Waving The Red Flag
Waving The Red Flag
If you read my previous article on trying to predict the English Premier League through modelling, you would know the first two weeks have been steady if not spectacular and as a result, I have seen six out of ten correct picks being returned each week.
Considering we are playing only on a percentage success rate basis only, the current average of 60% (if only because both weeks have seen that particular return) can be considered in the top half of the table and knocking on the door of a Europa League place.
While because of the relative success of the first two weeks, the logical notion would be ‘if it ain’t broke, then don’t fix it’. Although, with that said, I do have a feeling that not being willing to change, is something that could prove to my undoing this week.
Why is that you ask, well first let’s have a look at what ‘PremBot’ (that’s it’s new working name) has offered up this weekend:
Now as always, there a couple of potential results that I don’t really like the look of and although Tottenham’s and Manchester United’s woes this season have been well documented, you would think that they would be able to get the better of Sheffield United and Brighton respectively.
Now of course, you can argue that Sheffield United are higher in the table than Tottenham and have better form than this weekend’s opponents, as a good enough reason for them to come out on top in North London.
However, the fact that they are 6.0 to win at Tottenham’s new stadium, says one of two things:
1) either the model is wrong
2) there is an incredible bit of value on offer
Now the reason it doesn’t sit well with me, is something that I’ve mentioned a number of teams already and that is previous bias. Quite simply, I’m under the perhaps incorrect belief that ‘Big Six team at home = win”
However, the very notion of the ‘Big Six’ is one that is certainly crumbling in front of our this season and although this could be another Leicester 2016 off season, there is every chance that the landscape of the division is changing.
Whether that is actually is the case, is something we won’t really know until the 2020/21 season and if the same established clubs are still in a state of flux, then we will know that the status quo has morphed into the huge two and just about everyone else.
To be honest, that is nothing more than conjecture and therefore, let’s return back to this weekend and the simple fact that Brighton have not won 33 of their last 37 away Premier League matches – a stat that questions how on earth they have been tipped to win on Sunday.
NOT FIT FOR PURPOSE?
Now the reason that Brighton have been tipped to win this weekend, is because overall their form is currently better than that of Manchester United’s, for that there is absolutely no doubt. However, the good run of form for Graham Potter’s men has come solely at home.
This means because they have been great at The AMEX but pretty rotten on the road, it is skewing the outcome that ‘PremBot’ has offered me this weekend and here I have to decide whether to leave as is, or go against the grain.
Although if I do that, then I may as well go back to gut instincts and this whole predictive modelling project then becomes moot. With that said, because I like to throw caution to the wind and the fact that there’s only one more week before the international break, I will leave as is.
To be honest, the urge to change is lightly less because there is no money on the line and to be honest I need to once again put my big club bias to one side. With that mind, I’m going to sit back and see how successful we are and hope that Liverpool vs Manchester City ends in a draw.
Happy punting and thanks for reading. Dan
If this has grabbed your interest and you would like to discuss/feedback then please feel free to drop me a message at email@example.com. While I am always looking for new football/data projects to work on and if you feel that my skills would be of use, I can be contacted at the same address.