Hey there !
I think posting your current set of filter can help identifying is there is any particular flaw in your methodology. Would also be interesting to have an idea on the type of league you are betting on (lot of secondary leagues or mainly first divisions/second division of big countries).
Regarding the delay between bet and actual start, it just exposes you to more movement of the odds. That is because between your bets and the actual start there is more room for new info to come in the market. Especially if you bet on secondary markets (2nd division, etc), you then expose yourself to more volatile odds movements and more space for these movements to occur. This may translate in more variance or even losses if those movements occur too often in the wrong direction.
With the current deepdive capability of the software, what I would do is to check sports that have been hurting your profitability (especially if the sample is over 500 bets for those sports). For those sports you check the last 20-30 events on which you placed a bet. You do that manually by comparing odds at placement vs final odds (there is plenty of free website reporting the final odds - I am not sure if I can provide a link regarding this forum policy) → it can give a first hint to check if you are beating the closing line most of the time or not. Until now, RebelBetting has been beating the closing line most of the time every time I did a sanity check, but that’s my just personal experience.