Does Boston Round Your Qualifying Time?

Published August 7, 2026 · Estimates, updated each registration cycle

Short answer: no. The Boston Athletic Association does not round your marathon time in either direction and does not hand out a grace second. Your qualifying time is your official net time, recorded to the second, and it has to land at or under your age group standard. A man aged 30 on race day has a standard of 2:55:00. He qualifies at 2:55:00 exactly. He does not qualify at 2:55:01. One second is the whole difference between an application and no application.

The 59 second grace period ended in 2013

This question keeps coming back because the rule really did exist. Through the 2012 race the BAA allowed a 59 second window on top of every standard, so a runner facing 3:10:00 could finish in 3:10:59 and still apply. That window was removed for the 2013 race, in the same change that cut five minutes off the standards themselves. Nothing has brought it back in the thirteen years since. If a training partner, a forum thread or an old calculator still talks about the extra 59 seconds, it is quoting a rule that has been dead for over a decade. Treat the published standard as a hard line and read the current standards table rather than an older one.

Net time, not gun time

The BAA is explicit that qualifying times are based on the official submitted net time, also called chip time. That is the clock that starts when you cross the mat at the start line, not the gun that fired for the front of the field. For anyone starting in a later wave of a big race that gap can run to several minutes, and it is the reason a result that looks short of the standard on the gun clock can still be a clean qualifier.

The catch is that the race has to publish a net time in its official results. Some smaller races only post gun times, and in that case gun time is what you submit and what gets verified. Check the results format before you enter a race you plan to qualify at, not after. The note on the certified course requirement covers what else makes a result usable for Boston.

Seconds decide the cutoff too, not just eligibility

Meeting the standard only earns the right to apply, and a qualifying time is not an entry. Because the race is oversubscribed, the BAA ranks every verified applicant by how far under their own age group standard they ran, then accepts from the largest margin downward until the qualifier places are full. The last accepted margin becomes that year's cutoff, and that ranking is counted in seconds as well. The 2026 race cut at 4:34 under the standard, which means an applicant sitting 4:34 under got in and one a single second behind that did not. Our central estimate for the 2027 race is about 5:15 under, which is our estimate and not a BAA figure. The buffer guide explains how to read your own margin against numbers like these.

The new 2027 draw does not soften the line

For the September 2026 registration week the BAA has added a random selection component. After the cutoff is set, qualifiers who beat their standard but finished outside the cutoff go into a draw for roughly 1,000 extra places. It is genuinely good news for thin buffers, but it changes none of the arithmetic above. You have to have beaten your standard to be in the draw at all, so a time one second over leaves you out of the acceptance list and out of the lottery as well. The selection explainer walks through the full order of operations.

What to do with this

Two practical habits follow. First, never aim at the standard. Set goal pace at the standard minus a deliberate buffer, so that a bad mile or a long tangent costs you seconds you can afford rather than the whole application. Second, when you read a race result, read the seconds column and check whether the time shown is net or gun before you celebrate. Then put your age, gender and finish time into the cutoff predictor to see your exact buffer to the second, your rough odds against the 2027 estimate, and whether that same buffer would have been accepted in each of the last thirteen Boston cycles.