Sometime towards the end of last year after I clicked the update button on the database, instead of walking off to make a cup of tea and roll a smoke as I do every single time, I sat at the PC and watched everything process for the next 10 minutes. Once the database had updated and run all the calculations, it began to process the various selection systems when I noticed something which to that point in time had escaped me...

PROCESSING 1 OF 1,311 RECORDS

What that means is I had over 1,300 systems active in the database. No wonder it was taking 30 minutes per day to update! My approach to system build and design over the years was what I can only describe as gung fucking ho. I was constantly spotting trends or patterns that I wanted to explore further and have the ability within my database to back test any concept or half baked idea via setting up a new system.

Unfortunately even with the database I do not have the time to develop and test every single idea. Even with the trained datasets the actual system building and development is a very time consuming endeavor solely based on the huge number of variables that make up finding the winner in a race. I needed to decide how I was going to move forward and where to focus the limited time I have each day in order to maximize profits.

I began to review the systems one by one averaging around 1.5 systems per day over the first couple of weeks. I done the math and realised it was going to take me two and a half years to work my way through all of them. I decided I needed a more ruthless approach so I set up a rating scale based on a number of factors including profitability. Then came the purge...

This wasn't an easy process as many systems were a single tick away from being safe and some had been a mainstay in my arsenal for years but in the end I reduced the number of systems down to just over 100 which includes 45 complete, active and profitable systems and another 60 or so systems that ticked all the boxes but still require further development.

It is the selections from these 45 systems that I will share here for this campaign run. At least half of these systems are built for Fairbot but so it is unlikely I will share these. However I think I maybe said in my last post that the top 5 systems at least can work without Automation for the most part. I still got 10 days to figure things out.

Anyway as promised here is the first preview of my top five systems ahead of the new season. A few quick notes before we dive in...

All system tests are conducted at level stakes @ 1pt.

All systems are back tested over 20 years of historical data.

All prices are taken at Betfair Starting Price.

Now on to the preview of my 5th most profitable backing system... System 23 - The Hold Up Specialists

This is a pace based system built around hold-up style horses (those runners who drop to the rear of the field at the start and make their move late into the race). It looks only at handicap races class 4 and above with runners who are aged between 3-8 years of age and have a favorable draw bias. There are flat and all weather variants of this system, the results below are for both systems combined.

I believe the results speak for themselves. An average yearly return of 132pts at level stakes with an average win BSP of 8.32 and a strike rate of 13.13%. Profit consistency is very important in all my system builds and it shows here - This system has proven profitable 8 out of the last 10 years with minimal losses on the losing years.

It produces on average around 5 selections per day but comes with long losing streaks the longest ever being 49 bets so some patience is required. If anyone is interested in seeing how the above translates to actual runners I've included below the system selections for Southwell AW today.

16:52 - Thurso

17:25 - Star Legend / Swinging London

18:00 - Bearaway / Without Compromise

18:30 - Rhythm Master

20:30 - Dashing Harry

20:30 - Ultramarine

Feedback or comments always welcomed. I will post the review for the 4th ranked system whenever I get the chance to pull it together sometime later this week.

Thanks,

Darth Niloc