Lies, damned lies, and footy statistics

A series for coaches who'd rather read the game than be fooled by a number. No stats degree required — just enough to stop the numbers being used against you.

"There are three kinds of lies," the saying goes: "lies, damned lies, and statistics." Mark Twain put the line in the world's mouth — though, fittingly for a series about not trusting confident sources, even that is uncertain. He credited it to the British politician Benjamin Disraeli, and historians have never been able to prove Disraeli said any such thing. A famous quote about statistics being misleading, itself misattributed for over a century. You couldn't write it better.

Most people use the line to wave numbers away. Stats can be twisted to say anything, so why bother? But that's exactly the wrong lesson. The reason statistics can be used to mislead you is that most people can't tell a strong number from a weak one. The fix isn't to ignore the numbers — it's to understand just enough that nobody can hide behind them.

That's what this series is about.

You don't need a stats degree

Let's get this out of the way: you do not need to understand any of this to use Powercoach. The app already does the hard part — it ranks the stats, colours them by strength, flags what's solid and quietly hides what isn't. You can coach off that alone, and most people do.

But footy is drowning in numbers, and not all of them deserve your trust. Every Saturday a commentator tells you a team "won the clearances" as if that settles it. A rival coach swears by a stat that won them one game and has lost them ten. Your own eyes tell you a story at three-quarter time that the scoreboard flatly contradicts. Some of those numbers are real signal. Plenty are noise dressed up as insight.

A coach who understands a little — what makes a link strong, when a sample is too small to trust, why a stat that sounds important can be almost worthless — is a coach who can't be sold a dud. That's the whole pitch. Not to turn you into a statistician. To stop you being lied to.

What the numbers actually say about winning

Here's the fun part: when you stop taking the loud stats on faith and actually check them against thousands of real captured games, footy gets more interesting, not less. A lot of the things coaches yell about at the breaks turn out to be weak predictors of winning. A few quieter things turn out to be enormously powerful. Across this series we follow the data wherever it leads:

• What actually predicts winning — and why "win the clearances!" is one of the weakest instructions on the ground. (Process vs proxy)
• Where scores really come from — most of them start with a turnover, not a stoppage. (Where do scores come from?)
• Why field position is worth more than possession — the same turnover can be six times more valuable depending on where you win it. (Field position)
• How pressure becomes points — the full chain from effort to the scoreboard, link by link. (How pressure becomes points)
• Whether momentum is real or a trick of the mind — footy's version of basketball's famous "hot hand" debate, put to the test. (Is footy momentum real?)
• Why there's no universal formula — the platform-wide average can't tell you what wins for your team. (Find your team's winning stat)
• The principle behind every finding above — correlation isn't causation, and one stat that proves the point. (Correlation is not causation)

Every one of those findings comes from real games captured by Powercoach, and every one is spelled out in plain English. You won't need a calculator. You will, now and then, have a "well, I always assumed the opposite" moment. Good — that's the series working.

(There's also a standalone companion piece outside the series, for coaches who've specifically asked us why we don't show total metres gained on the dashboard: Metres gained: useful by source, not as a total.)

Start here

If you only read one before the rest, make it the next one: Stats made simple for coaches. We'll be honest with you up front — it's the driest read in the series, the one that feels a touch like homework. But it's the foundation everything else is built on: a friendly, jargon-free guide to the handful of ideas the rest of the series leans on — what an r-value is, why some insights stay locked until you've captured enough games, and how to tell a result worth acting on from one that's just luck. Push through those fifteen minutes and the payoff is the whole series: every article that follows — and every dashboard in the app — reads sharper, and the genuinely interesting stuff lands harder.

Then work through the rest in order, or jump to whatever's nagging you this week. The numbers are on your side. Let's make sure they stay that way.