Certifying a Yardage Sheet

In the test facility where I worked before this, nobody was permitted to report a measurement without also reporting three other things: what instrument took it, when that instrument was last checked against a reference, and what the environment was doing at the time. A figure arriving without those was not treated as a weak measurement. It was treated as not a measurement at all, and it went back.

It took me about a year of golf in this valley to work out that the entire equipment industry operates the other way round. Distances get quoted constantly, confidently, and almost never with any of that attached. A player will tell you they carry a seven iron a hundred and sixty yards with the certainty of somebody quoting their own height, and if you ask when that was established, under what conditions, with which ball, the answer is generally a shrug and a story about a par three in 2019.

This note is about the middle bit: how a morning of raw readings becomes something you can defend. It is our house method, it is not the only reasonable one, and where we have made a judgement call I have tried to say so rather than presenting it as physics.

The Header Comes First, and It Is Not Optional

Six fields go across the top of every sheet before a club is taken out of a bag. Nothing below them means anything without them.

• Date and start time. Time of day matters more than people expect, because a bay that has been open since seven in the morning is not thermally identical to one that has been running since noon in August.
• Bay temperature. Ours is held near seventy-four degrees, but held near is not the same as fixed at, and the actual reading goes down.
• Station pressure. Recorded rather than assumed from the elevation, because weather moves it around by enough to matter over a season.
• Ball model. The specific ball, not the category. A premium urethane ball and a firm distance ball are different instruments as far as spin is concerned.
• Radar unit identity. We run two. They agree closely and they are not identical, so the sheet says which one produced these figures.
• Date of last reference check. Ours run every Monday before opening. A sheet produced on a Friday states that its instrument was verified four days earlier.

None of this is difficult and none of it takes long. It exists so that a reading taken today can be compared to a reading taken in 2019, or to a reading taken by somebody else somewhere else, without either party having to guess at the difference.

Warm-Up Is Not Data

The first shots of any session are excluded by rule, not by judgement. The client warms up as long as they like, and none of it is recorded. When they say they are ready, recording starts.

This sounds obvious and is routinely got wrong, usually by accident. A monitor left running through a warm-up accumulates rows that will quietly drag an average around later, and by the time anybody is looking at a spreadsheet the distinction between warm-up and recorded shots has been lost. Deciding at the start which rows are evidence is far easier than trying to decide it afterwards.

Blocks, Not Sequences

Shots are recorded in blocks of eight to twelve with the same club and the same configuration, then the configuration changes and a new block starts. Blocks are the unit of analysis; an individual shot is not.

The reason is that a single golf shot carries very little information about a player. The useful properties, the centre and the spread, only exist at the level of a group. A player who has hit one drive has not told you anything about their driving; a player who has hit ten has told you a great deal, including how much they vary, which is frequently the more important half.

Ten is roughly where the returns start flattening for most amateurs. Below about seven, the centre moves around too much to be worth quoting. Above about fourteen, fatigue starts widening the spread and you are measuring endurance.

What Gets Discarded, and the Rule We Apply

This is the part that makes people uncomfortable, and it should. Any process that lets the person analysing the data decide which data to remove is a process that can produce whatever answer that person wants. The protection is not to avoid excluding shots, because a topped ball genuinely does not describe your seven iron. The protection is to fix the rule in advance and show the working.

Ours has three parts.

• Called shots. If the player says "no" before or at impact, it is out. Their judgement, not ours, and it has to be immediate rather than after seeing the number.
• Contact failures. Anything that catches the mat before the ball, anything off the extreme heel or toe far outside that player's own established pattern. Judged on strike location, which is measured, rather than on the resulting distance.
• Nothing else. In particular, a shot is never excluded for being short, long, or inconveniently placed. If it was struck acceptably it counts, however much it spoils the block.

Every excluded shot stays printed on the sheet in grey, with a one-word reason beside it. You can count them. If a block shows five exclusions in twelve, that is itself a finding, and it is usually a more important one than whatever the surviving seven averaged.

The Median, Not the Mean

Once a block is clean, the figure we report as the carry centre is the median rather than the arithmetic average.

This is a small technical choice with a large practical effect. An average is dragged by extremes; a median is not. If nine of your ten seven irons carry between a hundred and fifty and a hundred and fifty-eight yards, and the tenth catches a gust of enthusiasm and goes a hundred and seventy-two, the average reports something you achieved once and the median reports something you achieve half the time.

The second figure is the one you want when you are standing over a shot with water short. A yardage card built on averages is a card that will leave you short of the green slightly more often than it leaves you long, and the miss it produces is the expensive one.

Beside the median we print the spread: the range covering the middle bulk of the block, which is the honest expression of how much you vary. A player with a five-yard spread and a player with a twenty-two-yard spread can share an identical median and are playing entirely different games.

Then the Second Column

Everything so far produces the valley column: measured figures, in this bay, under the conditions in the header. That column is honest and it is local.

The travel column beside it is modelled. It takes the measured launch conditions and asks what the same shot would do in cooler, denser air nearer sea level. It is an estimate, the sheet labels it an estimate in the column heading, and it carries a stated basis so that anybody can see what assumptions produced it.

We include it because roughly two in five of our clients play a meaningful share of their golf somewhere other than the Coachella Valley, and a card built purely from readings taken at four hundred and sixty-six feet in warm, dry air will systematically over-promise the moment it leaves. Better to hand somebody two numbers with their provenance attached than one number that is right in one place and quietly wrong everywhere else.

What Certified Does and Does Not Mean

It does not mean the numbers are guaranteed. Golf is not that kind of activity and anybody selling you certainty about it is selling you something else.

It means every figure on the page can be traced back: to a block, to a set of exclusions with reasons attached, to a stated instrument with a stated verification date, and to a recorded environment. If you come back in four years and the numbers have moved, we can tell you something useful about why, because both sheets carry their own conditions and both are on file.

That traceability is the entire product. The distances are just what it is made of.