Pre-List: A re-cap.
The Dutching Methodology on the blog is 90% there off race selection.
Race selection is based on Ten Year Trend stats for each race type. These are probably the only rock solid stats in racing? Annual fluctuation is very low indeed. Applying the variable Dutch Frame method I use and a betting ratio within defined risk/reward limits is enough to generate an edge directly on betfair markets. Further filters are essentially to control volatility and I don't recommend running without them.
I was given a copy of 'Braddocks' Horse Race Selection & Betting 1987 a good while ago. Peter Braddock lays a lot of emphasis on the Type of race stats in his selection process and gives details by particular for inst Novice Hurdles, Novice Chases. His samples are only 2 year 1985/1986. I say only but considering he had no computer records and had to hand ball the stats - a hard task!
The Ten year rolling trend stats I use are remarkably unchanged over the intervening 28 years; from before betfair even existed. I doubt you could offer any other racing stat. about which the same could be said?
The only added stat in the table since the days when 'Braddock' was the bible is the 'Uplift' stat I added on the RH side. This is a figure I devised to drive Filters on the proposed Dutch Frame.
There is some skew in the stats as presented. The highest rank #1 Novice chases has improved over the years due to decreasing field sizes for these events. Novice Hurdles, I believe due to the much higher fitness regime for these type of races (think Martin Pipe) return higher fav. figures compared with Braddock's stats. Our figure of 45.9% favs.compares with Braddock's 39.4% favs -off his smaller sample.
So looking at the high strike Novice Hurdle 79.8% 1st/2nd/3rd in market compared to say turf handicaps down at 56.3% with favs only hitting 26% - Do the betfair markets fully reflect this variance? Of course they do to an extent but in my view nowhere near fully. On this basis I expect to continue to grind out a long term edge direct from the betfair markets without the need to chase bookmaker Value odds which are increasingly fictituous and which you are simply closed down on after limited success.
The daily pre-list is basically straight off the Green ranked race types on the above chart.
Each race type has some outlier stats. In class 2 (top) maidens for inst the strike rate falls away quite dramatically. Simillarly with many Grade 1 races in National Hunt.
Todays pre-list has 2 races eliminated from the Curragh. Also todays Group races are out of the frame for the side bets I sometimes follow in the Group contests.
Leaving 5 no: races to check against the Market filters I apply.
The 15;15 race 3 runner Dutch Frame has fallen below the 45 return minimum as a straight Dutch bet.
Here is an optional weighted Dutch bet which I would take scraping up the 45 return
On marginal situations it's ok to use hedging down the third horse to take the bet. Of course anyone looking at these races may have a strong view on hedging down a particular horse. If your numbers say the fav. is a flop? Thats ok too. Hedge him down and take the bet as you see it.
Here is an optional weighted Dutch bet which I would take scraping up the 45 return
On marginal situations it's ok to use hedging down the third horse to take the bet. Of course anyone looking at these races may have a strong view on hedging down a particular horse. If your numbers say the fav. is a flop? Thats ok too. Hedge him down and take the bet as you see it.
No comments:
Post a Comment