Improving Beaten favourites -system

A while back I was reading through old Smartsig issues. One of the features there is simple to operate systems called KISS’s. In issue 0701 there was a mention of a system about beaten favourites, but it didn’t haven any data to back it up. Long story short, sounded interesting and wanted to give it a go.

Rules of the system are simple, back any beaten favourite which improved between second to last time out and last time out.

I took data from 2012 and 2013 and results are presented below.

n3786
Winners552
Strikerate14.58%
Average odds22.40
Median odds8.96
P/L95.37
Return on Turnover %2.52%
Std deviation4.10
t1.16
p0.1226

A small profit over two years, but according to p-value, there is still over 12% chance that theses results are due to luck alone. And it can be considered that results are not statistically significant. Chart below show how bank shrinks and grows throughout the selections. Over two year the selections are mostly on making losses.

Exacta system test

A while back I was looking through results of my bayesian forecasting system and noticed that my first and second rated selections finish either first or second pretty consistently. About 30% of a time and pretty much regardless of race type or number of runners (unless really high number of runners).

Naturally betting exactas came into mind and I wanted to test if there was any profit to be found from these. Unfortunately I don’t have tote results in handy database so I had to approach testing this manually. For this reason I restricted this to only UK races in classes 2-5 and number of runners in race to 4-12. This way I got roughly four suitable races per day to check results from. For each race I paper played two bets, 1-2 and 2-1 so two points worth of bets per race.

I now have data for the past two months and over 186 races I have had 55 winning exactas for a strike rate of almost 30%. Unfortunately profits weren’t there, loss of 27.4 points. I am not willing to give up yet as class 5 seems to be making profits as well as shorter distances, meaning turf and all weather flat races.

I am going to continue papertrading atleast over the next month to see if there is anything worth backing here.