How the marathon guides are built

The checkpoint splits on each guide are effort-even, not pace-even. Given a goal finish time and the course elevation profile, each mile's clock pace is set so that effort stays constant: slower on climbs, faster on descents, with the total still equal to the goal time. The cost of each grade comes from Minetti et al.'s metabolic cost-of-running model, the basis of modern grade-adjusted pace calculations.

That is a model, not a promise. It assumes even effort and ignores wind, heat, and how well your legs handle sustained descent, which on a steep course is often the deciding factor. Treat the tables as a starting plan, then adjust for the day you get.

Where a course holds a USA Track & Field certificate, the certification panel transcribes it directly: the certification code, the date it took effect, the year it lapses, the start and finish elevations (converted from the meters printed on the certificate), the net drop in meters per kilometer, the start/finish separation as a percentage of race distance, and the course configuration. These are the figures the certification process actually measures, which is why they can differ from a GPS watch or a mapping site. Those measure a route you ran; the certificate measures the route that was certified.

Older certificates are scanned documents with no data table. For those we record the code and its dates but not elevation figures, because reading elevations off a profile graph would not be verification. Where the certificate is silent, so are we.

Certifications lapse on December 31 of their expiry year. Recertification is routine, but if a guide shows a lapsed or soon-to-expire certificate and you are targeting a later edition, confirm with the organizer that a current certificate is in place before counting on a time for Boston.

The weather panel shows what the last five editions were actually like at the start line, not what the month is usually like. The observations come from the Open-Meteo historical reanalysis archive (ERA5), sampled from 7 to 11am local time at the start line, so the table reflects race morning rather than a daily average.

Dew point gets its own column because it is the number that costs pace. Past about 60°F, sweat stops evaporating efficiently and most runners slow down no matter what the thermometer says. Reanalysis data is modeled from station observations, so treat it as a close record of race morning rather than an official one.

A Boston-qualifying time must be run on a certified course, which is why the certification code on each guide is load-bearing rather than decorative. For the 2027 Boston Marathon, the B.A.A. also applies a net-downhill adjustment: because runners of comparable fitness post faster times on steep point-to-point descents, times from those courses are penalized, and the steepest courses are excluded outright. The adjustment applies to net elevation loss (start elevation minus finish elevation), not to cumulative descent.

Each guide's Boston verdict is computed from that course's own certified drop against these bands. The standards, qualifying window (opened 2025-09-13), and registration dates (2026-09-14 to 2026-09-18) were read from the B.A.A.'s qualifying page on 2026-07-26. Meeting the standard earns the right to apply, not entry: in recent oversubscribed years a cutoff was applied (2024: 5:29 under the standard), (2025: 6:51 under the standard), (2026: 4:34 under the standard). Work out your own target with the Boston qualifier calculator.

Elevation profiles on the guides are illustrative of each course's established character, not surveyed GPX, so absolute values matter less than the shape. Logistics like start times, field caps, and cutoffs move year to year; each guide notes when its logistics were last checked, but confirm specifics on the official race site before booking travel. And the training science behind the plans themselves, VDOT pacing, Riegel prediction, and the safety framework, lives on the About page.