It's that time of year again when we need to ruminate over the draw and decide which team, if either, is really the home team for each game.
This year these decisions are even more complex than they usually are since, according to the AFL's official home team designations, 8 teams have two home grounds and 2 more have three home grounds across the season. The nomadic teams include 4 that play only a single game at one of their designated home grounds. I don't think it makes sense to designate a venue as home for a team if that team's players would, unaccompanied by team management, need to ask directions to find it.
Sensibly determining home team status is very important for MAFL as this status serves both as an input into most of the underlying fund algorithms and as a determinant of whether or not a wager should be made on the team selected by these algorithms. The question then is: from a modelling perspective, what makes home status important?
To me, home status matters because it represents advantage. This advantage can come in either or both of two ways. Firstly, the home team can reasonably expect greater crowd support than its opponents when it plays at home and, secondly, the home team should be more familiar with the ground than will its opponents.
With these ideas of advantage in mind and a guiding principle that 3 designated home games should be enough to breed familiarity, if not contempt, for a venue, I've felt compelled to alter the AFL pronouncements on home team for seven games as set out in the table below.
Let's consider the merits of each of these changes.
In the Round 8 clash between the Dogs and the Swans neither team will have played at Manuka Oval earlier in the season and, while Canberra is nearer Sydney than Melbourne, it seems unlikely that the crowd size will be sufficient for any pro-Swans bias to have significant affect on the outcome. (Manuka Oval pulled crowds of only about 7,500 and 12,500 in the two matches there last season.) So, neither team's at home.
Similarly, in the Round 9 match pitting Melbourne against Port Adelaide at Marrara Oval, both teams will be playing there for the first time in the season. Neither team can count on any substantial crowd bias at the venue and the crowd is likely to be quite small, with last year's only match at the venue attracting a crowd of just over 11,000. So, another game without a home team.
Round 11 includes the pairing of Collingwood and the Dogs at Docklands, a contest that the AFL has designated a Pies home fixture. By that point in the season Collingwood will have played at Docklands twice, both away fixtures, and the Dogs will have played there six times, four of them as the home team. My view is that any crowd bias enjoyed by the Pies will be more than offset by the Dogs' familiarity with the nuances of the ground in season 2010, so I favour switching this to be a Dogs home game. This is the call I'm least convinced of this season.
Next we consider the Round 13 Cats versus Saints matchup which has been designated a Saints home game by the AFL thus ensuring at least some likely pro-Saints bias in the crowd. However, by that point in the season, the Cats will already have played 4 games at the G, 1 of them a home fixture, and the Saints will not have played there at all so, on balance, I think it's fairer to call this a Cats home game.
In Round 14, Collingwood takes on West Coast at Docklands in another clash deemed a Collingwood home game by the AFL. Doubtless, Collingwood will enjoy a substantial excess of crowd support in this game and the Pies will by that point have played three games at the venue compared to West Coast's two, but I still feel uncomfortable designating it a home game for the Pies when the familiarity difference is so slight. For better or worse then, I've made this a "no home team" game.
The Dogs face Port Adelaide in Round 16 in the season's second and final game at Marrara Oval. For reasons similar to those I've noted for the Round 9 Melbourne and Port Adelaide clash, I'm making this another "no home team" game, notwithstanding that it will be Port's second game for the season at the venue. One contest does not a home ground make.
Finally, Round 21 sees Carlton taking on Geelong at Docklands in a game designated home for the Cats by the AFL. This despite the fact that, by that point in the season, Carlton will already have played 8 games at Docklands, 6 of them at home, and the Cats will have played just one away fixture there. A Carlton home game for mine.
After making those changes the MAFL home team designations by team can be summarised as shown in the following table.
Carlton ends up with an extra home game, the only team to enjoy this advantage, by virtue of my treating its Round 21 clash with the Cats as a Carlton home game.
Collingwood winds up with just nine home games, two fewer than it could reasonably expect and the only team facing this predicament, a consequence of my refusal to recognise Docklands as a home ground for them.
Three other teams - Melbourne, St Kilda and the Western Bulldogs - are short-changed by a single home game, for each team a consequence of my transforming one of their AFL-designated home games into either an away game or a game at a neutral venue.
On such decisions let's hope that profitability does not hingeth.