Masters crews of very different ages race each other, so a governing body hands the older crew a time allowance. Eight bodies have published one and no two of them agree — not on the shape of the curve, not on which boats share a constant, not even on whether the allowance scales with the distance rowed or with the time the race takes. Switch the systems on below, put your crew in a boat, and see how far apart they land — or add the crew you are racing and see what start you owe each other under each one.
Eight seats, not nine: no handicap system counts the coxswain's age.
Put an age in the boat to compare the systems.
Every one is transcribed from the body’s own published rules. Eight of the ten nations that have consistently put boats on an Olympic rowing podium are covered either directly or by a system their federation adopts; the rest — Romania, New Zealand, Italy, Croatia, Poland — run masters racing in World Rowing age categories without publishing a handicap of their own.
Add the second crew and the page switches to answering that instead. What each crew is allowed is measured from an imaginary 27-year-old who is not in your race; the number that gets paid out on the water is the difference between the two, and that is what the box on the chart and the list under it show once there are two boats.
Both crews race the same event, so they share the boat, the crew type and the distance — only the ages differ. Whichever crew is older starts ahead under every system here, because every curve rises with age; what the systems disagree about is by how much, and that disagreement is usually a larger number than crews expect. A regatta uses one system and one only, so treat the spread as the cost of the choice rather than as a range you get to pick from.
Because that is the difference between the systems, not a drawing choice. USRowing, World Rowing and the Dutch factors are defined at every age: a formula, a fitted curve, and a factor printed per year of age. Everyone else publishes one number per age category, so between 65 and 69 their allowance does not move at all and then jumps at 70. Smoothing those would invent allowances at ages nobody published, so they are drawn as the staircases they are.
It also means the category boundaries are where a crew has most to gain or lose: under a staircase system a crew averaging 69.9 and one averaging 70.0 are handed noticeably different races.
The systems split almost evenly, and this is the deepest disagreement between them. USRowing, World Rowing, Australia, Germany and Henley all publish an allowance per 1,000m and scale it with the course — five seconds over 1,000m is forty over 8,000m. British Rowing, Scotland and the Netherlands make it a share of how long the race actually takes instead, which is why they need a Standard Time for the boat before they can say anything at all.
The two are not the same thing. A slow boat and a fast boat over the same course get the same allowance under the first group and different allowances under the second. For the time-based systems this page uses each body’s own published standard times for 1,000m and scales them with the distance, so a number far from 1,000m is indicative. World Rowing’s linear scaling is also an extension on our part: their model was fitted to 1,000m racing, which is the only distance their masters regatta rows.
Because a parabola leaves its vertex flat and an exponential never does. Just above 27 USRowing is still almost at zero while World Rowing has already started charging for the years, so World Rowing leads. The square then catches it somewhere in the thirties and stays ahead through the middle of a masters career — the stretch most people reading this are in. Then the exponential takes over for good in the sixties or seventies and runs away with it. The women’s 8+ is the one boat that never crosses at all: World Rowing’s fit for it leads the whole way.
Three of the eight are gender-neutral by construction. USRowing has one formula and three constants for everybody. Germany prints a single set of increments. The Dutch women’s factors are exactly 0.9 times the men’s at every age, so the 0.9 cancels out of the ratio a handicap is made of and the age allowance comes out identical. The other five fitted or tabulated men, women and mixed crews separately, so switching the crew moves those lines and leaves the first three exactly where they were.
Because their bodies published nothing for that combination, and drawing a flat zero would be a lie rather than a gap. Henley has no coxless four in any of its three boat groups. Australia runs no mixed single or pair. World Rowing fitted mixed crews only for the 2x, 4x and 8+. Scotland and the Netherlands publish men’s and women’s tables but no mixed ones. A note under the toggles names whichever switched-on system has gone quiet.
Age coverage is ragged too. Scotland stops at 74 and its line simply flattens there, which is what its table does. Germany and the Netherlands are the only two that keep going past 85, splitting the top into three-year bands to 89.
Because it moves as you type, and a number that only changes at whole years makes the seats above it feel dead. A regatta would round: your crew enters the category its truncated average falls in, and most of these systems then read the handicap straight off that category. So a crew averaging 49.8 races C rather than D, and the staircase systems hand it a category C allowance — which is why the lines here step where they do, and why the chart is worth reading either side of a boundary.
Every rower’s, and only the rowers’. No body counts the coxswain towards the crew average, and USRowing puts no age limit on the seat at all. That is why an eight has eight boxes here and not nine.
Each table was checked against something its own publisher states before it went in. Germany’s two worked examples reproduce exactly. Henley’s three reproduce exactly. Australia’s triangular matrices are internally consistent as difference tables. Every one of the 91 cells in Scotland’s published table is its flat percentage of that row’s standard time. The Dutch model recovers all sixteen columns of their start-difference table to within 1.1s, on a table printed in whole seconds.
Two caveats worth knowing. World Rowing’s paper prints a coefficient for the men’s 4− that does not reproduce the men’s 4− column of its own table, running about 8% light; this widget uses the coefficient that matches the published table, since the table is what a regatta would actually use. And Scotland’s regatta table is published as an image this page could not read, so its line is built from the head-race table, which is the same percentages and is what Scotland’s own instruction to scale proportionately implies.