I started off with a spreadsheet with 2 years of R6 top rated horse data (2018 & 2019.)

I added in a column to show how many places would be paid.

Obviously, I didn’t want to add in too many filters and, all of the filters that I did use, had to make sense. If you’re starting with a good data set then you don’t need too many filters to make a profitable system.

The first thing that I did was to set a starting point. Without any filters I created a pivot table and found the following…

- There were a total of 21787 top rated (on R6) horses.
- This equates to 30 selections a day.
- The strike rate (for place wins) was 54.56%

### Number of Runners

The obvious starting point for a place betting system is to look at the number of runners.

I did a pivot table on the number of runners and found that the most accurate races were those with between 5 and 10 runners.

Interestingly, I found that we were less profitable for 7 runner races. This makes sense as only 2 places are paid and, if all the runners were equally good or bad, you would have a 2 out of 7 chance of getting a Place Win.

So, the first rule was… **Races with 5,6,8,9 or 10 Runners.**

With this filter in place I found the following…

- There were 14 selections a day.
- The strike rate (for place wins) was 60.14%

### Maximum Forecast Price

I create a Forecast Price for every runner every day.

Sometimes this is very accurate but sometimes less so.

It is, however, more accurate than the Racing Post Forecast Price about 83% of the time.

Now, it seems logical to me that, the lower the forecast price, the more likely the horse will win or, at least get placed.

After all, more 2/1 horses win (or get placed) than 20/1 horses.

So, I did some analysis and found that, if I restricted my selections to those with a forecast price of 6.0 or less I could improve the strike rate by a decent amount.

Thus, the second rule was… **Maximum Forecast Price of 6.0.**

With this filter in place I found the following…

- There were 12.65 selections a day.
- The strike rate (for place wins) was 62.56%

### Confidence Level

Now, Focus Ratings has a unique number that we call confidence.

Unless you use Focus Ratings it is a bit difficult to explain but, basically, the higher the confidence is, the higher the top rated horse stands out from the other horses in the race and, as a result, is more likely to win or get placed.

This was an easy one but I did do some analysis to make sure.

Basically, for a horse to be a selection it has to have a confidence level of 100% or greater (the maximum confidence level is 200%.)

So, the third rule was… **Confidence of 100% or higher.**

With this filter in place I found the following…

- There were 11.40 selections a day.
- The strike rate (for place wins) was 63.76%

### Distance

Finally, the fourth rule was… **Distance of 5 to 23 furlongs (inclusive.)**

The rationale for cutting out the longer races is that, the longer races tend to be Chases and, the longer the race, the more fences there are and…

Every time a horse approaches a fence there’s a chance that it will pull up, fall, run off or unseat the jockey.

And, thus, not win or get placed.

With this filter in place I found the following…

- There were 10.66 selections a day.
- The strike rate (for place wins) was 64.31%

### Race Type

Finally I checked those race types that did better for place betting results.

I sorted the different race types into Strike Rate order (descending) and then selected those race types that, combined, give an overall strike rate of 80%.

Now, there are 93 different rate types as I differentiate between different race classes, as shown below…

And that’s it; 4 very simple and logical filters and selected race types.

By choosing the best race types (in terms of strike rate) I found the following…

- There were 2.62 selections a day.
- The strike rate (for place wins) was 80.00%

Thank you for taking the time to read this…

If you want to see the spreadsheet that I used to record the selections (and the results), please feel free to download… level-stakes-with-monthly-take-outs.xlsx

You will see that, had you started with a £2,000 bank (and, thus, bet to £100 stakes) and taken out profits every month (where the month end bank was at least £2,000) you would end up, after 2 years, with your bank intact (although a few pounds down) and you would have taken out, over those 2 years,

almost £14,000 in profitsalong the way.Not a bad return for a £2,000 investment!

If you want to read about how I was inspired to come up with this system, you can read about it at… When Life Gives You Lemons….

If you are interested in my philosophy on making small profits but, lots of them, I try to explain that at… The Hare and the Tortoise.

And, finally, if you want to see how I stake the selections, I explain that at… The Annuity System.

My kindest regards