autocol

Not at all. I love disagreement because it's an opportunity to learn. What I don't like is people who barely skim read carefully constructed arguments, read whatever they want to read into them, and then regurgitate their viewpoint without regard for what others have added to the conversation. It reminds me of the Monty Python sketch, and it's pointless.

Hate to be critical, but that's your subjective opinion. You have no concrete evidence to support your hypothesis.

Quite clearly you've not read anything I've written beyond a quick skim, you have no interest in discussion beyond pushing your own opinion, have no idea whatsoever about what I'm even talking about, or have gleaned any insight as to how analytical thinking can improve one's perception of a given situation. In short, I'm bored of arguing with you. Good day sir.

Since you guys are stuggling with the difference between objective and subjective observation, allow me to offer my subjective opinion. Dunn is not a creative footballer. "See the ball, hit the ball out of bounds" seems to be his default thought process. When he does get the ball, his ability to scan for options quickly is not fantastic. He's a good, accurate kick, but the guy who wants it needs to make all manner of large motions with flourescent semaphore flags if they expect him to notice them, if not, he'll revert to the most basic of defensive kicks - the bomb down the boundary line. That said, when the ball is delivered to a contest, Dunn is more likely than not to spoil the ball, and achieve one of his two basic aims - hit it out of bounds or bomb it back down the boundary line. That's his talent. Simple, negating football. These are my subjective observations, just as most (well, all, actually) of the opinions presented in this thread by the rest of you are. I could, if I had the time or inclination, attempt to turn my observations into something more closely resembling objective facts by engaging in research. This might involve watching the game tapes, categorising types of contests, types of results, and scoring Dunn and his opponent in various ways. I could even score the frequency with which Dunn forgets to have an opponent which a few of you claim to be a common occurrence. All these facts are knowable, you simply need to perform the research to obtain them. Last year on Bigfooty I posed a hypothesis about the likelihood of a score based on an inside 50 being from a low "stab" pass or a high "bomb" into attack, and questioned by one of the more astute posters, I performed an analysis just as I'm describing on two quarters of football picked at random by him. The results only somewhat supported my hypothesis. As a result, I was forced to concede that the objective facts did not match my subjective opinion and thus my opinion was altered. You should go and read it. I'm proud of that post (and it took me two hours to create it so do me the kindness of at least making it worthwhile). Those of you making statements like "Ahhhh, but it's how you choose to present and interpret those very statistics..." as though there's been some grand conspiracy to fudge the numbers in this thread to make Dunn look good would do well to note that I have not presented OR interpreted any statistics in this thread. I've provided links to the stats, and invited people to peruse them. This notion that "statistics are all lies" is a falsehood consistently propagated by those that struggle to deal with facts. The facts already existed, and were already true, I just happened to point them out for you. Also, those of you that don't think that the offices of every AFL club are absolutely brimming with geeks like me (but smarter and better trained) running statistical analyses on all manner of data that we don't have access to, with direct effects on tactics, recruitment and list management, are sorely mistaken.

It's true that data can be manipulated. As the saying goes, there's lies, damned lies, and statistics. However, it's become a tediously easy way for people to deny whatever facts are being shown to them to simply cry that "the data doesn't support my intuitive claim, therefore the data has been manipulated!" The easiest form of this is cherry picking, that is, only presenting data which supports your claim. You could accuse me of doing that, except the truth is that I haven't. I didn't pick Josh Gibson, someone else named him first, and as a regularly lauded player in a top team I expected an interesting comparison. I expected Dunny's stats to be far behind his, but reasonable enough to justify a claim that he might be able to perform a similar role in a similar team. I was surprised to find him pretty much on level pegging. Howled down by the "Gibson has to take a better opponent" brigade, I simply expanded the comparison to all of the players around Gibson. Note that I didn't choose Hawthorn or Gibson after checking that the stats would support my claim. I simply linked to the stats and read out the results. Everybody has access to the same statistics that I do, so anyone could easily search the 16 other teams to search for one which statistically proves that Dunn is a poor player. My hypothesis is that you won't find one, but I won't claim that to be fact until someone does the research. If my statistics had been manipulated, as you're suggesting they could have been, then you could easily disprove my claims by accessing the publicly available data and making an analysis of your own. You're likely to reach a similar conclusion to the one I've detailed here, because there's been no manipulation of figures. I really wish I could teach people that it's okay to have a hypothesis or prediction (GO YOU GOOD THING SPENCER!) that turns out to be wrong. It's okay to refer to the evidence and realise that your initial intuitive reaction needn't be the same as your reasoned critical analysis. Subjectively, I think the main reason people think Dunn sucks as a footballer is his persona, and the fact that when he gives away a free kick, the opposition typically score a goal. The rest of it is an irrational attempt to justify their opinion, because the evidence is that Dunn is not a sucky footballer. It's interesting (and/or frightening) to note that the majority of the votes cast on September the 7th will be made with equal disregard for evidence and facts. Thanks. I just wish people would actually read them before making claims about my lack of credibility.

Seriously mate. Read what i actually write.

If you're going to accuse me of arguing poorly, I'd appreciate you actually reading my posts. Err, no he doesn't. One of the points I made above is that Dunn doesn't take marks, and the statistics bear that out. Even when you're trying to give him credit, you're still wrong! I know you think you're an expert observer of football, free from the cognitive biases and afflict everyone with a human brain, but you're not. Well, I subsequently compared him to all of Hawthorn's defenders, and surely one of them is his 'equivalent' if Gibson is not, and Dunn compares very favourably with all of them with the exception of Grant Birchall, who is unarguably a tier above him as a player (and also, arguably, given less accountability as a defender than Dunn, which might help explain his superior disposal count). You might think I focus on nothing but numbers, but I think you focus on nothing but your own (highly suspect) view of reality. The main difference between Lynden Dunn and all the members of Hawthorn's back six is that their team usually wins, and his team usually loses, and you want someone to blame in the latter case.

No, a lot of observations have nothing to do with research, but I think you'll find that Alexander Fleming had done quite a lot of research to test his hypothesis before making any declarative statements to the world about the anti-bacterial qualities of the substance he named penicillin.

Okay, well let's assume then that Hawthorn and Melbourne have played exactly the same opposition as each other throughout the year (the fixture being uneven means that's almost certainly not true, but let's assume it is for the sake of the argument). On that basis, no matter who Dunn plays on or how far down the pecking order he is, someone in Hawthorn's back six has the same role as him, and plays against the same guys as him. Do you want to know how Dunn stacks up against everyone in Hawthorn's back six? Very well. He takes less marks than most of them, and his disposal efficiency is marginally below most of them, but I'd attribute both of those facts largely to the constant pressure and lack of "free possessions" that comes with playing in the most consistently poor team in the league. Subjectively, I'd say he tries to spoil more often than mark too, unlike Brian Lake for example, but I don't have any proof of that. Other than Grant Birchall - who accumulates seven more possessions per game - statistically Dunn compares very nicely to every single member of the Hawks back six. The talls, the smalls, the negaters and the attackers. His defensive capabilities in particular appear to be better than most of the Hawks back six (though again, that could be because the other team has the ball more often than we do). So here I remain, a card carrying member of the Lynden-Dunn-goes-half-alright club. Ask me what I think about Jake Spencer, future superstar.

Haha. Good argument. I notice you've instead brought a well researched reply which demonstrates Dunn's ineptitude with plenty of supporting evidence. Oh. No you didn't. Right. And where do we put Dunn? In the goal square. Of course if he gives away a free it's going to be a goal, his job is to stand in the frickin' goal square! If ever there was a more obvious case of the emotional cost of a result bearing on the perception of the test than this, I can't think of one. The free kicks Dunn gives away - 38% less often than the average player - often result in a goal because his job is to defend right in front of the goal. That doesn't mean his transgression was any stupider or less skilled than one by Luke Tapscott on the half forward flank, you're just more likely to remember it. You're exhibiting a classic cognitive bias caused by the availability heuristic. I don't have any evidence to support this claim, but I bet you that's exactly what they do. Especially guys in recruitment (who no doubt utilise many intuitive judgement call also).

You guys are welcome to construct a coherent and logical counter to my argument, but other than the claim that Gibson takes a better forward than Dunn (which I imagine might be true, not that anyone has done the research to check if it actually is or not, and I'm not getting the feeling any of you are going to), I haven't seen anything yet. You might scoff at the idea that someone would do any research to test their intuitions (and admit that their intuition is wrong on occasion, as they often are), but that's the manner in which all great scientific discoveries have taken place in the last two centuries. You'll note also that every single team pays Champion Data and a fleet of analysts huge wads of cash to measure and analyze every tiny facet of the game... You don't suppose that might be because... there's some merit in it!?

Dunn averages 0.5 free kicks against per game. League average is 0.8 (ish). This while part of a constantly beleaguered defence. Your subjective observations aren't statements of fact, I'm afraid.

Well, at least you seem to be talking some sense! I'd argue that it doesn't matter for this discussion that Dunn is taking the 'lesser' defenders. We've got Frawley, Garland, and TMac, that's going to be his role in the team in future regardless (along with the occasional swing into a forward or tagging role when the situation demands it). We're not measuring his worth as a gorilla defender, we're measuring his worth as a "utility" defender, because that's what he is. His main shortcoming, compared to Gibson (and statistically he has very few) is disposal efficiency and clangers, and my subjective view of that subject is that any defender would make more disposal errors in the Melbourne backline than playing for any other team, because there's simply no-one to kick to when you get the ball.

But while I'm "waffling", let me continue... Stat comparison Gibson just leading on disposals, marks, frees for, uncontested possessions and effective disposal %, but not by much. Dunn leading on goals, frees against, clearances, rebound 50's and bounces, but not by much. Gibson has a clear lead in one-percenters and clangers, Dunn has a clear lead in tackles and inside 50's. Statistically, they're basically the same footballer. Gibson in a better team (more uncontested possessions directed to open teammates, Dunn more tackles and disposal under pressure). Your judgement on the issue is clouded entirely by your emotional involvement in watching the game, and the outcome of the game itself. I posit that if the two players were swapped in an alternate universe, you would be calling for Gibson's delisting while simultaneously joining the chorus of praise for Dunn. How's that for an opinion to follow fact?

Waffle? Statements of fact are waffle? The last sentence was my opinion. The rest of it is facts. You're letting your opinion come before the facts, which is exactly what Daniel Kahneman said you would do (but you didn't read even a single paragraph synopsis of that either, I assume).

A pointless stat? The statistic that says that if two players are in a one-on-one contest, no player in the LEAGUE is more likely to win the contest than Lynden Dunn, you're calling that a pointless statistic in a thread about LYNDEN DUNN? Are you for real? Granted, Jordie Mackenzie is not going to beat Dunn in such a statistic because he's the poor sod that has to play on Gary Ablett or Jobe Watson every week which makes such things a bit unfair (though I do recall a certain L. Dunn completely nullifying Chris Judd some weeks back), but all the same, Dunn wins more one on one contests than any other player in the league. Fact. The statistics haven't fooled me at all, I just don't fall for the fallacious method of using analytical thinking to justify intuitive ideas (google Daniel Kahneman's work if you're interested in the psychology of analytical thinking). You're all saying that Dunn should be delisted, because he's 'below AFL standard'. We're the worst team in the league, last time I checked. That means that we probably have the greatest number of players on the field who are 'below AFL standard'. Every other team is better than us (even GWS, these days), therefore any other random player on any other team is more likely than any random player on our team to be 'of AFL standard'. Not only that, any random player on our team is likely to be surrounded by more players 'below AFL standard' than the random player of the opposing team. Thus, statistically speaking, Dunn is more likely to be up against a player of AFL standard who is surrounded by others of AFL standard, and more likely to be surrounded himself by players below AFL standard, and still leads the league in winning one-on-one contests, and I'm the one who doesn't understand statistics? Put Lynden Dunn into the Hawthorn back six and you'd all be lauding him as Josh Gibson reborn, but with a better kick.

It's my opinion that Spencer will grow into an excellent first ruck in 3 to 4 years. It's also my opinion that she shows a lot more endeavour than 95% of our list, and calling his effort into question for one shepherd is a bit unfair. He gives 100%, and runs harder than most of our midfielders who are half his height and a quarter his weight.

You realise that Dunn is the single most effective one-on-one contest winner in the league? (data accurate at round 18). Like most of our players, you put him in a better team and he'd look a lot better. It's hard to kick precisely to a free player when no-one is running so there are no free players to kick to. The stats say Dunn is a good footballer, and I believe them.

You blokes calling for Spencer's head will be shoveling huge loads of humble pie into your gobs in the next three years.

I'm neither surprised nor bothered. The guy has been a pea-heart since day dot. He's never going to be a gun in a Melbourne jumper, and probably not in any other colour either. We get the best trade we can for him and move on.

Which is one of the reasons why the likelihood of a CEO being a sociopath is about four times that of the general population.

... rejoice in watching last year's footage of Colin Kaepernick running beastly read-option plays, and looking forward to the NFL season starting.

He can't have done that, because the wording of the poll is not heavily loaded to skew the results in Rangey's favour.

Remind me... when was Mitch's contract signed?

When they signed the contract he's on, he was quite capable of taking the park. Which bit of that don't you understand?