Could use some advice regarding U/O 2.5 system

[DeadNova]

Hello everyone. I'm trying to come up with a system to try and predict U/O games. I'm taking into account: -last 4 home/away games -the average goal/game in these 4 (if most of the games are low scoring, max 3, but there's 1 with 5+ I don't take that into consideration, instead I treat it as an anomaly/error and take the next game) -average goal/game in played games home, respectively away for the away team this season -their standing/motivation for the respective match -poisson to compare the value with the ones the bookmakers offer. Been testing it for a couple of days now, and there were some big surprises. Games that were somewhat risky have won, others, on which I was absolutely sure, lost. My question is, could this be because the system isn't functioning as it should or because I was testing it mostly on end season games. Could use any and all help, advice :)

[Grex]

Re: Could use some advice regarding U/O 2.5 system I think "end-of-season" is a big factor..... I have stopped both my "Homes" and "Overs" systems for this very reason. I would suggest you back-test the system over the past few seasons and see if performance tailed-off for (say) the last 6 rounds.... and possible the first 6 rounds of each season as well :ok