March 6, 2015 at 23:37 #806163
Some keen TRF’ers have been beavering away writing previews for some of next week’s races. Well worth a read and some food for thought. Go here and all races with an asterisk on the menu have a preview posted – Cheltenham preview articles
I’ve done a trainer trends piece – take care with it though, although the trends are real the data set (last 12 years festival results) is very small and what happened in the past is not necessarily a guide to what will happen in the future Cormack’s Cheltenham Festival top 5 Trainer TrendsMarch 7, 2015 at 11:22 #806658
- Total Posts 5651
Your trainer trends one is striking, Corm, especially the Jonjo stat, but how would you argue the Mullins 0 from 48 one with a statistician? It would be interesting to get Prufrock’s take on it.
0 from 48 sounds pretty desperate, but it was spread over 18 races, making it around 2.5 runners per race, therefore 1.5 of those were already doomed to failure. Or is that more than compensated for by having multiple runners in a race?
I wish my brain were bigger!
Never argue with a fool. He will drag you down to his level and beat you with experience, then onlookers might not be able to tell the difference. https://lazybet.com/March 7, 2015 at 11:33 #806663
Yes, NONE of them would stack up from a statistical viewpoint really, small sample sizes, etc. Pru would have a field day!
They are meant more as something to think about and take into account, coffee table stats might be how to look at them.
From an expectation of number winners viewpoint (based on the market’s assessment of the % chance of each selection of the sample, I’ll post something later on that, doing the ironing just now!) the Mullins trend is pretty weak to be honest. A lot of those 48 started pretty big prices so although it’s a good ‘headline stat’ treat it with caution, although I do think there is some logic to it, as I mention briefly in the piece.
When I put these stats together and look for patterns in previous results I try to approach it from a starting point of establishing if I think there are logical reasons why the stat may have come to exist. But, of course, as with all back-fitting it can be easy to rationalise in hindsight.March 7, 2015 at 12:20 #806678
The 0 from 46 wasn’t over 18 races, the 18 races was the handicaps only SC.March 8, 2015 at 14:49 #808087
We’ve added a few more races (marked with asterisks) and I’d like to thank Thomas Rowe, Tom Boardman, Chris Flavell and Ben Stones for sterling work producing these well-written and thought-provoking articles for us. List of articles
I’ve also written another list of more general tactics you might consider to help narrow down the volume of runners you need to consider for each race Five Easy Ways to cut down on your Cheltenham HomeworkMarch 15, 2015 at 00:45 #833695
I had a quick look just now to see how the ‘trainer trends’ we put up did – glad to report you’d have shown a 9.5 pts profit if you’d followed each of the trends and backed/layed them all religiously.March 15, 2015 at 01:09 #833715
Even better on the other stats article – 22.8 pts profit if you’d followed the highlighted trends.
I’ve updated both articles to show how each stat got on individually.
You must be logged in to reply to this topic.