Bernie, as you know I'm very generous when it comes to giving advice

My suggestion to you, especially given that you've been following the nags for 45 years (as you said), is to get yourself a database. I've no doubt of two things here ..... 1) Those who have purchased a database will have become more conservative / selective with their punting as a result of them finding out what works and what doesn't, and become far more disciplined because they're aware of the run of outs that's a normal part of racing. 2) Get far more enjoyment out of this pursuit as they've got an endless supply of filters to combine and test.

The data base I've got Bernie is RaceCensus and I use it to find 1) "horses to follow". I input data that will give me a list of promising "horses to follow", and with this database you can then do a sreach on the individual horses that are produced in the list and it gives you their entire career (provided it's within the date range of the database). For example Bernie I input a couple of filters that select a lightly raced horse placed in a fairly high grade Metro race, indicating that the trainer may have seen something. The results show up as a slight POT (for the race they're entered for) but the hidden value is that it gives me a list of horses to follow (back tested), a stable if you like, and I follow these for a certain number of starts. I also have a similar system for getting a list of horses to follow which includes again a lightly raced horse with no obvious form going up in distance. This system also shows a small POT, but again it gives me a list of horses to follow. You could also look at how horses who won their first start fared for their next 10 runs !!, the options are limitless and logic can be applied.

Then there's the systems you can develop. My best system is one I developed, geez, some 20 years ago, and it still works today. I've only recently started punting it with every selection because in the past I just didn't have the time to update my records manually. I've also developed a couple of other good systems, and logical ones too. Also I've got another method that I'm currently learning, and can test, so there's a vast amount of data to be sorted, and without the database I just wouldn't be in a position to do this.

Also there's myths you can bust.

I also believe you can actually "see" the trends in racing coming out by breaking up your testing into different periods. I especially think someone like you Bernie would have more to gain from a database than someone who hasn't had the experince in racing that you've had, because you would have a feel for what works and what doesn't. You'll also benefit by not having to rely on what's posted on here as to whether a system is successful or not. For $405.00 you get RaceCensus database and for confirmation on it's worthiness I'm sure those on here who have the database will be only to happy to share their experiences.