Hmm.. you're asking about two things really.
First is whether your way of breeding is efficient at producing good horses. I wouldn't use the way you've described for this reason. You're committing yourself to a minimum of 7 years from start to finish of process. And essentially you're breeding the same pairs for 3 years. Let's start with a premise Thunderspark postulated awhile back that breeding a good horse is dictated by the RNG and that I think he mentioned something like only 1:20 offspring are any good. So for each 7 years you breed 15 horses and what if all 15 are crap?
I breed 1 stallion to 5 mares. Save at the end of the 1st year. Back up the save. Then I run through the next year and have 5 two year olds to look at. If none of them are good IMO then I reload the backup save I made earlier. Run through the next year and have a look at the 5 two year olds again. You get the idea. I can breed the same pair over and over again and just check at the start of the two year old season if I want to run them through to their 3 year old season or not (in your case 4 year olds).
The second question you're asking is how to breed better horses. Let me ask. In all the time you have put in, have you formulated any ideas/hypothesis as to what might make your horses better. Have you then tested those ideas and checked to see if it works? Any good idea should cut .3s off your best times in CK within 4/5 generations IMO. If it doesn't then try something else. Don't be afraid to restart breeding from game horses. After all you do have your ttf.dat files backed up right?
Here's how my ttf.dat files look. 275 files containing 1-5 horses each. I can pull any horse I have bred from when I got the game.
Breeding from the start is the easiest way to test your breeding ideas. If something works you'll see a clear .5s drop from the earliest horses. Whereas if you're using horses you've already bred and have made some headway into cutting CK times, it can be more difficult to know if something works or not.
Breeding in SO5 is much more selective than SO4. Whereas SO4 was about how many generations you could breed once you had good lines, SO5 IMO is different.
Some things to consider.
1 Breeding across distances - what works/doesn't? Do you breed them between 2f differences i.e. 1m2f and 1m or do you breed them across huge differences e.g. 1m7f and 5f. Which works better?
2 Finding good game horses - good colts are easy to find. Good fillies are rarer than hen's teeth. How do you find good fillies?
3 How do you ensure you're finding the best horses to breed once you have good lines? Why do you test on one course and not another. Why good going? Which horses perform better - horses that like firm running on soft or horses that like soft running on firm. How does breeding a horse that likes soft ground with one that likes firm ground affect the offsprings choice off going. Do offspring run better on soft/firm if bred from soft/firm combinations instead of from parents that prefer good/good? Both combinations will give your offspring that like good going.
There are a lot of questions I have still. Some I have tested. Some I will get around to. But there are no easy answers
. You'll have to find out for yourself.
