Thursday, December 18, 2008

Keeping in-game statistics

Back in school, I played in the school's 2nd XVIII team in Aussie Rules. For years the schools in B-Town had been um-ing and ah-ing about establishing a seconds competition, and in 2000 they finally bit the bullet and started it. As was the case with most other teams, my school's 2nd XVIII contained some Year 10 students being groomed for the 1st XVIII, and mostly Year 12s who either weren't good enough for the 1sts or simply didn't want to train as much. It was almost a social comp, since we all got along with the other teams, and only maybe six or so players from either team putting in a real effort to press for selection in the 1st XVIII. Where am I going with this story?

At the request of the 1sts coach, we kept stats in our games. Kicks, handballs, marks, tackles, goals, behinds and hit-outs for the ruckmen. They were collected by whoever happened to be on the bench at the time, and the next day they were given to the 1sts coach to go over. Quite often, the kid who got 30 possessions simply by following the ball around the field would get picked over the kid who got 15 possessions, but stopped three or four certain goals with his work across the half back line.

Which brings me to ultimate. I've kept stats before and I was given a bit of paper with four categories - goals, assists, blocks and throwaways. The stats on the paper and what I observed in the game often told a very different story. Player A caught 8 goals and Player B got 5 blocks - they must have played a great game! But they simply were in the right place at the right was Player C who got the break throws that led to goals, and the pressure on the mark from Player D who forced the risky throws.

So it got me thinking about what stats I would like to see for a team that I am the coach or selector for. Assuming here that I'd have sufficient helpers on hand to keep track.


Goals - this one's a given, really. It's the easiest for a non-player to follow.

Assists - similar to goals.

TTA (throws to advantage) - this one is for completed hucks and break throws, as these type of throws are what sets up goals and what defenders are trying to stop.

Throwaways - the hucks that go nowhere, the dump throws that get turfed...clearly the fault of the thrower. Getting handblocked would go in here.

Unforced drops - exactly what it says on the tin.

Forced drops - these are drops from contests. This doesn't count if the defender gets a clean block...only if a receiver gets their hand to it but doesn't take the catch. Drops after so-called "phantom Ds" would count here.

Contested catches - counting the number of catches per game wouldn't really tell us anything. Nor would the one-on-one catches, given the number of times they happen every point. What would be interesting to know is if someone regularly comes down with the disc from a pack-grab. So a contested catch would be against two or more other players (ie: at least 3 players going for it).


Blocks - hitting the disc to the ground before the offense gets it. Straight forward.

Intercepts - catching rather than blocking the disc. Worth more than blocks as the attack can get going straight away.

Forced turnovers - plays which lead to forced drops for the offense. Also includes throwaways on high stall counts, and stalling out on the mark.

There's plenty of stuff you could keep stats on, but these are the key ones I'd want to know about. Any others you would want to see?


JdR said...

Hi Simon,

First, it must have been hard to take that Coach seriously, if he selected players solely on stats (sight unseen). What a clown?

The other question for Ultimate stats is who is going to take them? AFL and cricket have strong histories of stats-takers at many levels, but I find that Ultimate fanciers just want to play.

The stat I think you're missing too is just a straight 'touches'.

Simon Talbot said...

It was a very informal setup, the ol' 2nd XVIII. We didn't have a coach, only a supervising teacher who knew very little about footy. The 1sts coach didn't only select based on stats, but it was a pretty big part of his process. He'd talk to some of the older players about who played well, and also observed at training who was putting in the hard yards.

Obviously we're not going to suddenly find a bunch of people who want to take some uniform stats, but it'd be up to teams looking for a bit more analysis into their game. I know HoS keep stats for scrimmage games at training (done by an injured player or whoever is on the sideline), and I imagine there's other teams who do it too.