Keno guide
Medusa Keno: the eight-spot ticket on an 80-number board
Medusa is just what this page is called; no myth or rule comes with it. The page covers an eight-number ticket in classic 80/20 keno and a simulation run under a sample schedule.
The distribution for eight spots
An eight-spot ticket on 80 balls with 20 drawn has an average catch of 2. The peak is two hits at 32.81%, followed by one hit at 26.65% and three hits at 21.48%. The board has C(80, 8) = 28,987,537,150 possible tickets.
| Hits | Probability | As odds |
|---|---|---|
| 8 | 0.0004% | 1 in 230,114.6 |
| 7 | 0.0160% | 1 in 6,232.3 |
| 6 | 0.2367% | 1 in 422.5 |
| 5 | 1.8303% | 1 in 54.6 |
| 4 | 8.1504% | 1 in 12.3 |
| 3 | 21.4786% | 1 in 4.7 |
| 2 | 32.8146% | 1 in 3.0 |
| 1 | 26.6464% | 1 in 3.8 |
| 0 | 8.8266% | 1 in 11.3 |
Six of eight, derived
Six of the eight spots must be among the drawn 20, and two among the undrawn 60. C(20, 6) = 38,760. C(60, 2) = 1,770. Multiply: 68,605,200. Divide by 28,987,537,150 to get 0.2367%, which is 1 in 422.5. The eight-of-eight catch is C(20, 8) = 125,970 tickets, or 1 in 230,115.
Compare the 40-ball board with 10 drawn: there, six of eight is 1 in 841.9 and eight of eight is 1 in 1,708,993. The traditional board makes high catches easier by a wide margin at this pick count, which is why pay schedules for the two formats cannot be swapped.
A simulated look at eight-spot sessions
The simulation below uses an invented schedule for 80/20 pick 8: 2x for 4 hits, 12x for 5, 100x for 6, 1,000x for 7 and 10,000x for 8, with nothing for 3 or fewer. Its average return is 82.32% of stakes. Eight sessions of 100 games each were generated with a fixed seed.
| Session | Rounds won | Net (units) | Lowest point | Longest losing run |
|---|---|---|---|---|
| 1 | 16 | -48.00 | -49.00 | 15 |
| 2 | 11 | -68.00 | -68.00 | 25 |
| 3 | 10 | 1,926.00 | -11.00 | 31 |
| 4 | 10 | -50.00 | -58.00 | 23 |
| 5 | 9 | -62.00 | -62.00 | 24 |
| 6 | 15 | -20.00 | -30.00 | 17 |
| 7 | 15 | -20.00 | -31.00 | 11 |
| 8 | 11 | -28.00 | -28.00 | 19 |
Seeded pseudo-random simulation (Python random.Random, seed 8020), 8 sessions of 100 one-unit rounds. Average session result 203.75 units. This is a simulation, not a record of real play; a different seed gives different numbers.
The figures come from a pseudo-random generator, not from real play. A game counts as won only with four or more hits, which has a chance of 10.23%, so long losing runs fill the right-hand column. One session in this seed caught one of the rare top lines and finished far ahead, which pulls the average of all eight sessions above zero. That is the trap of small samples: a single rare line can hide a schedule that returns less than its stakes. Seven of the eight sessions lost.
Points to keep in mind
- Four or more hits on 80/20 pick 8 is roughly a one-in-9.8 event per game.
- Most of the average return in a schedule like this sits in lines that a 100-game session usually never reaches.
- A seeded simulation shows variety, not a forecast. Change the seed and every row changes.