Four League Points Systems and How to Choose Yours
Four systems cover almost every league running today. They disagree about one question: how much a finish is worth when the field is twice the size it was last week.
Players optimize whatever you count, so the question is not which system is fairest but which behavior you want more of. Each one below is worked on the same two nights, a quiet 12 player game and a busy 30 player game. Check the arithmetic as you go.
Flat placement points
A fixed table by finishing position, identical every night: 10 for first, 8 for second, 7 for third, on down the table. Field size never enters it.
The arithmetic. Win the 12 player game, score 10. Finish fourth of 30, score 6. Beating 11 people pays 10, beating 26 pays 6.
Who it rewards. Consistent high finishes, and anyone who wants to work out their own score without asking.
The failure mode. A win in an 8 player game is worth as much as a win in a 40 player game. If players pick their nights, the arithmetic says pick the thin ones, and nobody has a reason to bring a friend.
Field size scaling
Points are a placement multiplier times the number of players. This is the Chiprally default: 10, 8, 7, 6, 5, 4, 3, 2, 1 for the top nine, and everyone below that scores zero. Nine is the length of the default table and nothing more than that. The table is a setting, and a league drawing 40 a night should make it longer.
The arithmetic. In the 30 player game, first scores 10 x 30 = 300 and fourth scores 6 x 30 = 180. In the 12 player game, first scores 10 x 12 = 120.
Under field size scaling, fourth in a field of thirty is worth more than a win in a field of twelve, and that is the argument for it in one line.
The defense is that fourth place outlasted 26 players and the winner outlasted 11. If you think a win should beat a fourth whatever the field, do not adopt this.
Who it rewards. Players who turn up when the room is full and go deep. It also gives everyone a stake in recruiting.
The failure mode. One enormous night can decide a season. A holiday game drawing 60 pays 10 x 60 = 600 for the win, while a regular finishing fifth on a 12 player night scores 5 x 12 = 60, so 24 nights of steady work is 1,440. One evening is 42 percent of a full season.
Fixes: cap the counted field at 30, or count only each player's best N results.
Percentage of the field beaten
Points are the share of the field you outlasted: players beaten divided by field size minus one, times 100. A win is always 100.
The arithmetic. Fourth of 12 is 8 of 11 beaten, or 72.7. Fourth of 30 is 26 of 29, or 89.7.
Who it rewards. Deep runs in big fields, without the volume of raw scaling, and a win still counts as a win anywhere.
The failure mode. The scale is linear, and that shows at both ends. At the top, a win is 100 and fourth is 90, so a season of near misses catches a season of wins. At the bottom, 20th of 30 still scores 10 of 29, or 34.5, a third of a win, and over 24 games that is 828 points. Scoring only the top third of the field fixes the bottom end. The top end is the price of the system.
Placement points plus attendance
Any placement table, plus a flat number of points for turning up. That flat number is the entire design decision.
The arithmetic. Flat table plus 5 points an appearance. Player A attends all 24 games and reaches the scoring places six times, averaging 4 points: 24 plus 120 = 144. Player B plays 12 games with three wins (30), two seconds (16) and two thirds (14): 60 plus 60 = 120. A wins by 24. Now drop the bonus to 2: A has 72, B has 84, and B is champion. One constant decided the season.
Who it rewards. The backbone of the league. If 30 people are on the roster and 11 turn up, this pays for the thing you are short of.
The failure mode. Attendance points are uncontested, so the title drifts toward whoever has the emptiest calendar. A player joining in March is out of it before they have settled into a seat, so they stop coming. Counting each player's best 16 of 24 results fixes both.
The four side by side
| System | 1st of 30 | 4th of 30 | 1st of 12 | Failure mode |
|---|---|---|---|---|
| Flat placement | 10 | 6 | 10 | A win in a short field counts the same as a win in a full one |
| Field size scaled | 300 | 180 | 120 | One huge night can outweigh a month of steady turnout |
| Percentage of field beaten | 100 | 90 | 100 | Everyone near the top sits within a few points of each other |
| Flat placement plus attendance | 15 | 11 | 15 | Turning up and going out early still scores, and somebody will say so |
Bounties, entry fees, and things that should not be points
Flat add ons weigh differently at different field sizes. A 50 point bounty under field scaling is 42 percent of a win on a 12 player night (50 of 120) and 17 percent on a 30 player night (50 of 300), so flat rewards are loudest on quiet nights.
Do not score on what a player takes away from the night. The take already scales with the field, so the standings end up counting field size twice, and on a 30 player night 27 people go home with nothing on the board and no reason to come back. An entry fee is venue income, and it belongs in the fee record rather than in the points table.
Keep hospitality outside the points table too. Whoever brings the food gets a free drink or first pick of seat, which costs nothing on the board and leaves the standings a record of finishing order and nothing else.
How to choose
- Name one behavior you want more of. More bodies, bigger fields, or better play. A system rewarding all three rewards none clearly.
- Measure your field spread. Divide last season's largest field by the smallest. Near 1, flat points cost nothing. The 12 against 30 example above is 2.5, wide enough that scaling flips a win and a fourth.
- Set the attendance constant deliberately, as the 5 against 2 example shows.
- Decide the counting rule and the tie break now. Best 16 of 24 is part of your scoring system whether you write it down or not.
- Apply the bar test. A player should be able to work out what tonight was worth without a spreadsheet.
Backtest before you announce
Recompute last season under the system you are considering. If the same champion wins, you changed nothing. If someone wins who nobody thinks earned it, you found the failure mode in August rather than in December.
Run this on your own league
Take a season you have already played, put the finishing order in, and switch systems. If the title changes hands, you have learned more in a minute than a season of arguing about it. The points simulator does exactly that. Chiprally itself runs the field size model, and the simulator is where the other three get their hearing.