How it Works

Trophies, league standings, goals scored, the best players, or even fair play awards… these are all fine ways to measure and compare teams. But we’ll give you a better one – TPO Rankings! Most of what you see on this page revolves around the TPO Rankings, so it’s only fair that we explain ourselves and open up our methods to constructive criticism.

I should say up front that neither Cody or I (Jake) are statisticians, and although I’m a bit of a nerd when it comes to numbers and making them do cool things in Excel, we are in no way qualified to produce these rankings. But everyone loves a good rankings system (especially when it’s purely for interest’s sake!) so we went ahead and did it anyway! With that said, let’s get into it.


The aim of the TPO Rankings is to include as many Australian Men’s football teams as possible, and as many historical results as possible. The rankings currently include 362 clubs from the top three tiers in Australia, with results from 2013 onwards (2005 for the A-League). The next phase of development for the TPO Rankings will be to expand to over 1,000 teams, 100,000 games, and 40 years of history.


Everyone who has a passing interest in football (and if you’re reading this there’s a good chance that includes you) has heard of the FIFA World Rankings. We like to think of the TPO Rankings as the same concept, but instead of ranking National Men’s football teams, it ranks Australian Men’s football clubs (i.e. A-League, NPL, State League, etc.).

Each team has a TPO Score, based on its past results, and the team with the highest TPO Score is #1 (i.e. currently the best team in the country), and so on down the rankings. One way to look at it is that the TPO Score is a measure of relative strength. I.e. when two teams play each other, they each have a TPO Score which indicates their strength. We can use the relative strength between the two teams to determine some probabilities about the outcome of that game.

And it’s at this point that I need to introduce you to ELO. I promised Cody that I’d keep the explanation as pain-free as possible, so here we go…


Being a rankings system with lots of numbers, you knew there’d be some complicated algorithm or formula behind it all didn’t you? Well you’re only half right. There’s a formula that calculates all the rankings, but it’s not overly complicated.

The ELO system was developed by Dr. Arpad Elo (hence the name) and is used by the international chess federation to rate chess players. In 1997 it was adapted to international football by making a few tweaks to account for football-specific variables, like home team advantage and goal difference. We have taken that football-specific ELO formula, made our own amendments (and tested to ensure they improve the rankings), and this is what the TPO Rankings are based on.

So how do they work? It sounds weird, but I find it useful to explain (and understand) how ELO works if I think of the whole ELO system as a person (let’s call him Dr ELO). Bear with me.

Picture Dr ELO sitting on a big, high chair looking down on everyone, and when two teams want to play a game of football they first go to Dr ELO. And when they go to Dr ELO, they are each carrying a sack filled with tokens that they’ve won from previous games (the total number of tokens in this example equals each team’s TPO Score) – the bigger the sack of tokens, the stronger the team.

Before the game starts, as the two teams are standing before him, Dr ELO will look at the sack of tokens each team is carrying (relative strength of the two teams) and make a prediction about the result of the game, in the form of a probability. For example, he might say, ‘I give Team A a 75% chance of winning and Team B a 25% of winning’.

After each game takes place, Dr ELO then compares his predictions to the actual result, to see who over-performed and who underperformed his expectations.

Dr ELO then forces the underperforming team to take some of the tokens out of its sack and hand them to the over-performing team and asks them to move on so he can focus on the next game.

The reason I use this example is to show that, although there’s a formula that pulls the rankings together based on thousands of games, if you simplify it to its basic level, all the ELO system really does is compare the actual result to the expected result, and transfers points from one team to the other.


This is the part where the ELO system gets a little more complicated, and where TPO puts its own spin on the standard formula. There are several factors that impact how many points are transferred, and I’ll explain each of them in more detail in the future, but for now, here is a quick summary:

• The difference in TPO Points between the two teams. As we just discussed, the difference in TPO Points determines the probability of each team winning, and this is compared to the actual result of the game. The further the actual result is from the expected result, the more points that are transferred between the teams. E.g. a team that has a 90% probability of winning won’t take many points from their opponent if they win, because it’s expected. But if the team has a 10% chance of winning and does (an upset!), more points are won/lost.

• Home Advantage is also seen to give a boost to the team playing at home, so we factor it in when calculating the probability of each team winning.

• The margin of victory – the score. A bigger win means more points are won/lost.

• The significance of the game. Simply put, we’ve placed slightly more significance (in terms of points transferred) on all Finals games and Knockout Cup Competitions. It’s only a small difference, but it means more TPO Points are transferred in those games than normal.

• Games between teams from different Leagues or different States (usually Knockout Cup games). This is where things get interesting for me! It’s a topic that needs an entire post of its own, so I’ll keep it brief. But suffice it to say that when Inter-League or Inter-State games take place, more points are transferred, and not just between the two teams playing – it also has a small effect on other teams in the respective League or State.


The TPO Rankings is possible thanks to those out there updating scores from around the country. The majority of our data is from the Socceraust website.