Review The Stats Thread 2011

[Jack2Johnston]

Is it possible to take the Gold Coast rnd 2 game for Carlton out of the picture and redo the stats?

 

[grimlock]

Is it possible to take the Gold Coast rnd 2 game for Carlton out of the picture and redo the stats?

Which stats exactly?

 

[Jack2Johnston]

Pretty much all of them, basically remove the Goldcoast game completely, and see what inside 50 differential, scoring differential, contested ball differential, clearences differential, inside 50 per goal, and disposal per goal.
For accumulative stats like tackles or hitouts take an (average - goldcoast) and use that in place of GC game as a sort of dummy variable.

If its too much work based on the way your equations/data is stored is all good, just think when a side triples another sides score you can be fairly confident that they are going to take a huge lead on those stats, and after only 11 games it would still be throwing out the accuracy by a bit imo.

 

[Dangas]

Pretty much all of them, basically remove the Goldcoast game completely, and see what inside 50 differential, scoring differential, contested ball differential, clearences differential, inside 50 per goal, and disposal per goal.
For accumulative stats like tackles or hitouts take an (average - goldcoast) and use that in place of GC game as a sort of dummy variable.

If its too much work based on the way your equations/data is stored is all good, just think when a side triples another sides score you can be fairly confident that they are going to take a huge lead on those stats, and after only 11 games it would still be throwing out the accuracy by a bit imo.

Over 11 rounds, GC Stats wont change the average much at all as it is an "average"

 

[Jack2Johnston]

Over 11 rounds, GC Stats wont change the average much at all as it is an "average"

Actually it does, any reasonably statistically significant conclusion usually requires at least 30 data samples for specifically that reason, that an outlier will affect the average too much with any less data.

Just looking at Carltons, for and against should be enough to prove that,

Currently 1112 : 808, and 137% if we take out the GC game, that falls significantly to 941 : 756, and a percentage of 124%, a 13% discrepancy.

 

[grimlock]

• Biggest positive differential in contested marks (23 vs 7) for the year
• Biggest negative differential in bounces (6 vs 23) for the year
• Lowest number of free fors (11) and biggest negative differential in free kicks (11 vs 24). Hi Leg End!
• Only the 2nd time this year we have lost the tackle count (70 vs 88). The other time was, yep you guessed it, against Carlton earlier in the year (88 vs 106).
• Ted spent only 79% TOG, I say only because he was our leading TOG% player before the round.
• JPK had a game high 8 assists, but sadly no assists. He also had 20 contested possessions and 9 clearances, also a game high.
• We had more inside 50s (46 vs 42) but struggled to convert them into goals (10 vs 15) Carlton's rate of 15 goals from 42 inside 50's (35.71%) was the highest conversion rate by any of our opponents this year.
  
  Pies preview tomorrow!

 

[grimlock]

• Our tall defenders will have their hands full: Travis Cloke leads the AFL in contested marks (4.1/game), Cloke, Chris Dawes and Leigh Brown are all in the top 50 in Marks inside 50. Dawes is 7th in frees for which would suggest his big frame is an issue for defenders who infringe on him consistently.
• Scott Pendlebury is the complete midfielder - both inside and outside. Wins the hard ball (2nd in contested possessions, 2nd in tackles, 2nd in clearances) but also uses it well (2nd in inside 50s, 4th in assists, 4th in disposal efficiency)
• Heath Shaw plays very loose (1st in uncontested marks, 3rd lowest contested possession % and 3rd in uncontested possessions) but gets a lot of it (4th in disposals and 1st in rebound 50s)
• Alan Didak has had a poor season, despite getting 66.2% of his possession uncontested (4th lowest at the club) his efficiency has been atrocious (62.1% which is 3rd lowest at the club)

 

[Dangas]

Actually it does, any reasonably statistically significant conclusion usually requires at least 30 data samples for specifically that reason, that an outlier will affect the average too much with any less data.

Just looking at Carltons, for and against should be enough to prove that,

Currently 1112 : 808, and 137% if we take out the GC game, that falls significantly to 941 : 756, and a percentage of 124%, a 13% discrepancy.

Fair enough

 

[grimlock]

We can wax lyrical about all the categories that Collingwood dominate in, but to me we only need to look at two key areas:

• Collingwood are ranked #1 in converting inside 50s to a goal (28.6%) and also ranked #1 in limiting opposition teams in converting their inside 50s to a goal (20.0%). The Swans are ranked #16 (22.2%) and #4 (23.0%) in those respectively categories. Similarly, they are ranked #1 in goalkicking accuracy (57.9%) putting to shame the Swans' profligate efforts ranked #16 (47.0%)
• Collingwood are ranked #1 in inside 50s (62.1) and while the Swans have improved in this stats from 2010, ranked #10 (52.3) we seem to work so hard for our goals, reflected in the stat 'disposals per goal'. Sydney are ranked #11 (28.7) while Collingwood are supremely efficient #1 (21.8). Given the Swans are dead last in the possession count, this is a sad indictment on our dysfuntional forward line.
• Collingwood measure up really well in the areas in which Sydney are supposedly strong in: clearances (41.6 v 40.8), contested possession (160.5 v 150.4), opposition disposals (346.5 v 345.4) and opposition inside 50s (47.3 v 48.5). However they blow the Swans out of the park in disposals (386.0 v 332.8), marks inside 50 (13.4 v 11.6), uncontested marks (77.9 v 61.5) and disposal efficiency (72.8% v 69.5%)
  
  Sigh, could be a long and cold night out at Stadium Australia

 

[matt18]

uve got Sammy Reid at 51 behinds on the swans stats haha

 

[grimlock]

He might get there at this rate. What was it, another 2 behinds tonight?

 

[grimlock]

Swans summary


Crows summary

 

[AlibiMonday]

Nice work Grim. Is this a spreadsheet or database that use to manage these?

 

[grimlock]

Nice work Grim. Is this a spreadsheet or database that use to manage these?

Yes, just a spreadsheet at the moment. I have had thoughts about a DB, but that will have to wait until next year.

 

[grimlock]

Summary



What we're better than them at
Mostly everything, but most importantly tackles - Sydney are 1st (81.3) and Adelaide are last (59.5), clearances - Sydney are 3rd (40.5) and Adelaide are 13th (35.3) and contested possessions - Sydney are 5th (150.5) while Adelaide are 16th (135.4)

What they're better than us at
Goalkicking accuracy - Adelaide 7th (53.8%) Sydney 16th (48.0%). Watch out for Paddy Dangerfield who has a terrific 18 goals 4 behinds this year at 81%

What we're both good at
Forcing ballups and throw-ins - Sydney ranked 1st and Adelaide ranked 3rd. Expect plenty of work for Seaby/White/LRT and Jacobs/Tippett/Moran.

What we're both terrible at
Converting inside 50s into goals - Sydney are 16th (22.9%) Adelaide 15th (23.6%). It promises to be one frustrating night for both sets of fans

 

[grimlock]

Swans summary


Suns summary


Stop Ablett.

 

[grimlock]

What we're better than them at
Almost everything! The Suns are dead last in many categories, especially in the defensive ones - they concede too many goals too easily and efficiency in the offensive half - they struggle put goals on the scoreboard.

What they're better than us at
Clearances - The recruitment of Ablett, Rischitelli, Harris and Brennan and the talent of Swallow serves them well in this category. They are the number 1 clearance side in the competition and the number 2 side in clearance differential.

What we're both good at
Besides clearances mentioned above, not much else because the Suns just don't rate well in the rest of the categories.

What we're both terrible at
Goalkicking accuracy - Both teams are in the bottom 2 in goalkicking accuracy.
Uncontested possessions - For different reasons, Gold Coast just don't get enough of the ball while Sydney's balance of possessions lie in the contested area.

If we don't beat them then we don't deserve to play finals. Their main strength is in their midfield and excluding Ablett I think we match up well on them. Their defence is leakier than a Labor Party caucus meeting and hopefully this will translate into a form boost for our goal-shy forwards Reid and White. Down back, our defensive unit has been a strength all year and with more changes expected to their forward setup, we should be able to contain them to a low score.

 

[grimlock]

Swans summary


Freo summary


Be thankful they are missing Sandilands and Mundy.
Mayne is their main marking forward, he's in doubt for this game.
Fyfe is an absolute gem, 2.1 contested marks for a man his size is extraordinary

 

[grimlock]

What we're better than them at
Defence - We are 2nd in both opposition inside 50s into scores (%) and opposition disposals per goal. Freo are 14th and 12 respectively.
One-percenters - Swans are first with 60.0 / game while Freo are dead last with 42.9 / game.

What they're better than us at
Marks inside 50 - As mentioned in the post above, Chris Mayne leads the way for them in this stat with 32.
Hitouts - but without Sandilands, the discrepancy won't be as great.

What we're both good at
Tackles - First and 3rd respectively. With the rain and at the SCG, we could be looking at world record numbers, both individually and as a team.

What we're both terrible at
Goalkicking accuracy - We are 16th in this cat, we played the last placed team last week, we now play the one above us this week.

 

[grimlock]

No time for analysis

 

[grimlock]

Swans


Saints

 

[grimlock]

What we're better than them at
Clearances - Despite their reputation for also being a tough, contested side, they are dead last in clearances (34.4), while we lead the competition in this area (43.3). If Mumford can get on top of Big Boy McEvoy again, then we should clean up in this area.
Contested possessions - Again appearances can be deceiving. The Saints actually are ranked 3rd in uncontested possesions, meaning they rank 2nd last in contested possession rate (36.7%) compared to our league leading (47.1%). It does highlight that despite what all the experts are saying, the two teams do play quite different games.

What they're better than us at
Ball use - Saints are much cleaner with the ball in hand, finishing 3rd in disposal efficiency (73.9%), while the Swans are last (68.7%). Again this discrepancy can be explained by the highly contested style of game preferred by the S
Forcing opposition errors - Courtesy of their well-developed press, they are up there with the likes of Geelong and Collingwood.

What we're both good at
Defence - Sydney ranked 6th (11.1) and St Kilda 4th (10.8) in goals allowed per game. This is also highlighted in the opposition inside 50 convertion rate - Swans 3rd (22.5%) and Saints 1st (21.5%)

What we're both terrible at
Scoring efficiency - Converting inside 50s to a score, Saints rank 13th (25.1%) and we rank 16th (22.5%)
Giving away free kicks - Razor will be licking his lips, he'll have plenty of opportunity to blow his whistle. Swans 12th (20.8) and Saints 13th (21.5) in frees against

 

[Dangas]

All in all Grim, going to be a battle right 'til the end

 

[grimlock]

Hawthorn summary

 

[grimlock]

Summary


As it was the case in the Round 9 game, this matchup brings together two differing styles of football. Hawthorn are the most uncontested team in the league, while Sydney are at the opposite end having the most contested possessions and highest contested possession rate by quite some distance. I mentioned at the time that the team that can impose their style of play will most likely come away with the victory - the Swans started well but their forward pressure faded as the game went on and guys like Suckling, Birchall and Guerra were almost unchallenged as they brought the ball out of defence in the 2nd half as the Hawks ran away with it.

Without Buddy (and Roughy) the Hawks will depend on a coterie of smaller forwards in Breust, Rioli, Osborne and Isaac Smith. And despite all the criticisms their defence received last week against Geelong, they are actually one of the best defensive units going round, conceding the least number of marks inside 50 to the opposition and conceding the 2nd least number of goals behind Collingwood. Going in with a tall forward line (Reid, White, Spang and Goodes) will work if those talls can mark it but it is a double-edged sword as if the ball goes to ground, it will play into Hawthorn's rebounding defenders' hands and their efficient and effective transfer of play up the ground.