Congratulations luvfooty (DT) and whatlez (SC) – the 2012 Premiers!

Author Archive

Moving on up…


Things are starting to look better for the poor beleaguered odds system!  The favourite won in 34 games this week, and in one game two teams were both equal “favourite” so I give myself half a point for that – which makes it a grand total of 34.5 points!  If I repeat what I did last week, the picture now looks like this:

Still not great – I should have got around 39 right this time.  The chances of getting 34 or less right if my odds were accurate are 12.6% – so, a bad week, but not horrible.  Kinda like my actual footy tipping!  I’ve got some ideas to try and improve the tipping scheme, so we’ll see what kind of difference that makes in the weeks to come.


What are the odds?


So, one round in with our new shiny odds – how did they go?  Well, not as accurate as I would have liked!  Using my odds, the favourite won in only 31 of the 64 games this week, so you could do as well by tossing a coin!  But could it have just been a fluke?  Well, we have a way of finding out.  I used what’s called a randomisation test to work out how many times the favourites should have won if my odds were accurate.  And I got a pretty picture:

Hello!

So looking at this picture, I should have got around 40 right.  The chances of getting 31 or less right if my odds were accurate are only 2.0%, and that’s what we in the trade call “statistically significant”.  This tells us that my odds probably aren’t yet an accurate reflection on the teams’ chances, otherwise my odds would have tipped better!

The reason it’s not working yet is because you’re all unpredictable.  I can’t say when an underperforming team might decide to make lots of trades, go for broke, and suddenly start scoring better – and telling a coach that the odds are stacked against them is the surest way to make him/her work their asses off to win 😉  As Ossie has said – these odds will get more accurate as the season goes on, scores start to settle, and the patterns become more clear.  In the meantime, if you have any suggestions for improvement or anything like that, let me know!


The League – now with Odds!


Hi everyone!

I’m nick_the_insane, a friend of ossie85 in the real world and part-time editor of this fine publication. If you look through and find one of your posts suddenly has more punctuation than you remember putting in it, that was probably me!

Our glorious leader ossie85 has given me the job of announcing the newest feature to this here blog. So here it is: finally, odds will be available, starting this week. He’ll be putting them up very soon, so keep your eyes peeled!

I wrote the code for this up a while ago for our own personal SuperCoach league last year, just for fun. It’s a pretty basic system – it works based on the scores you’ve made so far this season in either Dream Team and SuperCoach, and assumes that your scores are going to go pretty much on the same track as compared to everyone else’s. Obviously this isn’t hugely accurate, but I’m not getting paid for this so you’ll have to make do for now 😛 It runs a simulation of the entire season, including the finals, thousands of times (the technical term is Monte Carlo simulation), and from that it works out the chances of particular events happening. And from that, we can work out the odds a betting agency would give you for them (taking a little cream off the top, of course 😉 )

The odds that will be available will be for all matches played from here on in, and also the odds of any given team making the finals, winning the premiership, winning the wooden spoon – that kind of stuff. Hopefully it’ll give you a reasonable idea of how your team is faring. If you have any ideas or comments, or would like to know more information about how this works, let me know 🙂

– Nick

Admin’s Note

If you’re a fan of Nick’s work, head over to his blog. He is a budding scientist, musician and artist 🙂

 – ossie85