Algorithm - factoring sample size weighted average

Question

I've created an algorithm used to measure season performance across multiple soccer teams. Each team plays a different # of games. I have assigned weights to time periods to calculate a weighted average so that more of the calculation is based on time periods with more games, therefore more consistent / larger sample. My issue is that each team plays a different number of games so I'm having trouble determining how to factor the number of games into the equation to accurately compare teams with different # of games. i.e. If team A plays 100 games and Team B plays 10, the calculation will be more stable for team A since the sample size is larger, therefore I want that to be factored into the weighted average in some way. How do I factor sample size (#of games) into an equation that will affect the weighted average when comparing teams with different number of games / sample size?

i.e. 

A(x)+ B(y)+C(z)+D(n) = Weighted %

Time periods:

A = Goals in last 20 days

B = Goals in last 7 days

C = Goals in last 3 days

D = Goals in last day

Weight:

x = 40%

y = 25%

z = 25%

n = 10%

Answer 1

0.4A+0.25B+0.25C+0.10D=100% of points as a weighted average