Ok so quick update as promised. I have been working flat out day and night on the setup ever since I got back from London and am delighted to say that after weeks of seemingly going round in circles I have made a huge amount of progress over the last few days. 

I had been working on my old systems and trying to implement them as a selection method into the new set-up. However with the ability to run back tests and query the historic data like I never could before trying to implement the old systems quickly turned into building new ones. 

I have built 3 systems (6 if you include their all weather variants) all heavily focused around my speed ratings. These have all been tested multiple times and I don't have the figures to hand but the results are looking very promising. I wouldn't say they are finished (is anything on a project like this ever fully finished?) but they are at the stage where I cannot fine tune them any further without backfitting the data to suit. Time to live test them. 

They are all based on the principles of one of my old systems. For example the new Top Speed system is based on my old Top Rated system although the profiler transforms the selection process. Before on average I had well over 1,000 bets a year with this system. Back testing the new system it averages now around 150 bets a year with a much improved strike rate and ROI. I used to do okay at it but this does much better than anything I ever achieved doing it by hand.

My old Horses for Courses and Star Horse systems form the foundation for the other two systems. Yes the names for some of my old systems are ridiculous but it made perfect sense years ago when I dreamed them up. I have several others in varying degrees of development their current status based on how quickly I hit a roadblock with analysing the data.

Some systems like the Top Speed system were by comparison easy to setup as the data done the work itself. What I mean is there was a clear cut defined line in the data that could be filtered out each time, reanalyzed, rinse and repeat. Based on the core principles the system largely developed itself. Same can be said for the other two forementioned systems. 

The majority of the other systems are not so straight forward to implement as there are no defined lines. It is just a blur of percentages which will require much more time spent analysing and testing to break open. There are also other variants of systems I want to explore further based on trends I've spotted and made a note on. The possibilities with this setup are endless - I would need a team of crack stat guys working on this full time to develop and test every idea I have.

However development will need to slow down for the next while at least as I had to go back to the office today. I have been working from home for several weeks spending 110% of my time on this so the office is in literal chaos. Need to focus some of my energy on sorting that shitshow before I get sacked 🤣. I also need to slow down again as I have been at this non stop 18-20 hour days for weeks. Life needs me to step in and sort some things out so need to spend some time away from this and limit any development time to a hour or two a day for the next few weeks. My brain feels like literal mush now, need to slow down a bit to recover but try to retain some momentum.

What time I do have for this will be spent developing the above systems and especially the system I mentioned yesterday. It is based on a highly modified version of my danger horse and value seeker systems alongside trends in the data. Trends I spotted repeatedly so I pulled together a set of data and the back test results were phenomenal. It is still early days and I need to get much deeper into the data but the raw ROI of this one unpolished like it is has me very excited.

Moving forward now we have working systems to live test I will post the system selections for the 3 (6) speed rating systems. The place strike rate on these is very high so we will include a place accumulator bet. I will also post any selections from the yet to be named system. Be warned as it stands currently the strike rate overall is around 1 in 10 and the losing streaks can be long but the ROI exceeds all other systems by a significant amount. Gives you an idea of the price range we are targeting with some selections here so some patience will be required. I am excited to get into this data on this one even deeper than I have done already. 

Ok so the selections for tomorrow are below. With regards to stakes I will bet these at half normal campaign stakes until I officially restart my campaign next week. Despite all I said above if I can get a full day at the weekend to review the speed rating systems and complete any final fine tuning I believe these should be good to go at full campaign stakes.

HFC Selections

Kempton 19.00 - Dubai Leader @ 4.0

Kempton 20.45 - Lilikian @ 7.00

Newbury 19.50 - Flying Secret @ 4.00

Dubai Leader is also a top speed selection so should go very well. An accumulator pays better than 40/1 and the above 3 in a place accumulator pays better than 7/1 so definitely worth a shot. 

The selections for the as yet unnamed system are below;

Yarmouth 15.30 - Aryaah @ 34.0 

Yarmouth 16.00 - Monopolise @ 6.0

Kempton 19.35 - Fantasy Believer @ 4.33

I would suggest backing these as each way singles as back test results suggest they do place often but based on current strike rate I would advise small stakes and avoid any multiple bets at this time. Also is Danlue a stupid name for this system? The systems all need to have a name. It's a mix of danger and value just FYI incase anyone can't work it out . Maybe Valger is better? Maybe I should stop talking and just go to bed 🤣.

Thanks,

Darth Niloc