Moving Ever Quicker
Moving Ever Quicker
If you read my previous article, you will know that with there being so much festive football, the pursuit for data driven betting excellence was one that had to be constructed in double quick time and now in this article, I am going to both set the scene and recap my predictions for last weekend.
THE SCENE SETTER
As per usual, it is a case of gut instinct versus data driven football predictions and here are the picks that came from simply yours truly and more importantly, how they fared across the course of last weekend:
Liverpool – Y
Manchester City – Y
Everton – Y
Newcastle – N
Brighton – N
Tottenham – N
Manchester United – Y
West Brom/Aston Villa Draw – N
Burnley/Wolves Draw – N
Chelsea – Y
A middle of the road 5 out of 10, which means I am now on 65 out of 137 on 47.4% for the season so far.
Now lets look at how “PremBot” fared last weekend:
A lesser off 4 out of 10, while there was a slight change in the PremBot modelling for PremBot 2 and this allowed a decent start to the slight tweak:
Using the form guide for the last six home matches, versus the last six away matches as more of a comparison it meant a slight improvement and this week “PremBot Jr” managed to get 6 out of 10 correct.
Let’s now see how the data driven football predictions are faring against gut instinct:
Pure Gut Instinct: 65/137 (47.44%)
PremBot 1: 29/69 (42.02%)
PremBot 2: 27/59 (45.76%)
Hybrid Model (Combining First 68 matches and then PremBot only): 67/137 (48.90%)
As you can see, the percentages are not quite where we want them to be and therefore, more work needs to be done. I wil let the Christmas period wash out and then, make some tweaks while also scaling up productions with some Euro models also.
Anyway Christmas is here, so all hands to the pump for Boxing Day.
Happy punting and thanks for reading – to discuss this or any other projects you have, feel free to message me at either @dantracey1983 on Twitter or email me: email@example.com
(THESE ARE NOT TIPS PER SE, THESE ARE JUST DATA DRIVEN FOOTBALL PREDICTIONS IN A PURE TEST AND LEARN MODELLING CAPACITY)