Predictortron Jr
You have all been thrilled by the CCN predictortrons (I know this because of the stunned silence whenever I bring the subject up).
Now I’m aiming for the stars and adapting the tools for individual games!
The added advantage of this is that I’ll be able to simulate games between any teams in English football history by the time I’m done. Ooooh.
How?
I’ll keep it under my hat for now, but using simple techniques I can evaluate how strong a team is relative to the rest of the league, then (and this is the clever bit), look up what tends to happen when two teams of these relative strengths play one another.
So for our game away at City, Predictortron Jr said that we had an 81% chance of losing, a 16% chance of drawing and a 2% chance of winning!
The system suggested a 3-0 defeat.
Let’s take a moment to consider that.
Done?
For Stoke tomorrow it says 1-1, 46% home, 30% draw, 24% away.
So there we are. Lump on. Or not – it is early days.
It’s all manual for now but if anyone out there can programme simple macros in Excel I’d be interested in hearing from them. If you know what you’re doing I think it’d be very simple to automate.
That’s all for now.
12 Comments on “Predictortron Jr”
Leave a Reply
Fulham’s Top 50 at The Times
If you enjoyed this and are in a position to dole out writing work, please get in touch!
Winner! 2009 Best Blog at the New Football Pools Fanzine Awards
When Football Was Football: Fulham
Rich's first book. Buy it from all the usual places, including now the club shop!
Buy the Fulham Review!
Or check Amazon for Kindle editions at amazing prices!
oh god 1-1, you have doomed me out..
Just going back. We get 1-1 for the WBA game….
Stoke 1-0 maybe. Need to decide whether to use average or mode score….
Say I get a long list of results that meet my criteria. What do I do, take the one result that occurs most often? I think that makes sense.
This gives us:
0-0 15
0-1 10
0-2 1
1-0 17
1-1 14
1-2 3
1-3 3
1-4 1
2-0 11
2-1 18
2-2 4
3-0 6
3-1 12
3-2 3
4-2 1
5-2 1
(I’ll add more seasons of data to bolster the above over the weekend).
Which suggests a 2-1 win now! Or 1-0. hmmmm.
Come on!!! feel better already
And the following for City:
0-0 4
1-0 4
1-1 3
2-0 5
2-1 3
3-0 3
3-1 1
4-0 2
4-1 1
4-2 1
4-3 1
5-0 2
7-0 1
9-1 1
Which now suggests that 2-0 was the most common result but strong likelihood of a walloping. I need to add more seasons to my repository.
So you gadget thought it was as likely that we would lose 3-1 as 9-1?
/Trollin’
; )
This definitely has potential. Have you sussed out how the Fink Tank system works or have you come up with something new?
Expanded the database:
Stoke predictortron II:
0-0 25
0-1 16
0-2 3
0-3 1
1-0 26
1-1 27
1-2 5
1-3 4
1-4 1
2-0 20
2-1 27
2-2 8
3-0 9
3-1 13
3-2 7
4-0 1
4-1 1
4-2 2
5-2 1
Toughie… should be a tight game.
shall i put a fiver on it? hmmm
No! we’re still working things out…. The fact that it can’t decide between four results says it all. 0-0, 1-0, 1-1, 2-1. Hmmm.
If it helps I have same problem…
Oh cheers Rich. There was me gunna trust your predictor and put a fiver on 2-1 (see above) and you tell me not too!
As it happens, it got it bang on – “2-1 27″ (well was the choice between 1-1 and 2-1)
Get it down to one result and you may be onto something :P