNAUa2+cgcMHZQCSrEGwvuJH5SKOtj1aWRSPTb8s8EN+Koj/EU/Axs8u8B/wBXS/ET8J5qy1PdbwgAqcG8DNZDzzHpQcvMB/1dL+r5RQ5b4E/83S8T8p5pNZhvuO28cTFHiYyImY9MJytwZ3Yqj/mAfGP09vYZt2Ion/FT5zzPTxp4yTTx54yZEMx6ap4hG9V1PYyn4GO5Oqeb8JtBr6b+qX2z9uV09V3XsY/kY7smc7hBOa4Dl3WGjNm/i1PnrNVszlalSwfQ8Ru8JHBo0pJl/ChU3DC6kEcRFZZg0JgtF5YMsAbtCyx3LBlgXGcsIpHssBWARysQVkkpElYBGKxNpIKRJWCjBESVkgiJtKBjLBkj2WDLAJOWGFi7Q7QYuIywwsVaHlgBWgtImM2otMbix4A2+MpKPLUO+RUym9jffB6RpykrpDXKDaVSriWwlJ3pU6aK2IrJ6NR2qXyUKT+x6IJZxruAtqZSYqlToU8lNMlNbkqgJ7SbXJJO8nU9JlttZWZ+cpC77mW4GZeFzYX7SO0TN8o8TjFoE4amy1GNnDLY5NfUciwN7eJtN3stNzGS0lm26FBj+USkkKrHylY+MD7hY8DKuttTHA/aIWP3qdGp5lSfORH2xVzAtRCi+pWkyG3Taxt5TzTqp6nbPwbg1FNM2XJGt9qyHcwHnej/APNfumQx1Pm8c6++PMf/AJPjLfYfKWkGIAOdtNRuABI87HuEh/SEo59a1PQVBmU8AwzAfET0k7STPCms1Ca5Wf4EWkfEY1U6bngPz4SvbE1mQeiQLetbLm67n8pCyHiPxD5z0ucqiScRjS/ZwG7+8aV4VPDknXQcQM3kI+uCHF+6nr5tbzkujWVhLVk3AoHJBNjbTtkJcI53Kx/kb8gRHkwVX9234TFw4slpUKNY6EGaPZ2Lzrfp6RMyNn1j/wANz3GTMPgcQuopuOwGW5lxbNhSIO+PojLqh7vlMzhKmJX2H7CpPxk8coGp/wC9UKPvAp4X3y3MZWbTYvKdkNiSOPyIm62btdKoGoDeR7Pl8ZyOltWhVFy6oegl1+N9e+SMHt3m3sHU26VdTfuvMyimajJo6/RxSOWCOrFTZgrBipHQ1joeox605BQp1Kdb61gTarqalA+rWB1IXUanhfrBHqnpPJnlAMZR5wIUKnKwPQ4HpDuOn5CeFmtzpko2Ti/UtrQWh2gg8wssLLFQoAkrElY5EwBsrElI6REkQUaKQskdIiYA3lhZYsiCASLQQQWgyACNY6sbWEXXrrTRnY2VFLMTuAUEknwnnzbO38RtGs1Z8Q9CjmPMU1Yg5QbZiARr949NwLATLvfQ3FLdnYqyE3vKXGYVKTiq1wL2YqpbTiQOgcZVfR5ymeozYTENnqIuanUP/EQbwfvAa9xmo2zsOlWAaqagFMlvs3dCdOnJ6R7BMOXB7nbBpfQLFY2lSCl6ihXF0bXI3To9st7a7+uO0agcXRgw4qQw8RF1sEuTmyoKWC5DqLDQfDfMbtjki9MmphSxA1yAnnF/hI9cefbvhzkiwhGWl7M1dbCq3rKrfxKG+ImN5e0RhqC1KOGoauA7GghCgg20AtqdLn85X0OW9ekcrVC1uioub+rf5y7wf0h03FqtIEHfkYEW60b5y96mjfh2nqrmE5I7O+sYsVKlJaiksWXVEULYkhUGoAOi9JInR22fh6lvsEKrohKi3STlHQAT437S7Q5R4ErlFRaWa91IFPtBK6C/aN8tqNFHUGmVZegoQy26ACukReu4npLRWRR/7NYUkk0UBPUPzBtDp8lKJNk50Hghp6dvoad8uKOGNRrL6oJDNxI0Kr2dJ6N2pvZzbfKXCbNpjn3CFrlUALO3EhRqe02HXNO3Ey6rjsVjcg1tdsQ6dopVD/2AfGRq2wcMn/MVSeIWiPgkgLy/weObm0xD0ajGyCugRGJ3AMpIB7SO+Uu1Uq0qhSoCGHn1g9ImJSUeB04eMq3zeVi7qYamPUr/AI6ZPmjj4RzC4SuwJpolcKbHmqjKw0vqrgndwvMe1dpK2dthqR3ldb50sG7Docy9XhwmI1It2eh1VcPUjC9NXfJ/o0w2mENqtCspHRnW/gyLJdDbWDPrGunbSDDxRjF7L5bh7JiVBvYBrZqbEi9kcga2vp1dMuH2VhawvkAv0jQd9rW7Dae7pS3Wp8uOMot2nFxf19SPg6uCqGy4pCeDHI3g9pbU9gU/3h8pmto8gqbaIT1A2bwGhPaLzPtyfxWHJOHqNpvVGuO+k3ynk21ujqVKFRfBU8zpLcnKXSb9qqR5iKHJ2h0hT/KnynNsPy2r0zasp/jp+if5kOh/WksBylaouam4fs9Fh2ru8hKpxMSwVVbs31HYGHXdTTuVR8BLGllUWA07ZySvyvqp6xYdVjfyjQ+kJh7Tfgf5TSmjylgZ8zsuZenT9cZneVPKSrgxnXB1cTRtc1KDqzJxz0rXt94XHG0wtH6Sz0lvwP8AKFgOX9UYleau6N6yMGVR1hmAynr+M1mR5eElzRLo/T9hGIH1fEXJsAObYk8AMwvN9sna/PUhUajVoA7lrhVqEcSisSvY1j1Sq2eMKK/1lKVNalQenUVUzn+IgesNbkHXpvpLTaGHLLmU3B4TRzuNtGrEr64nvCOBwdxmFx2KKHUytPK9qZ0O43B4HjMtpHusK5L4TpkTKrYHKili0GUhalrsl9x3G3VeWpmmrHHfgFCMO8ImQokxMUWibwCTDhQQQyP0qY809mui+tiHShfgtQkufwKw75h+RDZTiaIFkFKnS09Y1Cx1BPAhQAdPEzoHLrBGtTw6C1ziVIvu9GjWY37gZzvZFM4fFMCGtlo1MxUhXqJic9RUb1WyqQNDxlSuJPRIzmzdt5atDF01sQ1yg9UH1XQHgejgGE6TsX6SaFQhawNFm3ZtVPRbN0H5zn+2+R9XA0KnOWCnFYhaWoJKU8uVj0AEAW6d/VKepilC662O63Qw/wDr5zmrRu7n0sGozg1I9CKwYAqQQdxBuD3xp6U4fsPlpVwrfZOcvTTbVD3dHdOocm+X+HxdlJFOr7jHRj91unsOswm47mqlBrWOovbvJaliQSwyVP3ijefvjp7d85ztvktVwzfaC6E+jUGqnv6D1HWdmelGTRvcEAg7wRcHtB3zSs9jMKzhozhmHwzOwSmpZmNgoFyTOk8m+Q4pUzzjFalQoXZGyMoVgwpq4116SOjdxmrw2zaNIM6pTpixLMFVbKBqSQN2l5yLlbyhfFYgulRkpjSkg09AEjMRf1mIJN+zoklpqzohN13ljojqvKDlDRwGFaqct1FqdMEDM1vRUAbh8AJwSiKu0cTUxGJa49Z2Y5UA6Fzeyo3ADU7hG9t1mKBS7OzsAMx+A6zaWW1sIKVIYe9lopnrEe1UYaju1Ufwn3p7UdVc4cXFU5ZF9y429yD5gBaiUtUD3pE5gjbmOYAkdBOtum2kl8mdoNiabYOt6VfDqXoVGOtSgPWpnS7FdLdXZrJ+krHVQUelbNRxAQEi9lOGpnI3FTZwR0zK4jGHD1qOKpCxplaire/2betTJ6QPTS/UZ6TjmjY8MPWdKopo0z7M6oxU2f1TeYnCI6CoiNUVgHGQrmysAylVawbQ7r+MoRi8K5I57m2uQVqo1MgjeCbEA98+XKLP2lLEKaul+TN/V7aEArroQDa+mZb6A/rhadg8a9IG7saIUkVbuXpBQPQrOFAObXKNTuGtxe2bZAfWmyVB9x1fyBvKrH7ENiGXQ7wRoe6e1HETpaPVHFjez6OM+KDtP/v19QsNy/xVEU+dph6VYB6au3p5CSFZiAQl7aa3+Mv9mcv6de2VSWB1pVlzMpHuVRc+PjMWwejky2PNsrU+cQVAjL6pGbWwsPRvbTUSBshuZa59I3uTcgknUm1t/fO9VqVTifAeAxmHekL/AE1R1LHbWwmJutZMlQaE7iD1mZDamwAjZ6D34FTZv7yBtXaiVCGS4a1j1jt/W+RF2i3GcVWSUrb9T7+EpSdNSs1zT/ZOp7bYejWH81tP5h0dskc7TPsrKWtis28SOta27dMKZ0OiaMc37o8I4joNyiZv65+rmGuO/VzNKZh0DXUdpZd2kvNi8sDSNm1Q7xw6xMJSBNE1FYHKbOlyGUHcw19IdnXwMRTx/wCrmekahyyoQqJo6Zt9EqLnpkEHXTrnP9p0TrH8Dtmwy3NuFzDr1FbWackzFOi4aEPZHKdsEVyUxdmYNUylmIOWyjoA14E69U7FsHbYxOHWqBY6K63vlO69+no7iD0TkHNizKRdWFmXjwI4MOg9vGafkrtlMNcMWfPe6hbCx3jU2Atp0zohOLjZnxcVhKkauaKunyOlExJaNYasGRSDe6qb9o+YMWZDnas7AMF4kmFeCC/rkI42ZL9qmEdqGasCz5TY9hzZSxZBiaigi4zJha1rjpGp0mPNKjVxVRQjpVFJHrUqb3pMz0lqpmUnNuc+lZrANqTZZd0cTzmIoKdzGuluOfC1haVONdPr9NabWfF06dSoRvtRwRp0KXZzilzxOX3RJZ2djL3MJyn2xXqc4K5F15uyC+SmpFWyID0WsSd5JJJMzmFa419pfMaj4W75o/pHrnn6mYm5GHUXJ9RMLSI39dQ+MyeEqeiOIP53kmtDtwUrSa6EkN+rRatI9WqM2hHHfuv0QucnnY+hn5G25O/SPiMNZXPPUhplcnMB91947DcToWzfpIwVVbtUNJulaikeDKCp8ZwgVo4uInm6fIzKEZ7nR+Xf0iCsn1fC3NPTnKhGXnLeyoOuW+pvv00tvwZxzdKyOK8XY2/RkcDop5YKyLDk5hfrO0sJTOgz5j2J6ZJ7ll/y22GwWpTpqajmrzleohVwtIJdboDnW511FjcWJkD6Ogf2krL61PD4h16mWmxHnNrtDF0sVTotWvQxbYdMSr4cVCUV2Zcy5AXT0l1ChhrqF3nop6RPi4t/3myJtbDHFUsUya/+XxCW4BWR93SA3lM1tbYdWnhKLVUKipzyprqVGRxcb19J338Zs8Ti6WDQVXd6+carRzBqzPzbKh1ygMfTJsd9lAF5kOUvLKti1Zaqogo1FyJTvZM6VMy5zqx9Fbnje1ppNPVHNZrclcj+Xtahh1pnLUVRlUPe6hbiwIO7dobyJtfaRxNV6rAKzkEhdwsAvT1CZjZFfRh94+YEs1rT51aLzNH73s10u5jUsszVmN4lmXURulynrpotVwOGYkeB0jtWpeVmJo3moJcTGMct4Mtf9tqp0qLTftXKfFLfCR6nKKmd9Mr2MCPAgSj5slrL3nhJtLZdP2nF+tan/tUzpWHi9T8/PtOtTeVMlHb1Pg3l84obcp9fgPnIGKwNP2Xp+LL/ANyiQfqf3k/GsvhokXbFfp5F8ds0+J8Ig7Vp+95GVeKwhOWzJ6oH+8QbrjjGDgm4p/m0/wDVHh4l/rFbki6/atP3vIwjtSn73kZVUcIbNqnq/vKfEfejBwZ4p/mU/wDVL4eJH2xW5L39y7/alP3vI/KLXbFP3vIyp2fsxnqKLpa4uecpiwvqdTwjWLwgDnKwykkqSyglbmxNo7iJ5vtaq+C9/c0K7epD2vIx9eU9Ie0fwmZOlh9Rdl3j2vlFNhwSTnQX62/JZe5iYfadXkjXDlfRHvn+Uf6oteXdIbqbnvUfOY+nQUX+0U6dAfp093rj9LBoUJzE210TgOLESqlEw+0Kr5Hffo35VnGYdmKBFpsKa+lmZrjMxJsOI07ZrDiBxE5R9FtXm8Ba2rVahPdlX8psPrre6fGaSOWcszuzTc+OIiTiBxmdGKbh5iEcW3DzlsZMkdpj3vjD/a33m8/nM221VHE9wiDt63HwX5T0MmnXbvNvTq3J5qrTqWN9ysM3T7paVXKRKy7QXE0abMlGtSqZFB5wUKZVUYJvNNlzel1yoq7fBBBvbcfV3eEOp9JtdKIpWpsUUpSrlAayIehKns6eHRaZZGiJ9J+1krY5xT9SmFpgj2iihSx46i3YBM5gsYiWzKT3qR+BhY95kOtVLEk7zEXgKVtjX4HlRSBsSoXg2EoW/pvL2m+zMQBmFFWPTTdsM1/4W9HynO8FQzMC3qg69fVNNV2sjCzU1PaoP5ShX3RdYj6O0cZsPX06BUXMv+bSuP6ZR47kdiqVyaRdR7dK1Ve05dR3gRhalIHMqlDxQsh/pIlhhuU+Ipepiqn+Jlq+bC/nMuCZ7xr1I9TPgeW/q7eEMCaqpy15w/8AiMPhqxG51L0Kw7KikmQ8RtvD71pE/wDp1lV/DE0ClT8StPJwfA64YuPzIs/omF9q01Pt0cQnjSJ/KbDFbMU4rAOxK012cDUcex9Vrq57ySFHW0wvJ7lPhqePwNWlSekUr/bZqgqLkeyegcoIFi17zoHLzFtQw31Y2UtVrgMN64UPzjG/QNUt2TUV8Nmcteou9coD2zKi4qlRZ0tTrGuqAPUFRaNJGRFaqGuzEorE7iCNN041zg5ms17F6u47wFB3np/3nlOofRltYYmgVAt9XxaFVA3Ua6Cn8abdmgnMuUmGNA8w3rKzFgegk6DT7i0z3zaSS0OZXvqQNm1bZu785OGIlXgzv7vzkjPOepG8mffwlZxpRXvcmHESJisV0Df8IipWsOuOYTDg6ky06fFmMXjWlljuDCYmooAFiB0Ecd9jvkpcehNmGU+H9j5dslph1Av+u6M1cOLekNd/Z1Ce+XkfJVThLX3zI+KRLXJa3QVQMO85x5XleaNP94e9D+RMtcBgFZt7Ktjcrrr0Cx3w8fspEUsXU8LrkY9lrgmebrJSyvc7I4GU6fex0XX3+CuxNJDltUA9HpR/ebqjH1dP3q/hqf6YdSshtfNoLDdxJ/OI+z4t4D5z0zdDjydUOpRWzfaDd7r+8OqMmiv7wfhb5QwyWIuderv4xOVOJ8P7y3JkfNeY7hwqm+a/8rbxqPMRus4ZizPcnecu+EMg6T4D5wgE4t4D5xczk6rzFU1S/rN0+wOHW0ASn7z/AIF/1wZqf3vIQlqIPZJ/m/tFxlXNDic2B7ZBP3V3ePES4wKgr6CkAC4DHMXI1F7ADIDru9Ii3G1ZgRncKlIEno1Jt0nU28Z1XZeyMNXoFcJmNVVBq0aoAxAZfay21U7vR0F9OrLZuMVZtasqOSeMZMHTF9SXY6m5JdtT4S3/AGk/Hz/vKcYN8KBQq2SpTADKTuO/sO/eIPrnWPGaRh7l0Npv7xHfaEdrP77eJ+cqUq3/ALGL7L+B/ITRDHmpxt4QvR6vCOmrf3e+IJHV4wBsoOAjDYRT7I8JJI7POIK9fxgpEbADgPONtgeqTsp/V4214IQDhf1+hC5qTTeJL9cFIRXrgvJZiSBAIueHz0fKCNtT64AkVz1eE63s4namDz5ueqU1WnXw4OWrzYBtWpm/pNc7tNQRrcX5EbjojuD2nVouHpO9NxuZCVI7CNZGZaOpbE2O2xalXF84HwzoUWnUD0qzVMwdKeRlsSCCCwOgJOm6cx2xtFq9Z6jm7OzMx4sxubdVz5QbS29XxDZq1V6jWsGd2cgcBc6DslfeCbEinXK7rddwD8RJNHbLr7FJu2kn5ASvzQAzLinuj1jWnH+LZdU+UPGjRPYrL8DHBtymfWw69z2+KmUN4YmO6hyOhY6uvm/0n+C//alH9ww7Kq/IQ/2jSPsVfxAj4yhDmHzhjuo9fM142pxt/ivQu622CBalTI62tp/KJXvUqMbspJ4kyJmPXAGPXNRpxjsjyrYmrW/m9OXAkHDsfY8xGzgW93+oRvnG64fPNxM2c1hf7Pfh/UsH1FuH9QiOdbiYDVbifGBYcGz26vxCMVEymxHnDzHifGEVJglhN+qDN1Dz+cPmjFcyYJZknAbUqUj9lZSdLhQzHq1vLvDvXZkqNWqLUTVWWylOxgJQUaZBuCQeq4k6jiWHTftJg0lY1u0dvVq7B6zZnCqmbKFJC7iQoAv1yKuLPE+LfOUybS4qI59dv0ebfOLGmXS7Q4n+r+8UdodOYGUYxvVF/W+r4SkIYaHmEj6wrSluSSRCzSPrBrBB8vE5o1BIB0mJyxIh2goZSJ5uKAh5YII5qFzB6o6AYZWAMGgYXMmSIIKRuZg5jqkm8STAI5w/VB9W6hH79UUD1eZgEb6t1CF9Xku3V8YNYBF5iGKHVJQvFi/DygEQUYoYYSSbjjBngDIwYhjACPZ7RS1+qUg0uzOyH+zOtY+Kxiuc4gyAjrszrWKGzrdIjpqDgYDWHbAGDhbcPGEKPVHxU/WUQE9kAZ5rqhFBFt2fGFl/V4KBRxBhWEULncB4wPRI9kQLibjj5wieuEVPAQCkeHlBBswoIJQCHaFBIAwIeWHBABlgvBBADh2hwQABod4IIKGbQ+bEEEALmRFrhgeMOCCCmwgELm1HRBBIUGUcIXMwQQUbyQw5ggggtBfh4SSqD3R+u6CCaA39WB6Ik4ThBBAEinrDeh1mCCQCRhr9PxgOCPH4wQQBOS0AhwQAxFBhwHhBBIBS1Rw+EJip4+UEEAbNuuBW7fGCCCn/2Q==
 
Last edited:
dice-istock-de.jpg


I've been reading various things on this forum for years and have learned lots of useful information. This countdown has kept me going throughout these anxious days while waiting for the match results and I couldn't resist taking part. Happy to be a member now.

Lucky Number 7... just one week away from Match Day!
 
I suppose finding you're unmatched could be the equivalent of finding your wife's head in a box. But probably not. Besides, the 'box' day is in 3 days, not 7.

Ooops. Um, spoiler alert. Sorry.

I :love: this movie. The ending of knocking out two deadly sins at once was brilliant.

johnnydrama said:
Seven.jpg


What's in the box(/envelope)? Haha, thinking about this, could be incredibly relevant to match day...

(If you haven't seen this, serious spoiler alert.)

http://www.youtube.com/watch?v=1giVzxyoclE
Click to expand...
 
Last edited:
sunset823 said:
I suppose finding you're unmatched could be the equivalent of finding your wife's head in a box. But probably not. Besides, the 'box' day is in 3 days, not 7.

Ooops. Um, spoiler alert. Sorry.

I :love: this movie. The ending of knocking out two deadly sins at once was brilliant.
Click to expand...

There are a couple of programs I put on my match list that in retrospect I would prefer a severed head to on match day...
 
Mortal_Lessons said:
Interesting article on the NRMP system-it appears it only takes 10 minutes to run the algorithm.

http://www.readwriteweb.com/hack/2012/03/nervous-medical-students-await.php
Click to expand...

It should take a lot less. And is it seriously written in Visual Basic? Yikes.

With some of my 4th year downtime I wrote a quick script simulating the main match.

It took 12 seconds on my MacBook Air with the conditions in that article (27000 applicants, 4000 programs, programs have 2-12 spots for an average of 6, each applicant "interviews" at 20 programs and ranks them all, each program ranks every applicant who "interviews").

Adding in prelim positions should just be like running the whole match a second time with fewer applicants and programs.

Adding in the couple's match shouldn't add too much complexity either.

It should run in less than a minute on any reasonably powered machine/smartphone.
 
johnnydrama said:
It should take a lot less. And is it seriously written in Visual Basic? Yikes.

With some of my 4th year downtime I wrote a quick script simulating the main match.

It took 12 seconds on my MacBook Air with the conditions in that article (27000 applicants, 4000 programs, programs have 2-12 spots for an average of 6, each applicant "interviews" at 20 programs and ranks them all, each program ranks every applicant who "interviews").

Adding in prelim positions should just be like running the whole match a second time with fewer applicants and programs.

Adding in the couple's match shouldn't add too much complexity either.

It should run in less than a minute on any reasonably powered machine/smartphone.
Click to expand...


Yea. 10 minutes is actually pretty long for this kind of thing when you think about it. I wonder why they chose visual basic (& fortran, for god's sake)...
 
They wouldn't be able to justify us paying hundreds/thousands of dollars on the process if they just ran it a couple times (to make sure there weren't errors) in 10 minutes and reported the results?
 
Aesculapius said:
They wouldn't be able to justify us paying hundreds/thousands of dollars on the process if they just ran it a couple times (to make sure there weren't errors) in 10 minutes and reported the results?
Click to expand...

There's no point in running the algorithm more than once - it comes up with a unique solution.
 
High-Five.jpeg
 
johnnydrama said:
That was what I said... there is a unique solution. Doesn't matter how many times you run it, you'll always find the same one.
Click to expand...

But that's even more reason to run it more than once to make sure that regardless of the way data is analyzed or the order in which data was entered, the solution is indeed unique.... So if on the third run the algorithm comes up with some different answer, then you know there is a glitch....

Having said that, I do agree that it is ridiculous that we have to wait so long; All this checks and re-checks could've been done in 2-3 days (1wk max).... I highly doubt NRMP spent every hour of every day since Feb 22 checking the data....
 
Several things delay the match results.

First, some people are idiots and try to muck with their lists at the last moment. Inevitably, someone ends up decertifying their list and not recertifying it before the deadline (computer crashes, network problems, etc). The NRMP will often talk to these people, and might consider allowing them to recertify their last list. This takes some time.

Second, they need to wait for CaRMs to complete. Anyone who matches in CaRMs is withdrawn from the NRMP. CaRMs ROL deadline is virtually the same (Feb 21 this year), so that process needs to complete first, which further delays the process.

Third, as mentioned above, the NRMP wants time to ensure that it's done correctly. I bet they do run the algorithm multiple times and ensure that they get the same answer. I bet they also hand check some of it to make sure it all looks good.

Then, they need to get everything ready for SOAP/Scramble. This involves updating everyone's ERAS records (with whether you're SOAP eligible). This is all new this year, so I'm surprised they didn't work in extra time to deal with it.

Could it be done faster? Probably. But it would increase the chances of a mistake. I'd rather have a week delay and ensure it's done correctly -- just look at the mess created by the screwed up AUA match a few years ago.
 
Hey all have been following all your threads.
Thank you for sharing all the info.
I tried logging onto the nrmp website with my AaMC ID and password and for some reason it's not letting me logg on. Am unable to change the username password also. Last logged on a few weeks ago for certifying ROL. Wondering if anyone else is having the same problem?
Panicked ms4
 
^ Logged in now just fine. Go to bed. Try again tomorrow. Call the NRMP if it still doesn't work on Monday. Breathe.
 
I'm able to log in. When was the last time you logged into it?
 
bala565 said:
But that's even more reason to run it more than once to make sure that regardless of the way data is analyzed or the order in which data was entered, the solution is indeed unique.... So if on the third run the algorithm comes up with some different answer, then you know there is a glitch.
Click to expand...

Not really. Only thing that could catch would be a mistake in the implementation of the algorithm - that should have been troubleshooted well before the match.

Unless you change the input, the output won't change, and you won't catch input errors with repeated runs.
 
johnnydrama said:
Not really. Only thing that could catch would be a mistake in the implementation of the algorithm - that should have been troubleshooted well before the match.

Unless you change the input, the output won't change, and you won't catch input errors with repeated runs.
Click to expand...

Exactly. :thumbup:

One or two competent computer scientists should have this working automatically. They could test the program/algorithm all year long. When it comes time to run it for real, it shouldn't take any time at all. Then it would be set for all the following years unless the algorithm changed.
 
Nn81 said:
Hey all have been following all your threads.
Thank you for sharing all the info.
I tried logging onto the nrmp website with my AaMC ID and password and for some reason it's not letting me logg on. Am unable to change the username password also. Last logged on a few weeks ago for certifying ROL. Wondering if anyone else is having the same problem?
Panicked ms4
Click to expand...

tried logging in this morning and had the same issue. password and everything 100% correct. worked after the 3rd or 4th try.

very weird.
 
nismodrifter said:
tried logging in this morning and had the same issue. password and everything 100% correct. worked after the 3rd or 4th try.

very weird.
Click to expand...

flintstones_recordplayer.jpg


Stop logging in multiple times, there's nothing worth checking now. Imagine that the NRMP computer system is really being handled by one of those bird machines from the Flintstones - it probably isn't far from the truth.
 
johnnydrama said:
there is a unique solution. Doesn't matter how many times you run it, you'll always find the same one.
Click to expand...
Only if you run it in the same order everytime.

What's weird is that if you read Roth and Peranson's original algorithm article (http://www.bepress.com/cgi/viewcontent.cgi?article=1016&context=wilson), it states small variations in the matches if programs and applicants are processed in a different order. Meaning that I could have potentially go to a different program if they processed it in a different order (although quite low...I think they said somewhere around 2-3 program differences in filling).

I would hope they would run the program in every conceivable order to optimize number of matches (computation should still only take a day or so) so that the least # of slot go unfilled; but I expect they run it in the same order every year to have "consistency".

I actually don't care as long as I match, just thought order entry effecting the match was interesting.
 
cofcemt said:
Only if you run it in the same order everytime.

What's weird is that if you read Roth and Peranson's original algorithm article (http://www.bepress.com/cgi/viewcontent.cgi?article=1016&context=wilson), it states small variations in the matches if programs and applicants are processed in a different order. Meaning that I could have potentially go to a different program if they processed it in a different order (although quite low...I think they said somewhere around 2-3 program differences in filling).

I would hope they would run the program in every conceivable order to optimize number of matches (computation should still only take a day or so) so that the least # of slot go unfilled; but I expect they run it in the same order every year to have "consistency".

I actually don't care as long as I match, just thought order entry effecting the match was interesting.
Click to expand...

I remember reading that before. It isn't readily apparent to me why the order matters. Does anyone have a test case showing this effect?
 
rad0nc said:
I remember reading that before. It isn't readily apparent to me why the order matters. Does anyone have a test case showing this effect?
Click to expand...

I think it is a minimal effect as there are multiple stable "marriages" that can happen for some applicants. So if you enter everyone in different orders, you may end up with a few people at different positions or not matching.

Edit: At least that's how I understand it. I may just be misinterpreting the article.
 
Last edited:
cofcemt said:
Only if you run it in the same order everytime.

What's weird is that if you read Roth and Peranson's original algorithm article (http://www.bepress.com/cgi/viewcontent.cgi?article=1016&context=wilson), it states small variations in the matches if programs and applicants are processed in a different order. Meaning that I could have potentially go to a different program if they processed it in a different order (although quite low...I think they said somewhere around 2-3 program differences in filling).

I would hope they would run the program in every conceivable order to optimize number of matches (computation should still only take a day or so) so that the least # of slot go unfilled; but I expect they run it in the same order every year to have "consistency".

I actually don't care as long as I match, just thought order entry effecting the match was interesting.
Click to expand...

I believe when that article refers to stability, it means stability of rank list choices, not algorithm outcomes. With couples' matching, there may be no optimal strategy for a program when it comes to ranking candidates (as opposed to just ranking in order of individual applicant strength). Once all preferences are locked in though, there's still only one solution.

Just think about it - if there were multiple possible outcomes, how would the NRMP choose which one to use? That wouldn't work and if some initialization parameter like order mattered, people would sue them over unfavorable results.

Also, running in every conceivable order - you've just gone from a 10 second computation to one that would take much longer than the time from the big bang to the heat death of the universe.

EDIT: Found the "small changes" part of the article you mentioned - it's referring to changing the algorithm, not rerunning the algorithm. The previous algorithm did have differences when you changed the order.

The order of processing couples vs individual applicants apparently causes a few changes, but not many (the actual order of the applicants doesn't matter, and they decided to always run couples last).
 
Last edited by a moderator:
You must log in or register to reply here.

Similar threads

psychtb
  • Question
Couples Match - “No Match” Code Questions
Replies
9
Views
3K
NotAProgDirector
NotAProgDirector
H
Did not match ortho, advice for prelim surgery year?
Replies
3
Views
2K
j4pac
j4pac
H
Chances of matching if I'm restricted to H1b?
Replies
8
Views
2K
himalayan
H
A
Questioning a previous NRMP match result
Replies
11
Views
3K
NotAProgDirector
NotAProgDirector
A
  • Question
Prelim intern reapplying for match, how do you update an LOR from previous cycle from someone who saw you as a student and not a resident?
Replies
8
Views
2K
NotAProgDirector
NotAProgDirector
Top