A Family Feud

A Family Feud

A Family Feud

In my previous article, I mentioned that the use of the predictive model ‘PremBot’ was going to undertake some competition, in order to see whether a data driven betting approach is one that we should be using.

While although that article focuses on the pure full-time outcome, I’ve once again asked my brother to offer up his gut instinct on the Over/Under 2.5 goals market and see whether or not he can outfox a predictive model with a 67.2% success rate for the season so far.


Again as previously referenced my brother, is not blind but he is a football fan – although it must be said not one that is steeped in a bedrock of analytics, therefore he is the perfect specimen to go up the might of ‘GoalBot’

With that in mind, here is what he thinks will happen across the course of the next few days:

Which means, if both Arsenal vs Everton picks are to come good, he would need at least a 2-2 draw at the Emirates, not necessarily outside the realms of possibility but perhaps proof that the outcome picks have not led the goal picks.

Because if you are going for a draw as the result, you may have been more inclined to go for Under 2.5 goals also, because a 1-1 or 0-0 draw would be more illogical. Still we must not forget that these are blind picks, so it’s not up to me to over egg the pudding.
GOALS (Over/Under 2.5 in total)
Chelsea Tottenham – Under
Crystal Palace Newcastle United – Under
Burnley Bournemouth – Under
Sheffield United Brighton – Over
Southampton Aston Villa –  Under
Leicester Manchester City – Over
Manchester United Watford – Over
Wolverhampton Wanderers Norwich – Over
Arsenal Everton – Over
Liverpool West Ham – Over
While for additional context, here are the result picks he opted for:
Outcome (Home/D/Away)
Chelsea Tottenham – H

Crystal Palace Newcastle United – A
Burnley Bournemouth – D
Sheffield United Brighton – H
Southampton Aston Villa – H
Leicester Manchester City – A
Manchester United Watford – H
Wolverhampton Wanderers Norwich – H
Arsenal Everton – D
Liverpool West Ham – H

More importantly, we need to see how these stack up against my goal picks – a fearsome list that looks as follows:

A Family Feud
Now regular readers will be aware of the colour code system, but if you don’t I will explain once again:
Green – Back Over 2.5
Amber – Ignore
Red – Back Under 2.5
So, I’m going to need this data in two different variations:
All ten matches, I’ll use to compare against my brother
The eight Green or Red matches, I’ll use for my ‘GoalBot’ series.
Goalbot has currently been tested against 59 matches and 40 of them have been correct, meaning just short of a two thirds success rate. Therefore with eight matches being tested in my latest run of data driven football predictions, I’m going to need at least six to boost my overall percentage.
Now my predictive model ‘GoalBot’ is certainly showing signs of working, there is no doubt about that. However, the true measure of it’s success will be at the end of the season, as next season I will then have a proper benchmark to work against.
Until then, the two action points for this week are:
a) Beat my brother’s gut feel picks (comparing all 10 against each other)
b) get six out of eight in terms of recommended bets
Time will tell, if that proves to be the case.

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 dan@realfootballman.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.

Follow me on Facebook at Dan The Stat Man
You can also check out my Premier League Podcast on Soundcloud

Leave a Reply

Your email address will not be published. Required fields are marked *