Are you a lucky charm? How to calculate your team's record when you attend

Every stand has one: the supporter convinced the team never wins when they are there, and the one who swears they have not seen a home defeat in years. Both are usually wrong in the same way — they remember the matches that fit the story. The good news is that this is one superstition you can actually test.

Step 1: list the matches you attended

You need every competitive match of your team you attended, with the result and whether it was at home or away. Friendlies go in a separate pile — clubs do not play them to win, so they would blur the comparison. If you only have the matches you remember, stop here: a list built from memory is precisely the biased sample you are trying to escape.

Step 2: turn results into points per match

Win-percentage alone ignores draws, so use points, the way the league table does: 3 for a win, 1 for a draw, 0 for a defeat. Add them up and divide by the number of matches.

Example: you attended 12 matches — 7 wins, 3 draws, 2 defeats. That is 21 + 3 = 24 points, and 24 ÷ 12 = 2.00 points per match.

Step 3: compare like with like

Now calculate the same figure for your team over the same seasons. This is where most lucky-charm claims fall apart, because of one bias: home advantage. If most of your matches were home games, compare them with your team's home record, not its overall record. Home teams win more often everywhere, so a supporter who only goes to home games will always look lucky against the full season.

In our example, suppose the team took 1.80 points per match at home over those seasons. You are 0.20 above. Lucky charm?

Step 4: is the gap real or just luck?

Football results are noisy. A single match is worth 0, 1 or 3 points, so the typical spread around the average is roughly 1.3 points per match. That noise shrinks as your number of matches grows, and a simple rule of thumb gives the size of gap you need before it means anything — about 2.6 divided by the square root of the number of matches:

• 12 matches: a gap smaller than about 0.75 points per match is noise.
• 25 matches: about 0.5.
• 50 matches: about 0.37.
• 100 matches: about 0.26.

So the 0.20 in our example, over 12 matches, is well within luck. It would take well over a hundred matches before a gap that small started to look like a pattern. This is not a statistics exam — the rule is approximate — but it is the right order of magnitude, and it is why nearly every lucky-charm claim evaporates when it is checked.

Why it is still worth calculating

Because the number is fun even when it proves nothing, and because doing it properly settles the argument in the pub. It also tends to surface things you had forgotten: the unbeaten run you were part of, the season you saw almost every defeat, the ground where your team has never won in front of you.

All it takes is a dated list of your matches with results. Keep it in a spreadsheet, or let an app such as Kopio build it for you by scanning your tickets — either way, the record is only as trustworthy as its completeness.