Some GF and Flag stats for the final 8

[Pies4shaw]

Here's an interesting article from the AFL:

https://www.afl.com.au/news/2019-09-12/finals-flops-the-redhot-teams-who-crashed-out-of-september

Initially, I was keen on the idea that only 4 "minor premiers" from 1983 on had made the GF after losing their first final. On a moment's reflection, of course, that tells you nothing unless you know how the minor premiers have gone more generally.

Anyway, I was interested, after reading that, to have a look at how the "minor premiers" have gone overall (rather than just after losing their first final) since the final 8 was introduced.

The 8 started in 1994. From my review of other stats, it looks like from 1994 onwards the "minor premier" has won the flag on 8 occasions (1994, 1995, 2000, 2004, 2006, 2007, 2010 and 2013) out of 25: https://afltables.com/afl/teams/allteams/seasons.html

For completeness, the team finishing 2nd in the home and away on and from 1994 has won 9 GFs, the 3rd finisher has won 5, 4th has won 1 (Adelaide against St Kilda in 1997 - good luck, Collingwood!), 5th 1, 7th 1 and 6th and 8th have not won a GF. The same club won from 4th and 5th (Adelaide in 1997 and 1998) - and it is probably fair to say that it was the best finals side by some margin in both years (and maybe it under-performed during the year). In the last 20 seasons, though, only 1 team has won the Flag from outside the top 3 - Footscray in 2016.

On making the Grand Final, the stats are also sobering: in those 25 years of the final 8, teams finishing 1st to 3rd have filled 43 of the 50 GF spots. The only teams that have made the GF from below 3rd in or after 1994 are Geelong (1994), Adelaide (1997 and 1998), Carlton (1999), Collingwood (2002), Sydney (2006) and Footscray (2016). On the other hand, 3 of those 7 campaigns were successful. And in another case (2006), the GF could easily have gone to the lower-ranked team.

The other question that interests me is how teams have performed after winning their "week’s rest" final (the AFL mucked about with the system for a few years from 1994, so that there was always a double chance but the question of who went direct to PF on a week’s rest was more open than it is now). It's a little harder to assess that - because the system from 1994 to 1999 gave the teams that finished top 2 a week off and then gave them a big advantage in their next game. Anyway:
- In 1994, West Coast and Geelong both won their "big" final for a PF spot and went on to play in the GF.
- In 1995, ditto Carlton and Geelong.
- In 1996, ditto Sydney and North.
- In 1997, St Kilda got to the GF but Footscray did not (lost the PF by 2 points to Adelaide).
- In 1998, North got through but Footscray didn't (flogged this time by Adelaide).
- In 1999, North did but Essendon didn’t (lost the PF by 1 point to Carlton).
- In 2000, Essendon and Melbourne both got to the GF.
- In 2001, Essendon and Brisbane both got to the GF.
- In 2002, Collingwood and Brisbane both got to the GF.
- In 2003, Collingwood got to the GF but Sydney didn’t.
- In 2004, Port and Brisbane both got to the GF.
- In 2005, West Coast got to the GF but St Kilda didn’t.
- In 2006, Sydney got to the GF but Adelaide didn’t.
- In 2007, Geelong and Port both got to the GF.
- In 2008, Geelong and Hawthorn both got to the GF.
- In 2009, Geelong and St Kilda both got to the GF.
- In 2010, Collingwood and St Kilda both got to the GF.
- In 2011, Geelong and Collingwood both got to the GF.
- In 2012, Hawthorn and Sydney both got to the GF.
- In 2013, Hawthorn and Fremantle both got to the GF.
- In 2014, Hawthorn and Sydney both got to the GF.
- In 2015, West Coast got to the GF but Fremantle didn’t.
- In 2016, Neither Geelong nor GWS made the GF.
- In 2017, Adelaide and Richmond both got to the GF.
- In 2018, West Coast got to the GF but Richmond didn’t.

So, 41 of the 50 teams that won their way straight to the PF made the GF. In only one year (2016) has neither “rested” team got to the GF. In the years when one rested team made it to the GF and the other didn’t, the Premiership outcome has been generally in favour of the team that didn’t have the rest (Adelaide won the Flag in 1997 and 1998 under a very different system; Carlton lost in 1999; Brisbane won in 2003; Sydney won in 2005; West Coast won in 2006; Hawthorn won in 2016; and Collingwood lost in 2018).

To date, Collingwood has made the GF every time it has won the "week off" before the PF.

The sample size is too small to identify any statistically significant statements about how we’d want the other preliminary final to go – but I guess you’d have to think that if Richmond lose and Collingwood win, there’s apparently a greater prospect than you might want to hear about that Collingwood will find it hard to win against the “non-rested” opponent. So, if we get there, I suppose we want to be playing Richmond.

[RudeBoy]

Or not.

[Pies4shaw]

Why the eye roll, RB? I put quite a lot of work (I had to do the year by year result figures myself) and thought into that post.

[RudeBoy]

It was an excellent post P4S. I was simply trying to be funny.

Despite the extensive and informative statistical analysis, I'm not sure my preference would be to face the Tigers in the Grannie. But that's just me.

[23 YIPPEE!!!]

Very interesting post I was going to ask the same thing is it better to have the week off before the prelim also I would like to know out of the top 4 which teams have gone out straight sets also another interesting stat to know.

[Pies4shaw]

[David]

Thanks for your work on this, P4S! One interesting fact I've gleaned from your analysis is that, in the last four years, four out of eight teams that won a qualifying final have gone on to lose a prelim, whereas before that, that didn't happen to a single team for eight years in a row. Now, that might just be an example of random statistical distribution in action, or evidence that the competition is becoming more even, but I wonder if there's another explanation: perhaps the pre-finals bye, introduced in 2016, has been conferring less advantage on qualifying final winners?

If so, the two possible causes are that a) it's easy to be off your game if you've only played one match in the space of four weeks* (as will be the case for us when we line up on Saturday the 21st); or b) that the break means there's less of a disadvantage for teams that haven't had a break during the finals, as they're already pretty well-rested. Whatever the case, it's an interesting statistic, and it's hard to imagine that it's just a coincidence.

*Interestingly, I've heard this theory ages ago about why Essendon lost the 1990 grand final – because of the West Coast draw, they just hadn't played enough football leading up to the semi-final (it was their first game in three weeks), which we won by a significant margin. So maybe there really is something to it? (Having said that, teams that get week 2 off still have reached the grand final 50% of the time in the last three seasons, so it's not like we're necessarily at a disadvantage now; just that the huge advantage that seemed to exist before has disappeared.)

[Pies4shaw]

The bye week started in 2016, so it's actually 3 out of 6. I think your points are well-made - but I think the sample size is way too small to enable us to do much more than guess.

[David]

Yep – for this theory you'd have to consider the 2015 results an anomaly. But that still represents a shift from 87.5% from the implementation of the new finals series until 2015 to 50% since the bye was introduced. I agree it's a small sample size, but it is interesting that the new arrangement in 2016 was suddenly accompanied by this unprecedented event of both qualifying final winners losing their next matches. It's just conjecture, but interesting nonetheless.

[Pies2016]

An enjoyable ( and statistical ) stroll down memory lane 😀

I’m a firm believer coming into a finals series, that August form stands for more than August ladder position. The key to going deep into September is to be playing your best footy ( subject to availabilities ) in August. There is very little chance of winning a flag if you aren’t near your best at the end of the home and away season.
If I bang on about this long enough, eventually some posters will realise it doesn’t matter how bad you look winning by a point early in the season, just as long as you win. Having said that, when it comes to September, it’s obviously better to be on top of the ladder and playing good footy, than to finish eighth and be playing good footy.

When they introduced the bye weekend to all finalists at the end of round 23, it flipped things on its head ( including a couple clubs only playing two games in a month ) It will be interesting to see if the odd team wins the flag from outside the top four going forward.

[Pies4shaw]

Speaking of conjecture, I was just trying to analyse the recent form of Geelong and West Coast against each other. The crazy thing is that the last time they played each other at the MCG was 2011. In every match since, they've played at Kardinia Park or Subiaco/Perth Stadium. Games have gone with "home ground" advantage in those 8 matches , except for a 2013 game at Subiaco, in which Geelong thrashed the Wiggles. That's over 6 years ago. Their recent "form" against each other is probably quite irrelevant to assessing their chances in a match at the MCG.

Their respective form at the MCG against all opposition is interesting, though - the Wiggles have won 5 out of 6 at the 'G since they started playing decent footy last year. Their loss was against Richmond by 6 points in round 22. Geelong have won 4 of 6 there, this year - but lost the last 2. In 2018, Geelong lost their last 5 games in a row at the 'G. So, another way of looking at it is that they've lost 7 of their last 11 at the 'G (including both finals) - and their 4 wins all came in a rush in the first half of 2019, before their form declined.

It may also be that Geelong's poor finals record in recent years has more to do with playing at the MCG than "finals form". They've had 8 finals losses in 11 games at the MCG since they won the flag in 2011.

[Pies4shaw]

[David]

^ Me too, as long as it's not the Giants or the Eagles!