Trying It Again
Trying It Again
If you read my previous article you would be aware that my attempt at beating the bookmakers via a data driven approach had undergone a fundamental change and it is one, that must be said was not all that successful.
That’s because after building a predictive model in a bid to get the better of the English Premier League, it is clearly apparent that the English Premier League got the better of me and that’s because only 30% of the predictive picks were correct.
In summary that’s a real bad day at the office and when you consider that the accumulative odds for all 10 picks to be correct, was a staggering 22000/1 – you get the feeling that I was always going to be up against it.
Was I Being Too Clever?
That’s the question I need to be asking myself, because in previous weeks I have looked at just the comparison of league form when trying to predict results and last week, I added in comparative league positions also.
Not only that but it did not factor in a new manager bounce and with the news that Unai Emery has been sacked, I’m probably going to need this once again. Although somewhat annoyingly I have made the predictions just minutes after he was given his Arsenal marching orders.
Anyway I digress slightly and that’s not before I mention the relative upsets involving Arsenal and Everton home, as they failed to get the better of Norwich and Everton respectively. However there’s no point bemoaning my luck, let’s look ahead instead and that means another round of picks.
Picks that look like below:
This week, I’ve decided to make absolutely no changes to the model from the week previous and to be honest I would have to say I am confident that this week should be a better in terms of overall performance.
If only because a lot of the fixtures look like relative home bankers – looking at that list, you would have Liverpool, Chelsea and Leicester all nailed to win and that’s without considering Tottenham to continue their resurgence under Jose Mourinho.
Not only that but Manchester City will be well poised to exact revenge on Newcastle, after suffering a shock defeat to the Magpies almost 12 months ago and to be honest those picks above would be very much inline if I went with gut-feel alone. (Bar Arsenal after their managerial news)
Which means that because the picks seem quite logical, there is not a great deal of value to be found this week and a 10-fold acca would return 689/1 instead of the ludicrous 22000/1 that was being quoted last week.
Therefore with more picks mimicking the bookmakers, this should provide a better percentage return at the end of it and looking at this, I’ll be aiming for 60% (6 out of 10 correct) to be the strike rate come the end of the weekend.
If that does prove to be the case, then it as a method that I will continue with (although there is still plenty of scope for extra layers to be added) and if doesn’t prove to be the case, then this thing is going in the bin!
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 firstname.lastname@example.org. 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.
Follow me on Facebook at Dan The Stat Man
You can also check out my Premier League Podcast on Soundcloud