Certified Legendary Thread Race for the flag, in squiggly lines

[Dazb86]

Great work. Probably less confusing without the squiggly lines.

 

[Final Siren]

Superb stuff. Don't be discouraged by many, it's quite intuitive and simple to follow.

Would love to see the timelines for individual clubs over multiple years (eg comparing Hawks 08,10,12,13, Geelong 07-13, etc).

Uggghhh don't tempt me, I'm supposed to be working...

 

[lamaros]

Just looking over the charts, I don't see many examples of teams dashing into the ideal zone at the end of the season. Premiers tend to spend a lot of time in there.

The only recent counter-example I see is Hawthorn in 2008, which didn't really get in there until the start of the finals series. But according to this model, Hawthorn shouldn't have won at all. The Cats lived in that ideal zone all season.


As per the OP, it's a weighted average that works like this:

1. Most recent game: 9%
2. 2nd most recent game: 91% of 9% = 8.1%
3. 3rd most recent game: 91% of 8.1% = 7.4%
4. 4th most recent game: 91% of 7.4% = 6.7%
5. 5th most recent game: 91% of 6.7% = 6.1%
6. etc

And it's 91% and not some other value just because that has been the best predictor over the last 20 years.

That Hawks wouldn't have won 8 out of 10 that game I reckon.

Looks like it's Sydney or Geelong this year then. I'm prepared to say a side as poor defensively as the Hawks will fail once more, that Fremantle is too poor offensively, and that Essendon just isn't good enough.

 

[vonn]

Brilliant work.

But it reminds me of a board game I use to play as a kid, can't remember it now...

 

[Final Siren]

Obviously Hawthorn aren't as good defensively as Sydney or Freo, but I was very surprised to see them ranked the same as North, who are absolutely woeful at defending.

I guess you could argue that the reason Hawthorn allows teams to kick scores (last 6 games: Ess 87, WB 76, PA 79, Gee 82, Bris 85, WC 103) is that it hasn't usually mattered - they're usually well in front anyway. Maybe when they need to, they'll crack down.

Although I don't see evidence of that. Even in close games, the Hawks have still let the opposition score.

They may just prefer to win shoot-outs.

 

[clogged]

A little while ago, I remember someone on the Port board, think it may have been RussellEbertHandball, said the best way to know which team would challenge for the premiership the next year is to look at % (and other factors ofc, but he stressed %, as it's a better indicator than W/L).

This sort of demonstrates that.

Don't necessarily disagree, but since 2000, the side with the best H&A percentage has invariably lost.

 

[Flukey!]

I guess you could argue that the reason Hawthorn allows teams to kick scores (last 6 games: Ess 87, WB 76, PA 79, Gee 82, Bris 85, WC 103) is that it hasn't usually mattered - they're usually well in front anyway. Maybe when they need to, they'll crack down.

Although I don't see evidence of that. Even in close games, the Hawks have still let the opposition score.

They may just prefer to win shoot-outs.

I think that is the point.
We are confident that we can outscore the opposition most days.

 

[clogged]

I like to chart things for no good reason, so I decided to make a scatterplot of teams from their scores for and against over the course of the 2013 season. It looks like this:

After Round 18​

You can probably draw a line between Sydney 2005 and Essendon 1993 and say that is the minimum cutoff for attack-defence balance needed to win a premiership. Both of those premierships were won in fairly open fields, as you say, St Kilda 'should' have won in 2005 according to your system, but they finished fourth. And the year Essendon won there were very good sides only a game and a half behind them that missed the finals (Geelong).

That plot generally agrees with the general consensus that Hawthorn, Sydney and Geelong are the most likely contenders. What is notable are the positions of Fremantle and North, who are just outside the cutoff. Given North sit well outside the eight, but have had a number of close losses where their opposition has come from behind, it would be interesting to see what would happen if you made them a 1 or 2 goal better side defensively.

 

[Brishawk]

Interesting work. Clearly a lot of effort has gone into this. Can you see in your data the eveloution of the game over time? For example, in 2005 the Swans appear to be an outlier but were all/most teams playing with a greater emphasis on defence over attack in that period? OR were they truely an outlier? To ask the same question in a different way, do you find any correllation between the emphasis on offence or defence within a season? For example, in some season are there more emphasis on attack etc. across the league or do premiers tend to defy trends and largely operate in the range you describe regardless of what else is happening in the league at the time?

 

[SBD Gonzalez]

Some mornings the guy in my mirror looks like your chart.

 

[Brishawk]

In thinking about this a bit more, do expect this type of analysis to retain its accuracy as we move towards greater management of player work loads both within a match and within seasons? Cats have been rotating players out with 'soreness' for several seasons now in a bid to be right by finals. I'd say anecdotally Hawthorn are better this year than last but I suspect statistically we were better last year (on this measure, not win/loss). I put this down to being able to rotate players out of the side and limiting work load during matches for specific players in order to achieve wins not margins.

 

[philohk]

Really interesting stuff FS, thanks a lot for the effort. I guess like all stats, these are a starting point for analysis rather than an end point. As a few have pointed out already, you need to consider the results in the context of the season. Where the Hawks are concerned, are we too easy to score against or do we just let junk-time goals in? We have had a few close ones, yeah, but then how much attention do you pay to early season form (I know your weightings include that to a degree)? Even if you say we have been too easy to score against, we seem to have tweaked our game and defended a bit better against the slingshot the last few weeks. And are we keeping something in reserve as opposed to spending our tickets fighting for top two last year?

It looks like we’re an outlier, but it also looks like the Swans are, though maybe to a lesser degree. Cats less so again – they seem to be just working into a sweet spot (though that move looks to be the result of one week, which was a thrashing of a bottom team). Swans have players to come back, true, but so do we, and we’ve been the most consistent of the three. What to make of it? Looks like a fantastic finals series coming up!

Also, in general, one interesting thing I see is that while the maxim has always been that good defence wins flags, the premiers here look to be more powerful offensively than defensively.

 

[Jarlcorbz]

Holy poo WC don't know wtf they're doing

 

[deflated]

Interesting work. Clearly a lot of effort has gone into this. Can you see in your data the eveloution of the game over time? For example, in 2005 the Swans appear to be an outlier but were all/most teams playing with a greater emphasis on defence over attack in that period? OR were they truely an outlier?

What is really odd to me is that WC 2006 are right in the middle of the pack but there wasn't an inch between the Swans and West Coast those years. Did Sydney have a style that was underestimated by this model or did West Coast 'play down' to the field?

Really interesting how Hawthorn, Geelong, Sydney and Freo are all in quite different parts of the graph. Always great to see good teams with contrasting styles go at it in finals.

 

[Doss]

Final Siren- fantastic work, thanks.

 

[drummoyne]

That is some damn fine spreadsheetsmanship!

 

[Final Siren]

You can probably draw a line between Sydney 2005 and Essendon 1993 and say that is the minimum cutoff for attack-defence balance needed to win a premiership.

That's certainly been true for the last 20 years. It's a little too broad to draw great insights from, though, since most years several teams cross that line... it's like saying finishing in the Top 5 is the minimum required to win the flag. It's true, but doesn't help much.

Both of those premierships were won in fairly open fields, as you say, St Kilda 'should' have won in 2005 according to your system, but they finished fourth. And the year Essendon won there were very good sides only a game and a half behind them that missed the finals (Geelong).

It was a very even field, yes. Although Essendon 1993 looks like the "worst" premiership on this chart, its competition were further back. Thanks to a late run, Geelong finished up not too far away from where Hawthorn is now in 2013, as a high-attack, low-defense team.

That plot generally agrees with the general consensus that Hawthorn, Sydney and Geelong are the most likely contenders. What is notable are the positions of Fremantle and North, who are just outside the cutoff. Given North sit well outside the eight, but have had a number of close losses where their opposition has come from behind, it would be interesting to see what would happen if you made them a 1 or 2 goal better side defensively.

North (and Geelong) have had a good few weeks, but if I'd done this chart a month ago, neither would look nearly as good. So I'm interested in whether they hold these gains over the next few weeks or if they fall back to something like their "true" position.

 

[Big Blood]

Nice work. Hope your boss didn't mind.

 

[The Gtrain]

Is it possible to see winner and loser of the same yr grouped together? I notice that Hawks-12 loss is very close to where they are currently and syds win is also close to where they currently are.

It would be good to see what happened head to head...

 

[lamaros]

You can probably draw a line between Sydney 2005 and Essendon 1993 and say that is the minimum cutoff for attack-defence balance needed to win a premiership. Both of those premierships were won in fairly open fields, as you say, St Kilda 'should' have won in 2005 according to your system, but they finished fourth. And the year Essendon won there were very good sides only a game and a half behind them that missed the finals (Geelong).

That plot generally agrees with the general consensus that Hawthorn, Sydney and Geelong are the most likely contenders. What is notable are the positions of Fremantle and North, who are just outside the cutoff. Given North sit well outside the eight, but have had a number of close losses where their opposition has come from behind, it would be interesting to see what would happen if you made them a 1 or 2 goal better side defensively.

North are coming off a month of playing GWS, Richmond, Brisbane, Carlton, and Melbourne, and are about to play Geelong, Adelaide, Essendon, Hawthorn and Collingwood.

They won't be in that spot come years end.

 

[lamaros]

Just looking over the charts, I don't see many examples of teams dashing into the ideal zone at the end of the season. Premiers tend to spend a lot of time in there.

What about teams dashing out? In 2011 I think you'd have had Collingwood clear favourites prior to round 24?

 

[clogged]

North are coming off a month of playing GWS, Richmond, Brisbane, Carlton, and Melbourne, and are about to play Geelong, Adelaide, Essendon, Hawthorn and Collingwood.

They won't be in that spot come years end.

As I read it, wins against weak sides don't tend to help you in that graph. Their win against Melbourne hardly moved the needle.

 

[lamaros]

As I read it, wins against weak sides don't tend to help you in that graph. Their win against Melbourne hardly moved the needle.

It moved it massively? Are we looking at the same thing?

 

[clogged]

It moved it massively? Are we looking at the same thing?

Guess we have different definitions of massive.

 

[Final Siren]

Interesting work. Clearly a lot of effort has gone into this. Can you see in your data the eveloution of the game over time? For example, in 2005 the Swans appear to be an outlier but were all/most teams playing with a greater emphasis on defence over attack in that period? OR were they truely an outlier?

I think 2005 is the worst of my squiggles, in the sense that it doesn't just get the premier wrong, but gets it wrong by miles. Near the end of the H&A season, St. Kilda and West Coast are within striking distance of the "ideal" central cluster of premiership cups, while Sydney and Adelaide are both low and right, slightly above Sydney's eventual finishing position.

So looking at that, I probably would have tipped all three of St Kilda, West Coast, and Adelaide for the flag ahead of Sydney. I mean, Adelaide finished with a superior offense and defense to Sydney. But they didn't even make the Grand Final.

I guess I can console myself with the fact that it was close (Sydney beating WC by 4 points, West Coast beating Adelaide by 16 at home), but still, whatever was going on in this season, I'm not doing a very good job of picking it.

To ask the same question in a different way, do you find any correllation between the emphasis on offence or defence within a season? For example, in some season are there more emphasis on attack etc. across the league or do premiers tend to defy trends and largely operate in the range you describe regardless of what else is happening in the league at the time?

Each year is its own beautiful squiggle, for sure, which is kind of fascinating (to me). Some interesting ones:

• Geelong 2007 is so far in front of everybody else from about Round 6 onward that they might as well have not played the season.
  
  
• In 1993 you can see West Coast "warming up" for 1994: it's the only defensively focused team, a long way from everyone else, but still too low offensively.
  
  
• 2000 was a very attacking year. You can actually tell this just by looking at the "Points For" in the final ladder. Essendon averaged 128 points a game during the H&A season, which is just insane. And Carlton, Melbourne, Brisbane, and to a lesser degree North Melbourne that year were all highly attacking teams.
  
  
• By contrast, 2003 looks very defensive, even though the premier, Brisbane, was mostly an attacking team.

• Not even Fitzroy 1996 plotted as low as GWS are right now.
  
  
• Low-offense + high-defense teams are very rare. Most years don't have anyone in the general area that won flags for Sydney 2005 and West Coast 1994, unless they're coached by Ross Lyon. The main exception is 2005, as mentioned above.