Over the break I came across what must surely be amongst the simplest, most practical team rating systems.
It's based on the general premise that a team's rating should be proportional to the sum of the ratings of the teams that it has defeated. In the variant that I've used, each team's rating is proportional to the rating of those teams it has defeated on each occasion that it has faced them in a given season plus one-half of the rating of those teams with which it has drawn if they played only once, or with which it has won once and lost once if they have played twice during the season.
Saturday, January 31, 2009
Sunday, January 25, 2009
Surprisals in 2009
This year we'll once again be using surprisals as a way of quantifying how unpredictable the results of each round and each team have been.
In addition to measuring the surprisal of head-to-head results, which is what we did last year, we'll also look at how surprising each result has been from a line betting point of view - a measure of how accurately the bookies have been able to predict not just the winner but the margin too.
If you're interested in the details, please download this document.
Tuesday, January 20, 2009
Welcome to MAFL Stats
Okay, here's a betting strategy - a heuristic if you will - that would have made money had you employed it in the previous three seasons, which is the only period over which I can reliably test it.
To use the strategy you need just three (arguably four) pieces of information about a game:
- which team is the home team (true or notional)?
- what's their price (and are they favourites)?
- are they higher up on the competition ladder?
To use the strategy you need just three (arguably four) pieces of information about a game:
- which team is the home team (true or notional)?
- what's their price (and are they favourites)?
- are they higher up on the competition ladder?
Thursday, January 15, 2009
The 2009 Tipsters Have Arrived
I'd like you to think for a moment about the most recent non-trivial decision you made.
Do you recall how you went about making it? Did you, as the textbooks suggest, gather every piece of information that you could about each available option, carefully weight these pieces of information based on your fully-enunciated preferences, and then make the obvious choice, glowing with the self-satisfaction that comes with making a choice that's rational, completely explicable and just plain right? Oh, you didn't? Well join the very long queue labelled "most of us".
What you almost certainly did instead was to employ what the academics call an "heuristic" and what the rest of us call a "rule of thumb", a way of intelligently using what you know - some of it highly relevant, some of it barely so - to arrive at a decision that is satisfactory and, in its own way, sensible.
This year I've retired from tipping all of the MM and SMM Models, the Uber Model and the Simplified Uber Model, and replaced them with heuristics. Some of these heuristics you'll recognise - BKB, CTL and Shadow - but the remainder are new. If you want to know more, right-click the "Tipsters" link above and download the PDF, which describes the heuristics we'll be following this year (along with a revamped Chi and another new tipster, ELO, which is based on the Team Rating System that I developed last year).
Do you recall how you went about making it? Did you, as the textbooks suggest, gather every piece of information that you could about each available option, carefully weight these pieces of information based on your fully-enunciated preferences, and then make the obvious choice, glowing with the self-satisfaction that comes with making a choice that's rational, completely explicable and just plain right? Oh, you didn't? Well join the very long queue labelled "most of us".
What you almost certainly did instead was to employ what the academics call an "heuristic" and what the rest of us call a "rule of thumb", a way of intelligently using what you know - some of it highly relevant, some of it barely so - to arrive at a decision that is satisfactory and, in its own way, sensible.
This year I've retired from tipping all of the MM and SMM Models, the Uber Model and the Simplified Uber Model, and replaced them with heuristics. Some of these heuristics you'll recognise - BKB, CTL and Shadow - but the remainder are new. If you want to know more, right-click the "Tipsters" link above and download the PDF, which describes the heuristics we'll be following this year (along with a revamped Chi and another new tipster, ELO, which is based on the Team Rating System that I developed last year).
Thursday, January 1, 2009
Happy New Year and Welcome to MAFL Online
Hi everyone and welcome to a new footy season and welcome to MAFL Online.
Right now I can't tell you exactly what will appear here or how it will differ from the e-mail newsletters that you received last year except to say that MAFL Online will be the place you'll come to get the week's tips, wagers and results.
Right now I can't tell you exactly what will appear here or how it will differ from the e-mail newsletters that you received last year except to say that MAFL Online will be the place you'll come to get the week's tips, wagers and results.
There is, however, one important piece of information that's available now on this site: information about the Funds that will operate this year. It's a PDF and you can download it by right-clicking on the "Fund Descriptions" link in the top left-hand corner of the page.
Almost certainly, a bit later in the year there'll be another website, MAFL Stats, that will carry other of the regular features such as the Alternative Premierships, Team Ratings, Surprisal information as well as the one-off pieces I occasionally write with a statistical flavour. I'll let you know when there's something on that site that's worth a look.
Until then, if you're thinking about investing this year please have a look at the PDF.
(BTW: I've turned on commenting for now, so feel free to leave a comment on the site if you'd like to)
Subscribe to:
Posts (Atom)