Why track anything
Memory is a poor record. After a few weeks you will recall the dramatic games and forget the ordinary ones, and your sense of whether you are improving will be coloured by your last session. A simple log gives you something to check your impressions against, and it helps you notice patterns such as which map, time of day or role goes badly.
The danger is the opposite one: tracking so much that you stop. A sheet that takes ten minutes per match will be abandoned within a week. The principle is to log the fewest numbers that answer your questions.
Start with a question
Before choosing columns, decide what you want to learn. Examples:
- Am I improving at a particular skill?
- Do I play worse at certain times, or after a number of games in a row?
- Which matchups, maps or roles cause me trouble?
- Is a new strategy working?
Each question suggests a few numbers. If a column does not help answer any question, leave it out.
A minimal set of columns
A workable sheet can start with these:
| Column | Why it helps |
|---|---|
| Date and game number | Lets you spot trends and session effects |
| Mode, map or matchup | Shows where results differ |
| Role or character | Compares performance by role |
| Result (win/loss) | The basic outcome |
| One habit score | Tracks the thing you are working on |
| Mood or energy (1 to 3) | Reveals tilt and fatigue patterns |
| One-line note | Captures context that numbers miss |
That is seven columns, which is about the most you can fill in during the gap between matches. You can add more later if a specific question needs it.
Outcome numbers and behaviour numbers
It helps to separate two kinds of data.
Outcome numbers describe how the game went: win or loss, kills, deaths, damage, score, rank change. They are easy to find but heavily affected by teammates, opponents and luck.
Behaviour numbers describe what you did: how often you completed an opening, used an ability on time, took a fight alone, or checked the map. You usually have to record these yourself, perhaps from a replay review, but they reflect your choices much more directly. They also tie neatly to your improvement goals.
If you can only afford one extra column, pick a behaviour number connected to your current focus and count it per match.
Per-game stats that are often worth logging
What counts depends on the game, but some general candidates:
- Shooters: deaths from a specific situation, first-contact results, how many trade kills you took or enabled, how often you made a round economy decision on plan.
- MOBAs and team games: deaths before a key timing, objective participation, how well you hit economy or last hitting benchmarks at a fixed minute (such as a chosen time mark), ultimate economy used or wasted.
- Fighting games: how often you converted an opening into your planned combo, drops, number of times you were hit by a particular move.
- Card and strategy games: mulligan choices, turn at which you reached a milestone, whether you were ahead or behind early.
Prefer measures with a clear definition. If you cannot say exactly what counts, the number will drift from game to game.
Keeping the sheet maintainable
- Fill it in immediately. Doing it after each match, or at the end of a short session, is much easier than reconstructing from memory.
- Use dropdowns and simple codes. Typing less makes it more likely you will continue.
- Add a weekly summary. A few formulas, such as the count of games, the number of wins and a running average of your habit score, are enough. Wins divided by games gives win rate.
- Review on a schedule. Look at the sheet once a week, not after every game.
- Archive and restart. When a sheet gets long, start a new tab per month so it stays quick to read.
Reading what the numbers say
A few rules help avoid false conclusions.
Use totals over many games. A win rate over ten games is very noisy; over a hundred it is steadier, though still uncertain. Judging changes by sample size works the same way here as elsewhere, and results move around more than most players expect, partly because of random clumping and streaks.
Compare like with like. Split by role, mode or opponent level before drawing conclusions. A single average hides differences.
Beware of averages hiding variation. A steady average could mean consistent play or an even mix of very good and very poor games.
Do not read causes from correlations too quickly. If you play worse late at night, that might be fatigue, or it might be that you play ranked at that time against stronger opposition. Treat patterns as ideas to test, not proof.
Watch for rating effects. Rank changes can be affected by rating volatility, hidden adjustments and uncertain placements, so they are not a clean measure of skill.
Letting the sheet change your play, not your mood
Tracking can backfire if every number becomes a verdict. Wins and losses include luck, so a bad stretch on the sheet is not proof that you have got worse. Looking at process separates how well you played from how things turned out, which supports a healthier mental game.
Also check that the stat does not push you into counterproductive habits. If you track kills, you may start avoiding risky but useful plays. If a number starts to distort how you play, change or drop it.
Getting started
Make a new sheet with the seven columns above. Add a single behaviour score tied to what you are practising. Fill it in for the next twenty matches without analysing anything. After that, look for the first patterns and decide whether each column earned its place. Keep what helps, delete what does not, and let the sheet grow only when a real question demands it.