What this calculator does
Climbing a ladder is a race between what you gain for a win and what you lose for a defeat. If those two numbers are equal, you need to win more than half your games to move at all. If they are not, the break-even point shifts, and most systems quietly set it below 50 per cent for players who are climbing.
The useful output is rarely the number of games. It is the net points per game, because that single figure tells you whether the climb is a weekend or a season.
The formula
Each game is worth your points per win multiplied by your win rate, minus your points per loss multiplied by your loss rate. That gives net = w × gain − (1 − w) × loss. The games needed are the points you want divided by that net, rounded up. Setting net to zero and solving gives the break-even win rate, loss ÷ (gain + loss), which depends only on the two point values and not on your skill.
| Term | Meaning |
|---|---|
| Net points per game | The average movement per game. The whole climb hangs on this one number. |
| Break-even win rate | The win rate at which you stay exactly where you are. |
| Points needed | The gap to your target, in whatever unit the ladder uses. |
The inputs explained
| Field | What to enter |
|---|---|
| Points needed | How far you want to climb, in ladder points. Count across division boundaries if your system resets points at each one. |
| Points gained per win | Points for a win. Many systems vary this by how far your rating sits from your hidden rating. |
| Points lost per loss | Points for a loss, entered as a positive number. |
| Win rate (%) | Your win rate over a meaningful sample. Twenty games is not a meaningful sample. |
| Minutes per game | Average minutes per game including queue time, which is what turns games into an evening. |
When to use it
Working out whether a climb is realistic
At 55 per cent with 22 up and 18 down, a 400-point climb takes 100 games and 50 hours. At 50 per cent the same climb takes 200 games. A five-point swing in win rate halves the work, which is why improvement matters more than volume.
Finding the break-even point
With unequal point values the break-even win rate is not 50 per cent. Gaining 22 and losing 18 means you hold steady at 45 per cent, so anything above that climbs eventually. Equal values put break-even at exactly half.
Deciding whether to keep playing tonight
The time figure is the honest one. A hundred games at 30 minutes is 50 hours, which is not an evening, and seeing that before you start is better than discovering it at game forty.
Worked examples
Every figure in the tables below is produced by this page’s own calculator at build time, so the numbers and the tool always agree. Select any row to load that scenario.
How much does win rate change the climb?
Only the win rate changes down the rows.
| Win rate | Games to reach the target | Net points per game | Time at that pace |
|---|---|---|---|
| 45% | never, you tread water at this win rate | 0.00 | n/a |
| 48% | 334 | 1.20 | 167 h 0 min |
| 50% | 200 | 2.00 | 100 h 0 min |
| 55% | 100 | 4.00 | 50 h 0 min |
| 60% | 67 | 6.00 | 33 h 30 min |
| 70% | 41 | 10.00 | 20 h 30 min |
What if losses cost more than wins pay?
The gain is fixed and the loss gets steeper.
| Points lost per loss | Games to reach the target | Net points per game | Win rate needed to hold steady |
|---|---|---|---|
| 12 | 60 | 6.70 | 35.3% |
| 15 | 75 | 5.35 | 40.5% |
| 18 | 100 | 4.00 | 45.0% |
| 22 | 182 | 2.20 | 50.0% |
| 26 | 1,000 | 0.40 | 54.2% |
| 30 | never, you lose ground at this win rate | -1.40 | 57.7% |
Questions
Why is break-even not 50 per cent?
Because it depends entirely on the two point values, not on skill. Break-even is loss divided by the sum of both, so gaining more than you lose puts it below 50 and the reverse puts it above.
My points per win keep changing. What do I enter?
An average over recent games. Most ladders adjust gains based on the gap between your visible rating and the system's private estimate of you, so the figure drifts as you climb.
Is a 55 per cent win rate good?
It is enough to climb steadily in almost any system, which is the only thing that matters. Sustained win rates much above 60 usually mean you are rated well below where you belong, and that corrects itself.
Why does my win rate sit at 50 per cent no matter how much I improve?
Because matchmaking is designed to produce that. As you improve, you face stronger opponents until you are winning half again. The rating moves; the win rate does not. There is more on this in the piece on Elo.
How many games before my win rate means anything?
Hundreds. Over twenty games a coin flip regularly comes up 13 to 7, so a small sample cannot separate a 52 per cent player from a 58 per cent one.
For the rating system underneath most ladders, see Elo rating change. For what the other scoreboard ratios actually measure, there is the piece on measuring players.