"1y 1mo." That's the countdown in the top right corner of PRs by age group. In thirteen months I stop being an M45-49 and become an M50-54. Every road PR I have gets a new set of numbers to beat.
The page is a matrix. Distances down the side, age groups across the top. Every cell behind me holds my best result in that band. Every cell ahead of me holds the time I'd need to match it.
This is the second of a short series on reading the RaceRecords dashboards, after Targets to Beat. Same approach. My own numbers, because a dashboard is only worth explaining when the data is real.
What does the PRs by age group page show?
One row per distance you've raced, one column per age band, and your best result in every cell you've filled. For me that's eight road distances, 5K to marathon, across the eight bands.
The bands are the ones the sport actually uses. U20, a senior block covering 20 to 34, then five-year steps: 35-39, 40-44, 45-49, 50-54, 55-59, and 60+ with no upper edge. Underneath, RaceRecords buckets every race into strict five-year groups. U20, the senior block and 60+ are aggregations of those, for display only.
My current band is boxed. Everything left of it is history and can't change. Everything right of it is a target. A dash means I never raced that distance in that band. I started racing late, so most of the grid is empty.
Two markers sit on the cells. An orange label flags my best at that distance. A blue filled pill flags my best age grade of all, a 10 mile at 82.5%. The fill does the work rather than the colour, so the two still read apart in greyscale. Any cell holding a race opens that race when you tap it, and on a phone the grid scrolls itself to your current band.
The toggle changes what a cell means
Time shows the fastest race you ran in that band. Age-graded % shows the highest-scoring one. Usually that's the same race. When it isn't, the page is telling you something a results list never will.
My 10 mile row at 40-44 is the clean example. In Time mode the cell reads 59:54. Flip the toggle and the same cell leads with 77.1%, over a 1:00:08 underneath. Fourteen seconds slower on the clock, and it's the better race. It was run four years later, against a standard that had eased in the meantime.
A second pair of tabs switches between Road and Track. Those stay apart the way they do everywhere else in the app, with their own PRs and their own tables.
Why does the same cell show a different race?
Because a five-year bracket treats everyone inside it as the same age, and the tables never did.
The 59:54 came at 40 and graded 74.9%. The 1:00:08 came at 44 and graded 77.1%. Two points better, for a race that's fourteen seconds worse. That is the five-year bracket problem turning up in my own results.
My half marathon does it twice. A 1:19:47 at 40 scores 74.8%. A 1:20:23 at 44 scores 76.7%. Same band, slower time, better performance. It happens inside my current band too. A 1:18:02 at 45 and a 1:18:31 at 46 sit 29 seconds apart on the clock. Two tenths of a point apart the other way.
So which one is my real 40-44 half marathon? Both of them, depending on the question. Time mode answers "how fast was I". Age-graded mode answers "how well did I run". On a page organised by brackets they disagree more often than you'd think. Worth knowing that the orange marker follows whichever mode you're in, so it moves too.
What are the required times in the bands ahead?
The finish time that matches your current best age grade at that distance, priced at the entry age of that band on the current tables. Alan Jones 2025 for road, WMA 2023 for track. My 10K best is 80.5%, set with a 35:26 at 45. Here's what that same grade costs later, and the same sum for every distance I've raced.
| Distance | My best age grade | 50-54 | 55-59 | 60+ |
|---|---|---|---|---|
| 5K | 75.8% | 19:16 | 20:04 | 20:57 |
| 10K | 80.5% | 36:56 | 38:33 | 40:20 |
| 15K | 82.1% | 55:17 | 57:46 | 1:00:30 |
| 10 Mile | 82.5% | 59:07 | 1:01:47 | 1:04:42 |
| Half Marathon | 79.9% | 1:21:16 | 1:25:00 | 1:29:06 |
| 25K | 81.4% | 1:34:44 | 1:39:07 | 1:43:57 |
| 30K | 69.9% | 2:13:12 | 2:19:27 | 2:26:19 |
| Marathon | 80.1% | 2:47:52 | 2:55:54 | 3:04:46 |
Every one of those cells carries a small arrow and the word "required" underneath, so it can't be mistaken for something I've actually run.
These are not predictions. Nothing here claims I'll be able to run 2:47:52 at 50. They're the price of the same performance, quoted in a currency that weakens every year.
The page stops after three bands ahead. Past that the number is arithmetic without a purpose.
One thing worth noticing. My 5K and 30K rows are empty in my current band and show no required time either. Required times only appear for bands you haven't reached. The one you're in is still open, so there's nothing to predict.
How much the standard actually moves
More than most runners expect, and not evenly.
Take that 80.5% 10K. At 45 it's a 35:26. At 50 it's 36:56, at 55 it's 38:33, at 60 it's 40:20. Almost five minutes of slack across fifteen years, for an identical quality of running.
But the ladder isn't the whole story. Each of those is an entry-age figure. Inside the 50-54 band the same 80.5% runs from 36:56 at 50 all the way to 38:13 at 54. Seventy-seven seconds, one bracket, no change in the grade. That happens because the age factors move every year, not every five. USATF Masters sets out the method.
The spread widens as you go. It's 71 seconds in my current band and 93 seconds by the early sixties. That's why the chart above draws the app's open-ended 60+ column as a 60-64 band. So the number on the page is always the hardest version of it. Turn up on your fiftieth birthday and you owe 36:56. Wait four years and you owe 38:13.
How do I use PRs by age group?
Two ways, depending on which direction I'm facing.
Looking back, it's an audit. Flip to Age-graded mode and the strongest race of each period floats to the top, which is not always the one I remember. My best 10 mile by the clock and by grade are the same race, which is reassuring. My best half by each measure is two different mornings, a year apart.
Looking forward, it's a shopping list. The 50-54 column is thirteen months away and it hands me an entry time for every distance. 36:56 for the 10K. 2:47:52 for the marathon. Both are slower than the PRs I already hold. That's the strange gift of age grading, and the same arithmetic behind what any given age grade costs at your age.
The dash in my 5K row is the one that bothers me. It isn't a target, it's an omission. Some cells are empty because the distance is hard. That one is empty because I haven't bothered.
If you've logged races, open PRs by age group and flip the toggle. Watch which cells change. Those are the races where your clock and your age disagree, and the disagreement is usually the interesting part. To check a single race without an account, the calculator grades it on the same tables.
Sources
- Alan Jones, Age-Grade Tables (road, 2025 edition, approved by the USATF Masters Long Distance Running Council) — retrieved 2026-09-09, https://github.com/AlanLyttonJones/Age-Grade-Tables
- World Masters Athletics, Age-Grading Tables (2023 edition, track) — retrieved 2026-09-09, https://world-masters-athletics.com/wma-age-grading-tables/
- USATF Masters, Age Grading (method reference) — retrieved 2026-09-09, https://usatfmasters.org/age-grading/
- RaceRecords, Methodology — retrieved 2026-09-09, https://racerecords.run/methodology