It is possible. A real-life example is provided by Ken Ross. It involves the batting average of the baseball players Derek Jeter and David Justice during the years 1995 and 1996:
| Player | 1995 | 1996 | Combined |
| Derek Jeter | 12/48 = 0.250 | 183/582 = 0.314 | 195/630 = 0.310 |
| David Justice | 104/411 = 0.253 | 45/140 = 0.321 | 149/551 = 0.270 |
Strange but true. To see what is going on, have a look at this made-up example that makes things clear.
| Player | 1995 | 1996 | Combined |
| Player A | 0/1 | 98/99 | 98/100 = 98% |
| Player B | 1/99 | 1/1 | 2/100 = 2% |
What is happening here is that player A is scoring well in a large number of cases, whereas player B's averages are high when he plays little and low when he plays a lot.