Opinion "Help me out where I need faith!" - The Statistical Data Thread

[Janus]

What did you do?

I may or may not have gently mocked a poster who said that the most amount of likes they've ever received on Big Footy was a post on this board calling me out, and then I said that Adelaide's gameplan sucks and we'll see who calls out who come finals time

 

[El Zorro]

I may or may not have gently mocked a poster who said that the most amount of likes they've ever received on Big Footy was a post on this board calling me out, and then I said that Adelaide's gameplan sucks and we'll see who calls out who come finals time

It doesn't sound so bad. You were off for a long time...

 

[OneGreatClub]

I may or may not have gently mocked a poster who said that the most amount of likes they've ever received on Big Footy was a post on this board calling me out, and then I said that Adelaide's gameplan sucks and we'll see who calls out who come finals time

Well you must have really got under their thin skins.. remember..Quiet please

 

[RussellEbertHandball]

Hurling People ( guys who like hurling but do stats for footy and other sports) produce stats similar to what Janus has.

The Race for the Eight is tightening up, and the Dogs are on a surge
MAY 25, 2017 ~ EDITOR

Over the last 7 weeks HPN has been tracking the fortunes of teams using our Team Ratings, and some early trends appear to be emerging. But before we get to them, lets look at how the ratings sit after 9 weeks of footy:

For the second week in a row, Port Adelaide hold the top spot on the HPN Ratings, just a sliver ahead of cross-town rivals Adelaide. GWS sit on their own between the surging top two, and the cluster of 11 teams below. After their bad loss to Essendon, West Coast have lost a lot of ground.......
......
The Eagles currently sit in the bottom third of the league in their ratio of clearances to opponent clearances, and inside-50s to opponent inside-50s. Most of their companions in the lower range for these measures don’t look like finalists at this stage:

Over the last seven weeks the Dockers are the biggest movers up our ratings, a touch ahead of the Bulldogs. It’s taken some time for the Fremantle’s new additions to fit into Ross Lyon’s system, but it appears they are loosely in the mix for the bottom end of the eight........
https://hurlingpeoplenow.wordpress....is-tightening-up-and-the-dogs-are-on-a-surge/

 

[RussellEbertHandball]

This is what they wrote after the Shanghai game
https://hurlingpeoplenow.wordpress....-good-so-why-are-we-being-distracted-from-it/

...This season has shown that any team can beat any other, and the race for the finals is truly up in the air.

The HPN Ratings for the week reflect this.

Port Adelaide have leapfrogged their local rival into first in the rankings, driven by their stellar midfield and inexplicably stingy defence. Port’s midfield is particularly well rounded, with the inside grunt provided by Wines, Ebert and Powell-Pepper tempered by the run and creativity of Wingard, Wines, Gray and Polec. Ryder and Trengove have proven to be a ruck combination that provides more value around the ground than any potential losses in raw hitouts, and they never fail to provide a solid contest under the ball......

https://hurlingpeoplenow.wordpress....-good-so-why-are-we-being-distracted-from-it/

 

[RussellEbertHandball]

These are some stats before Rd 8, which haven't changed much in Rd 8 and 9.
https://hurlingpeoplenow.wordpress.com/2017/05/12/who-are-the-kings-of-style-of-the-2017-afl-season/

Kick First, Ask Questions Later
The kick v handball question is an old and nuanced issue. About a decade ago the pervasive thinking amongst football clubs moved from kicking where no other better option exists to looking for an extra handball to find a better target by foot. As The Arc pointed out earlier this week, Carlton in 2017 is going a different way. The Blues are turning the clock back to a slightly happier time on Lygon Street.

For GWS, Everything Is A Clearance Sale
We don’t think every team puts equal stock in winning clearances. A won clearance can still easily result in a turnover from the subsequent possession, and lead to the ball being rebounded back the other way. Clearances aren’t a great midfield strength measure in isolation, as they don’t correlate terribly well across the competition with inside-50 ratios or even with, you know, winning.

Adelaide and Port Adelaide, for instance, are doing pretty well on the scoreboard and in generating a predominance of inside-50 opportunities despite being indifferent break-even clearance sides. Port in particular have a very high ratio of inside-50s to opponent inside-50s despite not winning clearances, suggesting an outside midfield game that is working well.

However, a very high clearance ratio in a good team probably suggests they regard it as a strength and a tactical tool. Such may be the case with GWS, whose clearance ratio is a league high 122% and whose dominant wins in 2017 all match with games where they destroyed their opponents in the clearance stakes. A lot of analysis of GWS focuses on the wave running they do with their powerful and skilled runners, and they seem able to achieve this out of stoppages at will if teams aren’t sufficiently careful.

What Fremantle’s ruck dominance means

How do teams generate clearances? One obvious answer is by having a good ruckman. Fremantle, the side with the best ruckman, are also a good clearance-winning team. The top five sides for clearance ratios all also have positive hitout ratios (but not the other way around) which suggests rucks do still matter if you care about winning clearances.

Fremantle’s 115% clearance ratio is almost certainly partly a product of the towering presence of Aaron Sandilands. Gold Coast, GWS, Collingwood and Brisbane all also have pretty good primary ruckmen securing hitouts, and this may be part of what is translating into their relative clearance success. The Demons are a strange case which should probably be ignored for the moment – they’ve gone from having a really good ruckman to not really having one at all, so their season totals reflect both those situations.

This doesn’t mean that hitout domination necessarily translates linearly into midfield success, however. North Melbourne, for instance, are sixth in hitout differential, but lose the clearance counts on average. Fremantle are the key case study to what ruck dominance means. The Dockers themselves are generating more than two hitouts for every opponent hitout, which is an insane degree of dominance. But they’re also not translating those hitouts as efficiently into clearances as that might suggest.

The brute force effect of utterly dominating hitouts is only giving Freo a modest advantage in terms of winning the ball away from the area around a stoppage. This may suggest there is a saturation point of ruck dominance beyond which ruck contests get too predictable, and midfielders too evenly matched, to gain further advantage.
https://hurlingpeoplenow.wordpress.com/2017/05/12/who-are-the-kings-of-style-of-the-2017-afl-season/

 

[Janus]

A chart showing the variance in contested possession numbers - the blue line represents contested possessions for, the red line contested possessions against.

 

[RussellEbertHandball]

Some stats I have been keeping this year.

The stats suggest to me the West Coast loss when we were dominate side, will cost us come the end of the season. The Geelong game we should have pinched. The most important stat was their disposal efficiency 75% to ours at 64%. Might have to dig these up.

 

[RussellEbertHandball]

The Arc on his ESPN Footy Forensics blog has done a 2 part series on the 18 teams so far this year. Part 2 looks at the Hawks to Western Bulldogs. The Port bit;


PORT ADELAIDE - DOMINANT BETWEEN THE ARCS
Back in 2014, when Port Adelaide missed out on a spot in the Grand Final by less than a kick, it was their midfield dominance that stood out. The 2014 Power recorded 1.2 inside 50s for every inside 50 they conceded - the league's best ratio that year. But then they fell back to the pack in 2015 and stayed there in 2016, recording about as many inside 50s as they gave up to their opponents. This year, their record between the arcs has soared, beating even their 2014 performance. The Power have recorded 1.3 inside 50s for every entry they've conceded so far this year, easily the best inside 50 ratio in the competition.

Why has Port's performance soared? There are undoubtedly a lot of factors. Paddy Ryder can ply his trade without watching for the third man-up, Sam Powell-Pepper has burst out of the blocks, and Ollie Wines is a clearance machine. But one big factor might be the resurgence of Brad Ebert's performance around the contest. Ebert has gone from averaging just 1.8 clearances per game last year to a whopping 4.8 this year, the second best in the team. He's also racking up 11.1 contested possessions each match, up from 7.5 in 2016, again second in the squad.

The Power are the real deal this year, winning games off the back of their strength in the middle of the ground. Neutralising Ebert might hold the key to taking them on.

http://www.espn.com/espn/feature/st...orensics-your-afl-club-key-game-trend-part-ii

I agree with the Ebert bit. I reckon I have given him votes in 8 of the 9 games and is leading my B&F votes. I will do a tally after tonight's game against the Hawks to see if my perception is reality. Also agree with the Ryder bit. I just don't understand how so called footy experts totally ignored the Ryder impact when they assessed our opportunities in 2017 before the season started when doing their top 8. As did many of our fans especially when no trades happened and a lot of people on here thought we were a big chance to finish bottom 4 and blow our 2017 first round pick traded to Brisbane.

 

[RussellEbertHandball]

From today's Tsier

http://www.goldcoastbulletin.com.au...s/news-story/c7c2d569d2232a42655365152f0726d3

“I LOVE that strong defensive football when we lock the ball in our forward half. The ball did not look like getting through us.’’With those words, coach Ken Hinkley summed up Port Adelaide’s record-breaking first half against Hawthorn last week when it held Alastair Clarkson’s team - triple premiers from 2013-15 - to its lowest-ever half-time score of 0.3.“That was pretty brutal football, wasn’t it, when you can do that to a side like Hawthorn who were coming off a really strong win (against Sydney). “You’ve got to be really pleased with that.’’

http://www.goldcoastbulletin.com.au...s/news-story/c7c2d569d2232a42655365152f0726d3

The game is being played in our half for most games. We just aren't putting goals on the board in big numbers.

The Power’s stunning first half summed up just how well it is playing the territory game and why it has risen from non-finalist in consecutive seasons to genuine flag threat this year. It is the blueprint for the club’s success. No side has controlled play like Port has in the first half of the season.After allowing Hawthorn just 26 inside 50s at Adelaide Oval last Friday while penetrating its own forward 50 64 times, the Power illustrated how its pressure game is paying big dividends.

Th Numbers explained.

The numbers are compelling. After round 11, Port ranked a clear number one in inside 50 and time in forward half differential — two key pointers to success. Its inside 50 differential of +15.7 is nearly seven more than the next-ranked side Collingwood (+8.9). Its time in forward differential of +9:25 is more than two minutes greater than the next-best team Richmond (+7:22). In other key territory game pointers, the Power ranks first in points from forward half turnovers (43.6), forward half stoppages (35.2) and points from forward half stoppages (20.5). Overall, it leads the competition in average inside 50s (62.4) and is second in points scored (108). Defensively, Port ranks number one in points conceded (72.7) and inside 50s conceded (46.7).

Team defence mantra

As good as the Power’s back half has been this season, a chief reason for its ability to stop the opposition scoring is that the game is being played in its half of the ground. Port has taken an all-in approach to defence and the players have bought in.

Even when it was outscored in the second half by the Hawks last week, Hinkley felt his side was playing the game on its terms. “We still had a lot of forward 50 stoppages, a lot of tackles,’’ he said. “We still has the ball in our half, where we would like it, but just didn’t have the polish (to finish off the work).’’ Once the Power gets the ball in its half of the ground it just doesn’t let it out. This enables it to get repeat scoring opportunities while allowing it to set up its pressing game which in turns protects its defence.

and the skipper backs up Ken

“We pride ourselves on our team defence,’’ said Power captain Travis Boak. “We did a lot of work on it during the off-season. The big thing was making sure teams don’t score against us on turnovers. “Right now we sit as the best defensive side in the competition but we still feel that we’ve got a lot of improvement in that area. “We need to make sure that we bring four quarters of defence – both offensively and defensively – if we’re going to beat the best teams.’’

The forward pressure leaders.

Key forward Charlie Dixon – a revelation in his second season at Port — has played a key role in the Power becoming forward half ball hogs. He and Geelong’s Tom Hawkins lead the AFL in forward half tackles among key forwards, with 33. Midfielder Brad Ebert – enjoying a career-best season – leads Port in forward half tackles with 39 ahead of fellow onballer Ollie Wines (33) and Dixon.

Dixon is the Power’s leading inside 50 tackler with 21. “We just love mowing opponents down with a tackle,’’ small forward Jarman Impey said. The proof is in the pudding!

 

[TeeKray]

Polec being high on that list of forward half tacklers is really encouraging. Great turnaround for a guy who was dropped last year for lack of a defensive game.

 

[Forzaport]

Polec being high on that list of forward half tacklers is really encouraging. Great turnaround for a guy who was dropped last year for lack of a defensive game.

He's a complete player now. Gun.

 

[El Zorro]

Nailbiters: How your team fares in the close ones

http://www.afl.com.au/news/2017-06-28/the-nailbiters-how-your-team-has-fared-in-the-close-ones

AFL.com.au has gone through each club's respective records in close contests. We've tallied up every match your team has played during the past five years with a final margin of 10 points or less, including draws.

http://www.afl.com.au/news/2017-06-28/the-nailbiters-how-your-team-has-fared-in-the-close-ones

Of course, I had to do the tally:

 

[RussellEbertHandball]

Nailbiters: How your team fares in the close ones

http://www.afl.com.au/news/2017-06-28/the-nailbiters-how-your-team-has-fared-in-the-close-ones

AFL.com.au has gone through each club's respective records in close contests. We've tallied up every match your team has played during the past five years with a final margin of 10 points or less, including draws.

http://www.afl.com.au/news/2017-06-28/the-nailbiters-how-your-team-has-fared-in-the-close-ones

Of course, I had to do the tally:

View attachment 386589

Prima facie confirms our senior players are mentally soft. The kids at the GWS have done better.

Or we are a long way behind and kick a few in junk time to get close. or we let opo to kick junk time goals to get close.

 

[PepeSilvia]

Are Brisbane #1 because they're so terrible that their wins are always small and their losses are usually big?

 

[OneGreatClub]

Prima facie confirms our senior players are mentally soft. The kids at the GWS have done better.

Or we are a long way behind and kick a few in junk time to get close. or we let opo to kick junk time goals to get close.

So what would be a better outcome this saturday night? Win a close one (to show our players can do it) or win by a lot (to enhance our percentage further?)

 

[El Zorro]

Are Brisbane #1 because they're so terrible that their wins are always small and their losses are usually big?

Probably.

 

[RussellEbertHandball]

So what would be a better outcome this saturday night? Win a close one (to show our players can do it) or win by a lot (to enhance our percentage further?)

Just bloody win - however we can.

 

[Chrizzt]

Prima facie confirms our senior players are mentally soft. The kids at the GWS have done better.

Or we are a long way behind and kick a few in junk time to get close. or we let opo to kick junk time goals to get close.

Our poor record in close games is an interesting situation.

The Arc did an analysis on close games last year: Close games and coin tosses

On average, if a team won 75% of their games in one year, we’d expect them to win 64.6% of their games in the following year. If a team won 25% of their games one year, we’d expect them to win 35% the following year. Good teams tend to stay good, but they regress a little. Bad teams stay bad, but they usually improve a little. So far, so thoroughly unsurprising. We don’t need to do any kind of fancy statistical analysis to tell us that.

What about close games? If teams tend to win a large percentage of their close games in one year, should we expect them to carry over this trait to the following year? You can imagine reasons why the tendency to win close games might persist year to year. Teams might have a winning culture, or experienced players who know what to do in the tense final minutes of a close match, or they might have superior endurance that lets them run out the game better than their opponents.

But it turns out that this trait doesn’t seem to carry over from year to year at all. A team’s ability to win close games in one year has no relationship to its ability to win close games the following year. Here’s a version of the plot above, but this time just looking at close games, which we’ll define as those decided by 12 points or fewer.

This time, we have an almost perfectly flat trend line, with a slope of .01. Statistically, this is indistinguishable from a slope of zero, a purely flat line. In other words, knowing how good a team was at winning close games in one year provides no guide at all to how good they’ll be at winning close games the following year. The ability to win close games is a trait that does not persist in clubs from year to year.

This also goes back to an earlier analysis done by Matter of Stats: Close Games in VFL/AFL History: Do Successful Teams Win Them?

On balance, I think it's fair to say that the evidence for the proposition that successful teams win more than their share of close games is weak at best. Statistically speaking, if there is an effect, it's a very small one.

It's not true either, by the way, that successful teams avoid playing in close games altogether, since the correlation between the proportion of close games played by a team in an era and the number of Premierships it collected in that era is just +0.07, and between the proportion of close games played by a team in an era and the number of Runner-Up finishes just +0.05.

Here's an alternative hypothesis: close games are largely lotteries and the rate at which generally successful and unsuccessful teams win them will be very similar. As such, they don't provide much information about a team's ability. What will differentiate teams is the rate at which they win games that aren't close.

...

Our alternative hypothesis seems to be supported by the evidence then. Successful teams don't so much tend to win the close ones, but instead tend to win the ones that aren't close.

He also suggests that the results of close games are just largely down to luck, and that the most successful teams don't have a statistically significant difference in their close game win percentage. However, when successful teams win, they tend to win by large margins. Backing up the old idea that percentage is often a better indicator than W/L record.

 

[RussellEbertHandball]

Our poor record in close games is an interesting situation.

The Arc did an analysis on close games last year: Close games and coin tosses


This also goes back to an earlier analysis done by Matter of Stats: Close Games in VFL/AFL History: Do Successful Teams Win Them?


He also suggests that the results of close games are just largely down to luck, and that the most successful teams don't have a statistically significant difference in their close game win percentage. However, when successful teams win, they tend to win by large margins. Backing up the old idea that percentage is often a better indicator than W/L record.

When we lost to Geelong by 2 pts statistician sirswampthing put up a tweet about Joel Selwood and close results by under a goal. Since then Geelong have won another game by less than a goal by 2pts on Sunday v Freo. Ok Selwood was knocked out at around 2 minute mark and didnt contribute. Sure luck plays its part but senior players who know what they are doing play a big part IMO.

 

[Chrizzt]

When we lost to Geelong by 2 pts statistician sirswampthing put up a tweet about Joel Selwood and close results by under a goal. Since then Geelong have won another game by less than a goal by 2pts on Sunday v Freo. Ok Selwood was knocked out at around 2 minute mark and didnt contribute. Sure luck plays its part but senior players who know what they are doing play a big part IMO.

Qualitatively speaking, I wouldn't disagree - hence why it's something that has interested me for the last few months.

When looking at close games in the last few years you can discount clubs like Brisbane and Essendon who have a number of close wins but they are mostly due to not being good enough to win big. You can however compare both Port and Richmond who have both been middle of the table sides capable of winning big on their day. Port are 10 wins 19 losses from games decided by 10 points or less in the last five years, while Richmond are 8 wins 14 losses. Compare that to Hawthorn (15-7) and Geelong (19-7) and the story does seem to emerge that Port and Richmond are actually doing something wrong, instead of just falling victim to chance.

 

[Lagooner]

Are Brisbane #1 because they're so terrible that their wins are always small and their losses are usually big?

On the flipside when we win we win big and when we lose we usually lose small.

I can only think of a handful of blowouts under Hinkley off the top of my head (Freo '13, Bulldogs '15, Crows '16, Essendon '17).

 

[Janus]

With only a handful of games to go, I thought it might be good to look at how we are traveling. Which is better than most, if not all, of the competition. Let's take a look at a few key aspects:

1. Contested possession - 1st

At the start of the year, both Adelaide and GWS flew out of the blocks as contested ball winning sides - their differential was something like 17 per game. But as I said, this method of play is unsustainable because it relies too heavily on athleticism to get to the next contest.

Meanwhile, over at Port, we were sticking to structures and our revolutionary formation based game style. As of this moment, Port Adelaide has a contested ball differential which is the best in the competition of +9.6, which is .6 ahead of Geelong (9.0) and 3.4 ahead of Adelaide (6.2). Sorry Tom Lynch, but you're shit, and you know it.

2. Clearances - 2nd

GWS is way out in front for clearance differential with 8.6 per game, and understandably so. Port Adelaide is second with +3.6 per game. The next best is Gold Coast with +3.3 per game. Adelaide and Geelong? -0.3 and 1.4 per game respectively.

3. Inside 50s - 1st

Port Adelaide generates so many repeat entries, denying opposition shots inside their forward 50, that it's almost unfair. +13.6 per game. To put that in perspective, the next best side is Adelaide with +6.1.

4. Points For/Against - 2nd

Only Adelaide with +33 points per game has a better strike rate than Port Adelaide, who in spite of losing more games than the Crows has a scoring rate of +28.

"I'm not going faster, everyone else is just slowing down." - Cole Trickle, Days of Thunder

 

[Janus]

Last year, I made the statement that we needed to find another 3 goals and we'd be in flag contention. I said we weren't that bad, in spite of what the analysts from Lehman Brothers on this board said.

In 2016, we scored 300 goals (13.6 per game) and conceded 275 goals (12.5) over 22 games for a total of +25 goals for the season, or +1.1 goals a game.

In 2017, we've scored 240 goals (15) and conceded 176 goals (11) over 16 games for a total of +64 goals for the season, or +4 goals a game.

Funny how that works, huh?

 

[PepeSilvia]

Wow, prophetic.

You predicted that kicking more goals and allowing fewer would make the team better?