April 21, 2011 - Disclaimer: I'm stating in advance that most of the following (tailwind benefit excepted) are just my personal ideas based on an uncharacteristically (for me) hasty look at a relatively small number of course-to-course comparison stats, an inordinately extensive personal history with running with tailwinds in places like Waco, Texas (where there are miles-long stretches of straight road and where high winds occur often enough for anyone to gauge very accurately what kind of advantage big tailwinds give), and an awful lot of assuming, gut intuition, and guessing about what the runners might have been feeling in the early pack in Boston. Some of it is probably off the mark, so I will make no attempt to defend it if anyone calls it crazy talk, but indisputable science which gives a precise conversion for Boston's 2:03:02 to a windless day or on a different course probably isn't coming anytime soon, either, so I think my guess is as good as most.

In the end, we are all guessing about where the final result from the men's race at the 2011 Boston Marathon ranks among the all-time great marathon performances. But this you can take to the bank: The race was wind-assisted at ground level and significantly faster because of it. As the saying goes: Common sense ... so rare it's a superpower.

The misleading thing about an aiding wind is that runners subconsciously or inadvertently often throw common sense out the window and believe they simply made a huge breakthrough, as it's very hard to realize how much it's helping you unless you've done a lot of timed runs in very windy conditions. Even if you are isolated, you are not likely to notice the true strength of the wind at your back if you're running 12+ mph and you will greatly underestimate its benefit ... unless you try turning around and running into it for awhile. Hence, you see comments like you did from 2011 Boston winner, Geoffrey Mutai, about the wind: "It was at our back, but it wasn't such a big wind."

Don't think a 20 mph tailwind makes a substantial difference? On a day which is that windy, go find a straight-shot course, run 10 miles out-and-back at a strong clip (with and against the wind) or maybe try 6 x 1 mile hard out-and-back with 3-minute rest periods. YMMV slightly depending on your weight and the surface area your chest and back expose to the wind, but you'll probably find that much consistent wind assistance is worth between 10 and 12 seconds per mile, meaning you'll run 20 to 25 seconds per mile slower when you turn around and run into that wind.

If the forecasted wind wasn't absolutely guaranteed to make a huge difference at Boston, I never would have mentioned the likelihood of a world-best time. I didn't throw out the possibility of a 2:03 before the race because I had no doubt the top few runners were so much better than those in any other marathon ever contested; nor did I mention it because I knew this big secret that Boston's course, long viewed as one of toughest majors in the world, is really nothing but a piece-of-cake, freewheeling downhill and this just happened to be the year several runners were going to prove it. No, I took one look at the weather forecast. It's the wind, stupid.

How Difficult Is The Boston Course?
So how hard is Boston's course? Well, despite the net elevation loss of 450 feet, the terrain includes so many progressively-taxing, rhythm-breaking ups and grueling, quad-blasting downs in the second half that the virtually constant downhill in the first several miles - unrecognizable as "pounding" at the time - is actually considered by many runners to be a wolf in sheep's clothing, exacerbating the distress that comes later. Plenty of discomfort - along with probable glycogen depletion and possible cramping and hobbling - awaits the runners who are unprepared (or unsuited) for the terrain or who let the seemingly easy going in the early miles fool them into thinking it couldn't hurt to bank some extra time. Any of the dream-thwarting problems a normal marathon can spring on a runner in the latter stages are sharply magnified at Boston. The unique layout seeks out and tests every aspect of course-specific preparation, execution and determination, and its exacting nature unquestionably outweighs the large elevation drop. Boston presents a demanding course for those who get it right, a nightmare for those who get it wrong.

Despite all of its challenges and "sly old dog" surprises, the course hasn't traditionally been considered brutal, unfair or even the kind of course that was low on the list if a personal best was the goal. In the heyday of US marathoning (the 1970s through the early 1980s), runners didn't shy away from Boston, fearing it as "too hard" of a course on which to run a decent time. Reasonably demanding physically and challenging to execute well, like a historic marathon was expected to be - yes. Too slow to be a good option - no. Why has this perception changed? One reason is because in those glory years, there was a tremendously larger number of Americans who regularly ran marathons at a more competitive level, and almost all of them pointed toward Boston as The Race, from the elites dreaming of victory or a sub-2:10 to the guys who barely qualified for the race, and everyone in between. Americans rarely ran in such large numbers on the courses that are now known to be faster, so quite a few of the top marks (and many personal bests among the masses) each year came from Boston, except in years with miserable weather, like 1976 (95 big degrees for that death march). But in the intervening years, as more runners have sought out the races that guarantee the fastest times with the least discomfort, becoming spoiled by the relatively easy Chicagos and Rotterdams of the marathon world, Boston has acquired the stigma of being a bad course for a good time. Sure, it's noticeably tougher than any of the notoriously easy courses, as it can really bang up even a durable runner, and a marathoner who runs Chicago every year isn't likely to set a PR at Boston, but it's a fair course that isn't horribly slow, as long as you've trained properly for it.

My first estimate - admittedly no more than a semi-scientific one based on paltry data - is that under ideal, windless weather conditions for all courses, Boston is about 75 seconds slower than London, which is in turn about 30-40 seconds slower than Berlin (site of the official world record). Given the disparity in the quality of fields between some of these races, the presence of paid pacesetters in races like London and Berlin (which is bound to result in additionally faster top marks, discounting relative course difficulties, but could also possibly be responsible for more elites blowing up - who knows?), and taking into account variable weather from year to year, it would take an extremely detailed analysis (which would probably be restricted to small-sample statistics) to come up with course comparison figures having anywhere near universally-accepted accuracy.

How Much Difference Did The Wind Make?
What about that much-debated wind? Based on comparably wind-aided Bostons (1975, 1979, 1983, 1994), it looks like people have run between 2 and 4 minutes faster vis-a-vis years in which the Boston weather was similar temperature-wise but sans wind. Furthermore, personal experience with hundreds of timed tempo runs of up to 10 miles along straight highway access roads with all sorts of wind conditions has resulted in my running 10-12 seconds per mile faster when I've had a tailwind in the neighborhood of 20 mph. I'll be conservative and restrict that advantage to exactly 10 seconds per mile for the Boston terrain, some of which is not as out in the open and uniformly wind-blasted as a Texas highway. I'll make a further SWAG and surmise that the tight pack in the early stages of this year's Boston race slightly reduced the perceived effect of the tailwind for some of the runners - at least for those who weren't at the very caboose of the pack, who might have been holding back to avoid clipping the heels of the runners in front of them (a phenomenon I've also noticed in windy road races). Just arbitrarily, let's stick with the tailwind providing the full 10-seconds-per-mile advantage for the runners for the second half of the race. This results in a second half that is 131 seconds (2:11) faster than it would be with no tailwind. The effect of the tailwind in the first half might be a little dicier to figure, but it certainly wasn't negligible. Time-for-place figures for the mid-pack runners indicate that they ran roughly 2:00 faster than last year for the entire race, possibly meaning that the wind assistance while sheltered in a group was worth in the neighborhood of one minute for the first half (maybe a little less for the elites, who were on the course for less time). Then there's the question of whether racing with a tailwind for a very long time might result in additional heating and negate some of the wind's benefit. That one is absolutely, positively too iffy to address. So (once again) I'll be annoyingly arbitrary and throw out what I think is a reasonable round number of a guess (this is also based on results in previous windy Bostons) - the overall effect of the tailwind was to confer (drum roll) about a 3-minute improvement for the finishing times of the lead runners.

If all that speculating is as plausible as I think it is, what does it all mean? It means without the favorable wind, Geoffrey Mutai would have likely run in the very high 2:05s or low 2:06s in Boston - a similar performance to Cheruiyot's in 2010 (probably better, since 2010's race featured a crosswind with a minor assisting component from the West). By my course comparison estimate (- 75 seconds), this performance would be somewhere in the middle to high 2:04s on London's course. So I'm saying add 1:45 or thereabouts to the 2011 Boston times (for the top elites) to get equivalent 2011 London times (that gives Ryan Hall a 2:06-something, close to a PR effort). Yep, I can hear the railbirds now: "That's just great ... you're calling the winning Boston and London performances virtually equal and if you put them at Berlin, they'd be very close to Haile Gebrselassie's 2:03:59 world record. Way to be evasive and wishy-washy, Kell-Dog." But is it really a cop-out? I think it reflects the state of elite marathoning in the world today that the winners of two separate major races can run 2:04 on London's course and the equivalent of that on a wind-aided day across the big pond on Boston's trickier, quad-busting course. The best in the world these days can run 2:04s on loop courses if they have to and perhaps dip under the official world record on the easiest loop courses. Ryan Hall can and has run 2:06 in London. It adds up.

Then again, maybe I'm subconsciously fudging my course comparisons and other guesswork so the final numbers will conveniently jive with what I think the winners of these races are capable of running on a record-eligible course - 2:04. Maybe I don't want to admit the possibilities that this year's winning Boston time was actually superior to the WR or that it might have been worth "only" a 2:05-2:06 at London. But to reiterate, I'm just guessing and babbling like everybody else, and it's all but certain that even the nerdiest analysis of all the variables would still involve enough unknowns and require enough assumptions that ballpark figures and more babbling are unavoidably as good as we're going to get for now.

JK Confirms He's A Genius, But Didn't Predict 2:03
Alas, I must admit I didn't have the cojones to actually predict a 2:03 at Boston. If pressed for a predicted time, I would have weaseled out and gone with a mid-2:04 (still a huge CR), mostly because of the uncertainty of anyone volunteering to get the early pace going (but it turned out someone did). I just said I wouldn't be shocked at a 2:03, and I'm not one bit now that it's happened. Think about it - there was a 2:05-high last year with a marginally favorable crosswind. With a projected 20 mph direct tailwind, that could easily be more than 2 minutes faster, so if the runners got after it and put up a similar 2:05-type of effort, of course they'd run 2:03. It didn't take a genius to say that. A genius did say it, but it didn't take a genius. I actually thought they could run sub-2:03, but it's hard to believe that strongly enough to mention it as a possibility. When you say 2:02 or anything faster out loud, it just seems a little too far out of the box - tough to wrap your mind around - even if you know there'll be a big wind behind the runners and know how much of a difference it can make. But they came awfully close to the 2:02s. Obviously, it wasn't so wild a notion after all.

Mutai Over Mutai
Who would have won if all the main players had run in London? I think we all know Mutai would have beaten Mutai. Would one of the runners who didn't emerge victorious in either of these races have had his day? How fast would the winner have run? How about in Berlin or Rotterdam? World record? Those questions involve even more guessing, since late-race tactics (and how the courses suit the runners) would come into play. But we do know that in two back-to-back days, we saw two great races and two great winners, two course records and the fastest marathon ever run on any course under any conditions. And - oh, yes - it's not outrageous to think the winner in London and/or the top two in Boston could have bettered the WR had they put up their efforts in ideal weather in Berlin, although I think an equivalent race at that venue would have resulted in a near miss of about 2:04:10-2:04:20. That's the level at which the superstars of marathoning appear to be at the moment.

Back To Top