View Full Version : Premier League Predictor

09-07-2015, 06:48 AM
English Premier League Predictor

Hi Everyone, I am trying to create a predictor that will offer results for matchs that are at least 80 percent correct.

This has turned out to be a very costly exercise and I have set up a crowdfunding site to help me with the costs

Note that for every $20 Aud I will give a free week of Football results.

09-07-2015, 08:25 AM
This is entirely meant as constructive criticism so don't take it otherwise.....

If it works $20 a time isn't that expensive and someone willing to trust your system could easily make that back. However to reach that sort of funding you need to convince people (me at least) a lot more in your pitch being a viable method. Things like demonstrating it's accuracy using previous seasons at least.

As it stands there's nothing there that suggests your methods offer anything over what existing systems do.

09-07-2015, 11:14 AM
Hi Master Yoda,

I am going off what I have researched, my findings have been that all large professional gamblers use algorithms to provide them with outcomes of football and other sports outcomes. I am offering a minimum probability of at least 80%. I have come to this percentage based on what research I have made, I am going to be speaking with a gentleman called Luca Daniels who has done and sells betting algorithms. With the details I am requiring I will be contacting the sports medicine dept of every Premier league club. I am expecting that this will be rather expensive and will be where most of my funds will be spent.

Regarding a price of $20 a week, this is my initial thought as I really want this solution to make people like myself happy to make a couple of hundred pounds/dollars a weekend doing accumulator bets. My feeling is that $20 will be enough to pay my costs (ideally I will have several thousand subscribers).

I hope I have answered your questions. Any feedback is of interest to me.

09-07-2015, 11:31 AM
Sorry All Fired Up I would see no reason from clicking the link or your rationale above to give you $20. If you contacted a sports dept I was involved with I would either tell you to **** off or quote patient cofidentiality at you

09-07-2015, 11:38 AM
80% sounds very ambitious to me. Even if you were endlessly predicting Man United v West Brom, any algorithm is going to predict a home win every time on the balance of probabilities. But in reality West Brom are going to nick at least a win and a draw in every 10 games. So that's barely 80% accuracy on 'home bankers'. I'm not sure how a computer would be able to pick out when those upsets will happen - it's all down to luck on the day, refereeing decisions, perhaps an inspired performance from an otherwise ordinary player.

Then take a game like West Brom v Spurs, a game that could just as easily end in a home win, draw or away win. A computer isn't going to be able to predict whether a mentally-weak team like Spurs are going to be up for it and play them off the park, or concede an early goal from a set-piece and roll over.

I can't see Premier League clubs handing over details on the fatigue levels of their players, whatever the fee on offer.

Football is unpredictable, that's why it's the greatest sport in the world.

13-07-2015, 09:46 AM
Hi Cantspell and WorthingEagle,

I have spoken with my Italian Contact and he said that I am taking on quite a challenge, getting an algorithm that will offer 80%. I have taken note of this. He also passed on to me a stats company, I will look into this tonight and see if they offer the information I need. If they don't then I will call the football clubs. Starting with our club. Either way I am determined to get this solution created and I will do my best to have it up and running this season.