In another thread there was a suggestion that in some of the top marathons somebody was throwing down a 4:20 mile to break the race open in the 22 mile region. I hadn't previously seen any case where this had actually happened. The two specific references that came up thereafter were to the 2004 Olympic marathon and last year's NYC marathon.
In the case of the 2004 race at Athens, Baldini strung together an entire final 12K at 4:39 pace, which is pretty awesome, but I don't find any record of a 4:20 in there. Perhaps somebody has seen more detailed splits than the ones that are available for every 5 or 10 K for that race? I couldn't find any.
In the case of last year's NYC marathon, according to the story on the NYCM website (as pointed out by an alert poster on the other thread) Hendrick Ramaala dropped a 4:21 17th mile. I'd like that number better if I could find real data reflecting it instead of just a news story, but it is a pretty staggering feat if taken at face value.
Are there other cases with official results posted to back them up? Khannouchi was interviewed a few years back and talked about his long runs that are progressive, starting very slow and ending very fast, dipping down into 4:20 pace or thereabouts. But the finishing times that I can find for him show his breakaway splits from some of his Chicago victories to be in the 4:38 range.
Ramaala's split sounds almost unbelievable, and the NYCM site doesn't have mile splits recorded. So that leaves me wondering whether it can really be accurate...is anyone aware of the provenance of this split? The half to 20M and 20M to finish splits aren't much help, but they show the average pace over those miles to be 4:54 and 4:57 respectively.
Perhaps equally of interest, is anyone aware of a similar split elsewhere in the annals of marathoning? If KK at his prime was breaking away with high 4:30's, running 15+ seconds faster seems like it must be absolute suicide...
But I'm a (very) slow guy who just likes the sport, so what would I know about that?