My third run was the 1999 Chicago Marathon. My chiptime was 3:24:37. My goal was to beat 3:30 and I am happy with the result. I ran with my sister, Barbara Stevens, and her husband, Kent Stevens. We had trouble parking and were late getting to the line, so we started relatively far back. It took 10 minutes to cross the start. My course time was 3:34:40. According to the chip I was at 0 miles at 10:03, 9 miles at 1:12:36 (6 min 57 sec/mile leg pace), 13.1 miles at 1:49:18 (8 min 57 sec/mile pace), 17 miles at 2:18:14 (7 min 25 second/mile pace), and 26.2 miles at 3:34:40 (7 min 12 sec/mile pace). The 9 mile split seems too fast going out and too slow to the half. I stopped at the half for a restroom break, which probably cost 2-4 min on the second leg. The cumulative pace is 7 min 49 second miles. I ran hard the whole way and had stride trouble only in the last couple of miles. This was the only race I have ever run where I was never out of a crowd. I placed 3109 out of 24595 finishers. I was number 2729 among the men and 584 out of 2827 finishers in the 35-39 male group. I was some where in the top 20 of the 90 or so runners from Champaign and Urbana. 3 CU runners finished in under 3 hours. The runners ahead of me seemed a sea, but I guess that the greater sea was behind. As indicated by Khalid Khannouchi's world record of 2:05:42 (4 minutes 47.8 seconds/mile), it was a good day to run. Sunny and cool. Barbara finished at 4:47(chip) and Kent finished at 4:51(chip).

Another fellow named David Brady, this one from Naperville, ran 4:03. He crossed the start 2 minutes ahead of me and I was 12 minutes past him by mile 9, so we passed each other fairly early. Several people I know also ran, for example, my friend Kevin Miller crossed the finish at 4:03:40, 11 seconds ahead of the David Brady from Naperville. I saw no one by chance, however. To give some perspective of the finish, here is a plot of the number of finishers (in 10,000s) verses time (in hours):

This plot was generated by from Chicago Tribune data and a perl script. The times are chip times. My time and Barbara's are marked by vertical lines. The slope in the linear regime is about 208 runners per minute. That is a speedy crowd. Here is a histogram of runners verses chip time. Note the peaks at 3, 4 and 5 hours, showing that people are not Poisson.

If each volunteer takes 10 seconds to remove the chip from a shoe, one needs 35 chip removers, which is about how many there were. There were at least 5 photographers at the on-course photostation. I guess that they each had to shoot 1 frame/second. I wonder if they had multiple cameras and a reloader. The finish area was a zoo, one couldn't fall down. Traffic getting out was horrendous. Hence the focus on smaller races next year.