The International Break

The International Break

If you read my previous article, you will know that the pursuit for data driven betting excellence had taken a positive turn and with the Premier League stocking up enough in the way of the form guide, it was time to unleash my good friend “PremBot” on the world.

Which means that last weekend saw the first raft of data driven football predictions for this season and with us in the depths of a depressing international break, there is no better time to look at how they and my gut instinct both fared

HEAD OR HEART

First lets recap on what those picks actually were and their overall performace – first up the gut instinct:

Brighton – N
Southampton – Y
Everton/Manchester United Draw – N
Leeds – N
Chelsea – Y
West Ham – Y
West Brom/Tottenham Draw – N
Leicester – Y
Manchester City/Liverpool Draw – Y
Arsenal – N

A bang average 5 out of 10, which means that from this random method alone, I have managed 43 correct predictions out of the first 78 games on offer, giving me a gut instinct success rate of 55.1% (down 0.7% from last week)

It’s not enough to herald a decent level of bookmaker bashing, but it is not a complete disgrace either. With that said, we will leave it there, as the real interest is with how the data driven football predictions and “PremBot” fared.

From those picks in the prediction column, it also managed to get 5 out of 10 correct, which means it is no better than how my gut instinct fared and in this initial round of testing, neither method was better than the other.

A SCORE DRAW

If we look at the differential games between myself and the bookmakers:

Everton vs Manchester United
Crystal Palace vs Leeds
Manchester City vs Liverpool
Arsenal vs Aston Villa

My model managed to get none of these predictions correct, which means there is certainly some work to be done going forward. Still that’s not the worst thing in the world, as we can fine tune this along the way and seem some sembalnce of starting point.

What it does show though, is that it can certainly sniff out a ‘banker’ and it was Brighton not getting the better of Burnley which would have got all six of these correct. So there is at least some reason to be cheerful.

TAKING STOCK

While we are here, we should also take stock of how the bookmakers pre-match predictions fared last weekend:

Today, 06 Nov			1	X	2	B’s
16:30	Brighton – Burnley		1.89	3.49	4.57	13
19:00	Southampton – Newcastle		2.03	3.61	3.76	13
Tomorrow, 07 Nov			1	X	2	B’s
11:30	Everton – Manchester Utd		2.94	3.57	2.43	13
14:00	Crystal Palace – Leeds		2.94	3.41	2.49	13
16:30	Chelsea – Sheffield Utd		1.40	4.96	8.34	13
19:00	West Ham – Fulham		1.84	3.89	4.30	13
08 Nov 2020			1	X	2	B’s
11:00	West Brom – Tottenham		6.42	4.42	1.53	13
13:00	Leicester – Wolves		2.39	3.19	3.31	13
15:30	Manchester City – Liverpool		1.99	4.11	3.50	13
18:15	Arsenal – Aston Villa		1.71	4.04	5.04	13

The bookmakers were slightly sharper last weekend as they not only got 6 winners out of 10 correct but also priced up Manchester United correctly as winners, so I guess a tip of the hat has to be pointed in their direction.

This means that the bookmakers have managed to correctly predict 38 wins from the first 78 mathces of the season, 1 shy of half and 48.7% in terms of hit rate.

SCORES ON THE DOORS

Now we need to start putting some running totals in place for the models that are in play:

Pure Gut Instinct: 43/78 (55.1%)
PremBot: 5/10 (50%)
Hybrid Model (Combining First 68 matches and then PremBot only): 43/78 (55.1%)

Which means we are just about in the halfway house and next week, I am going to divise a second model to also get us going,using the data that has been collated from the first eight weeks of the season and just where the wins/losses are coming from exactly.

Happy punting and thanks for reading.

(THESE ARE NOT TIPS PER SE, THESE ARE JUST DATA DRIVEN FOOTBALL PREDICTIONS IN A PURE TEST AND LEARN MODELLING CAPACITY)