I read something a while back about building lay systems like this and it recommended trying to find something that had a negative ROI (or LOT if you want to call it that) of around -30%. Its doable but even some of those I found dont make great lay systems anyway.

I prefer to base my lay systems on the horses that are over-bet by punters and therefore their odds arent true representations of chance. Look for things that a lot of people will see as a good sign it's a winner and then check the stats to see if it backs up. More often then not it wont.

I also think with laying as it can get volatile sometimes that have a system that bets a smaller number per day but more days is better. If you were laying 50 horses a day andd had a "bad day" your bank roll takes a big hit. For example I now have 4 lay systems in action using real money. They've been live for about 3 weeks and without actually digging out the stats overall I'd say I've had just over 100 bets (so only 33/week).