I've mentioned Race Researcher (RR) in my recent posts and received a lot of questions asking exactly what it is, also if it's available to access.
In this post I'll explain more about RR and give you an example of what it does, and address these questions.
RR is an online platform created by Nigel Dove, the whizz kid behind Hedger Pro and Cloud Bet Bot.
It's loaded with years of racing data from every UK and Ireland meeting, and also Focus Ratings R5 selections.
All of this can be quickly filtered using any criteria we require.
The simpler the better, because too many filters can be convoluted, and create distorted results when attempting to build realistic profitable horse racing betting systems.
You could find a system through filtering the parameters that yields incredible profits, but when looking deeper the wins could have come from a small portion of high odds flukes, or carry with it terrifying drawdown periods.
When I filter the BSP is an one of my main starting points, as I don't want to build the foundation on rank outsiders.
Assessing the drawdown impact over time, not just the amount, but looking deeper at the losing spells within the bigger picture is important.
When we create systems using RR we're looking for less is more, with the smoothest line of growth over a reasonable number of races.
Let me give an example of a work in progress.
The above image shows a strategy output I've been tuning for a couple of weeks. Below this image within RR we have all the filters selected to work with. For now in this example we're focusing on the output section.
Straight off the bat you can see this is not the smoothest graph, which is why I wanted to use it in this RR explanation.
This is just for demonstration only and not a system we've put in play yet. We've a library of strategies Keith, Nigel, and myself are working on, but for the purpose of explaining RR I'm using the example named 'BacksimpleV2IRE'.
The strategy above is using a small set of filters covering Irish racing only, and the blue line gives us the P&L from January 2018 to the present.
You can see from the start of 2018 we've had some months breaking even, some dips in the balance, but also some periods of exceptional growth if you look at Oct 18 for example.
Then up to the end of 2018 no growth, a slight jump in profits at the start of 2019, then a dip followed by a good run of wins, and so on up to the present which is on a decent line of growth.
The stats to the right of the graph show that if we were running this system over the full two years, the Chi Square statistic, which is the Conf percentage, and in this example quite high at 99.97%. The Chi Square statistic is commonly used for testing relationships between categorical variables, and for the purpose of horse racing data analysis a very useful benchmark.
Then we've Runs, which in this example is a total of 2152 using the filters I've set up within RR. This is the amount of horses that have ran according to my filters.
Down from here the output is divided in to Back and Lay results, so in this example we're focusing on the backing side.
Given the number of runs over roughly 22 months, that's approximately 98 bets a month, which will be variable of course but it gives us a good idea.
The bets are calculated as straight flat stake one point backing bets per horse, so if we've multiple qualifying horses in each applicable race, they would be backed as such.
From this we can see an average number of 6 winners a month, as we've 129 in total spread over 22 months. As stated this will of course vary, which you can see in the graph. It would never just be six a month steadily.
So let's think about this, we could have effectively 20 races within the month covering the volume of horses we have, which produces the 6 winners in this example.
However, from these winners over time you can see the Points total is 900 over this period. Which is pretty good you'll have to agree for 22 months betting using 1 point per bet.
The strike is just 6% which you can see from the average winner assessment, but with a huge 42% total ROI.
A commission of 47 points from this profit, assuming everyone has now switched their Betfair rate to 2% as instructed in a previous blog you can find here: https://www.exponentialbet.co.uk/post/betting-system-creation
With a total Draw Down of 97 points at worst over the 22 months. If you'd just started running this system when the dip occurred to hit this losing run, and started with a 100 point bank, you would effectively wipe out your balance leaving just 3 points in the account.
In order to continue betting you'd have needed to re fuel the account, and then be able to carry on. You can see from the graph that on numerous occasions the balance has taken a hit, and then hit a run of wins to recover.
This strategy as it is shows an incredible ROI and points increase in just under two years, but to run it you need to be confident the wins will come to recover previous losses, and push the balance to a new high. Which you can see it has done over the period, but the Oct 18 jump I feel shows more luck than accurate data assessment. Hence why this strategy is still a work in progress.
Without this lucky run over the Oct 18 period I think we'd be closer to 500 points total growth, but not with much difference to the draw down number. Still a very impressive number over 22 months!
Other than the freak period of wins the strategy shows a lot of potential at this stage, and I'll keep tuning it to smooth out the graph. Simplifying the betting process, and creating a steadier line of growth.
I accept that some systems will have draw down levels that require either smaller percentage stakes to reach reduced percentages of the predicted points balance.
Keeping this to a minimum is more appealing to me obviously, as we'd always like to go for the low hanging fruit. We can combine strategies using RR after all cross market.
For the systems that do show more volatility over the years we'll highlight this with all systems we decide to introduce to the Hedger Pro library, or to run using Cloud Bet Bot.
The decision will then be as stated reduce staking to anticipate it, which means the points growth would need to be very high.
Alternatively, simply accept that it's not to be run as a betting system in a lot of ways, but more as an investment that over time could require some maintenance and a definite long term view.
The confidence to keep going if opting for this is based on data demonstrating recovery with continued growth, which helps us make these decisions.
It's a gamble of course, which is the case for all types of investing at some level, but you would have the luxury of making an informed choice based on cold hard facts.
If you were reinvesting at the roulette table using something stupid like a martingale on red or black. That's pure chance where the laws of probability do not apply, as much as we'd think red must land next after 16 black spins, there could in fact be another 16 or more black spins.
In using the data from the past two to four years with RR, which is a relevant time period, and uncomplicated selection filters that provide a clear idea of what to expect, this gives us a decent edge. It will help us decide what to stake, how to stake, and how we approach the entire betting process with some of these strategies.
I haven't touched on laying, laying flat or to fixed liability, Dutching across selections, or double dutching, which is a very smart way to reduce exposure, and not just something people say when they haven't a clue what someone else is saying to them.
Now bear in mind this is just one example today from many we're presently working on, as I wanted to explain more about what RR does, and give a working example of how we can utilize this for the good of our members.
The other question I've been asked is whether RR will be available for members to use, and the simple answer is yes at some point in the future.
We're looking to include a range of our exclusive back end tools for early adopters for our crowd funded software, you can read that post here: https://www.exponentialbet.co.uk/post/create-your-own-bet-trade-software
Moving on from this we will look to create a version which is available to purchase, as we understand a lot of members want this, but we'd also like it to be available at some point for you to create strategies to be included within the Hedger Pro library for other users.
For now though it's a tool we're using to set up a foundation of strategies to run using the cloud service, and to launch the Hedger Pro library.
It's still very much in development with Nigel making improvements to the user interface, and how it outputs.
The goods news is, after recent stress tests to compare the data output to other industry standard sources, RR got the thumbs up as being the most accurate horse racing data assessment tool.
What you're not seeing from our screenshots and videos are the workings of the platform, which for obvious reasons we're keeping under wraps. You'd be amazed how many copying bastards who would try to create poor imitations. Why would we give up all the hard work to competitor sites, when we've so many amazing plans to use this for your benefit. Been there and got the T shirt!
I'll post again later in the week with another couple of examples from RR, and my progress with this Irish backing system shown in today's example, should I decide it's got legs compared to other more lucrative systems I've been working on. I will try and smooth out this line of growth though and explain more about this, as it would be good to follow on from this post.
On a side note, we are looking to release a selection of free to use systems from RR for members on our mailing list. If you've not registered yet, you can sign up from most pages, including the home page near the footer.
Remote Dutch Betting (RDB) members, you'll have details of your new set up within Cloud Bet Bot, and a clear overview of your staking using the latest RR parameters before we go live on the 22nd October,so before the end of next week.
Anyone on the invite waiting list for RDB I'll notify once spaces become available, but this will be after it's been running for existing members for at least a couple of weeks, and the landing page information updated to reflect the latest performance expectations based against RR data.
We may not be adding new members to the service for some time, if we see existing members bets using new parameters impact market liquidity up to max capacity per race. However, with the revised staking plan Nigel and myself have been looking at to lower exposure, we will ideally have some opportunity for new members to join soon.
Tomorrow and Tuesday I'm working with Nigel and Keith to introduce development updates for RDB, so my next blog post will be over the next few days covering the progress we've made with this, and more details on stats as shown in today's example.
Thanks for reading. Any questions drop me an email to ryan@exponentialbet.co.uk and I'll get back to you asap.
Comments