A Gut Check
A Gut Check
As you will know by now, I have been testing out some predictive modelling for both the English and Russian Premier League competitions. For the former, it has been a case of just playing for a percentage hit rate and for the latter it is also about seeing a return on my investment.
Now with just two models going, it means the content schedule becomes a little top heavy and is geared more towards the weekend. Therefore, I thought I would try something different over the course of the past few days and that was some blind tests accumulators.
Just to explain a little further, this wasn’t completely blind in the sense of the most random competitions and not having an absolute clear who was top of the table or not, this was based on leagues that have very few teams (maximum 12)
With the theory being, that I go all in on a 5 or 6 game accumulator – covering every game in the division that weekend and see if I can scoop the lot, with the competitions in question being:
Scottish Championship
Scottish League 1
Scottish League 2
Austrian Bundesliga
Now you may ask why those in particular? Firstly because of the reason above but also in regard to Scottish football, my brother has moved to Edinburgh and therefore I have taken more of an interest in it and in terms of the Austrian Bundesliga, I have been doing infographs for a wonderful podcast.
That podcast is known as The Other Bundesliga (do check them out after you have read this) and this is what an Austrian Bundesliga infograph looks like:
Now on the basis of the above, I picked the following results for the weekend (correct picks in bold)
Sturm Graz vs Wolfsberger AC – Draw
Rapid Vienna vs SKN St. Poelten – Home Win
Mattersburg vs Red Bull Salzburg – Away Win
Admira Wacker vs Hartberg – Draw
WSG Swarovski Tirol vs Altach – Home Win
Lask vs Austria Vienna – Home Win
As you can see a relegation threatened two out of six and one of those, even your Grandma could have predicted correctly (that of course being Red Bull Salzburg to win) therefore it seems as if my gut was not strong when it comes to the Austrian Bundesliga.
But why is that the case, especially if a neat infograph could point you in the right directions in terms of results. Well this is where, a small league actually becomes your undoing (or so my theory/excuse goes)
Because with less teams in the league, the gap between top and bottom is obviously a lot smaller and this means that there is more volatility in the division, then compared to a 20 team league such as the English Premier League.
This is also means that although there are less ‘shocks’ because the middle group of teams are a lot more concentrated, it does make the attempt at picking result outcomes on a whim a lot more difficult.
However, this does also raise another question.
Would a model have done any different?
Looking at the data above and the form guides, I would imagine it would look something like:
Sturm Graz vs Wolfsberger AC – Draw
Rapid Vienna vs SKN St. Poelten – Home Win
Mattersburg vs Red Bull Salzburg – Away Win
Admira Wacker vs Hartberg – Home Win
WSG Swarovski Tirol vs Altach – Home Win
Lask vs Austria Vienna – Home Win
So there’s only one game that would have been different, had I built a model (using the same logic as either of the other two that are currently in operation) therefore perhaps I am being too hard on my own picks and it’s more of a bad week at the office, rather than any faulty logic.
Still two out of six, is not going to make you rich anytime soon and that’s before I give a very quick recap on how I fared in Scotland:
Scottish Championship: 3 out of 5 correct
Scottish League 1: 1 out of 5 correct
Scottish League 2: 3 out of 5 correct
Overall 7 out of 15 random Scottish picks were correct, with the Championship and League 2 being respectable in terms of return and League 1 being forgettable. Ultimately though, none of the trio saw me come close to winning a neat five-fold come Sunday evening.
Therefore, what have we learned from this?
We can categorically say that betting on ‘gut instinct’ alone is more of a fool’s pursuit and this is where you should put more credence into data driven betting. That is something that I will be doing going forward and from here on in, it is predictive modelling all the way (or bankruptcy, I’m not sure which)
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