Benchmarks: the board grades you can trust

On a board with tens of thousands of community-set problems, grades drift. Someone sets a problem, calls it V6 because it felt like V6 to them on a good day, and it sits there a touch soft or a touch sandbagged forever. Benchmarks exist to cut through that noise and give you a smaller set of problems whose grades you can actually trust.

What a benchmark is

A benchmark is a problem that has been singled out as a reliable marker for its grade. Rather than one setter's guess, a benchmark has been climbed by a large number of people, its grade has settled through that broad consensus, and it has been flagged as a steady reference point. When you see the benchmark tag on a board, you are looking at a problem the wider community broadly agrees on.

That matters because the rest of the database is genuinely all over the place. New problems with a handful of ascents can be wildly off, soft enough to flatter you or stiff enough to crush your confidence. Benchmarks are the opposite. They are the few hundred problems you can point to and say, with reasonable confidence, that this is what the grade feels like.

Why honest grades are worth chasing

The whole reason people train on boards is repeatability and honesty. A grade you can not trust undoes both. If you tick a soft V7, you walk away thinking you have moved up a grade when you have not, and your training drifts towards whatever feels good rather than what actually challenges you.

Benchmarks fix the reference. Because their grades are stable and the holds never move, a benchmark you climb this month means the same as one you climbed six months ago. That makes them the cleanest way to answer the only question training really asks: am I getting stronger.

How to use them

The simplest way to use benchmarks is as a ladder. Pick the benchmarks at and just above your current level and treat them as your honest test pieces:

• Project benchmarks at your limit grade, not random hard problems, so a send actually means something.
• Repeat a benchmark you have done before to feel whether it has got easier, which is real evidence of progress.
• Use a benchmark at a grade below your max as a barometer warm-up to gauge how you are climbing on the day.

Because everyone is climbing the same benchmark problems, they are also a fair point of comparison with friends and with your past self. Two climbers arguing about who is stronger can just go and try the same benchmark, which settles it faster than any conversation.

Do not chase the tag for its own sake

A word of caution. Benchmarks are useful precisely because they are reliable references, not because they are inherently better climbs. Some are brilliant movement, some are awkward and one-dimensional, set to test a specific style. If a benchmark suits you perfectly, your tick will read a little flattering; if it plays to your weakness, it will feel sandbagged even though the grade is fair.

Use them as the honest yardstick they are, spread across a few styles so no single weakness skews your read, and let them be the backbone of how you measure progress. Logging your benchmark sends over time, alongside the date and angle, turns a scattered pile of attempts into a clear line you can actually watch climb, which is exactly the kind of record Sendit keeps for you.