McGrath v Warne

As Gideon Haigh once put it 'Warne and McGrath. McGrath and Warne. Who is to decide their proper ordering? Should it be the man with the more wickets or the lower average?' Well, as I've argued here before, the lower average matters but the other relevant statistic for a bowler is not their wickets but their strike rate. A team's chance of winning a cricket match is a function of how quickly they score and how long they stay in, so how could a bowler who takes wickets more quickly and more cheaply than another bowler not be considered more valuable? And on both these measures, McGrath comes out ahead:

McGrath - Average: 21.64; Strike rate: 51.95
Warne - Average: 25.42; Strike rate: 57.49

But hang on, McGrath is a fast bowler and Warne is a spinner. How do they each compare to bowlers of similar characteristics? Well, if you take the average bowling average and strike rate of each of the top 10 spin bowlers and top 10 non-spin bowlers by wickets taken, you get these stats:

Non-spin bowlers - Average: 24.39; Strike rate: 54.88
Spin bowlers - Average: 28.98; Strike rate: 68.89

And if you take each of McGrath's and Warne's bowling averages and strike rates and transform them into a percentage of the corresponding averages just above, you get these figures:

McGrath - Average: -11.2; Strike rate: -5.3
Warne - Average: -12.3; Strike rate: -20.1

So Warne's average is slightly better compared to the top spin bowlers relative to McGrath's average compared to the top non-spin bowlers, but his strike rate is much, much better compared to the top spin bowlers relative to McGrath's strike rate compared to the top non-spin bowlers. Does this mean we're much more likely to see another McGrath than another Warne? It would seem so. Part of the reason why McGrath's figures, and those of non-spin bowlers in general, are better than those of Warne and other spin bowlers could be that non-spin bowlers face easier bowling conditions on average. Think about it: if bowling conditions are good then fast bowlers clean up before the spin bowlers get much of a chance but if bowling conditions are tough then both fast and spin bowlers have to rough it out. I haven't proven this theory yet, so let's just stick with the view that a spin bowler as good as Warne is rarer than a fast bowler as good as McGrath for now. Which, given the trouble Australia has had on settling on a new spinner since Warne's retirement, isn't all that surprising a finding..

The Finger Points Outwards - No. 37

At what point is it worth the NBA players caving in on the lockout?

The best statistics question ever.

A less than favourable review of the new Tintin movie.

The truth behind the Scrabble "cheating" scandal.

Who Has The Easiest AFL Draw in 2012?

With the release of the Australian Football League draw for 2012, analysts were quick to rate the difficulty of each team's fixture. Some went so far to come up with a numerical rating of this difficulty, generally based on who each team plays twice, road trips, and matches played against teams from interstate (see example here).

Using the AFL post-final power rankings posted on this blog a few weeks back, I have come up with the following ratings of the difficulty of each team's draw in 2012. Each team's rating is a sum over all matches of the reverse of the ranking points of their opponents (which themselves are based on those opponents' expected winning margins), adjusted for any home ground advantage. These adjustments for home ground advantage were explained here.

For now, I've assumed that Greater Western Sydney's ranking points are -50, and have treated their adjustments for travel in the same way as other non-Victorian teams (that is, +12 points for playing at home, -12 points for playing away) rather than Sydney (who have a smaller adjustment when playing Melbourne teams). I might change this before the season starts. Note also that I treat Hawthorn games in Tasmania as if they were playing in Melbourne (i.e. an advantage against non-Victorian teams, but not Victorian teams).

As an example, if Team X is playing West Coast in Perth then I deducted them -33.3 points: -21.3 due to West Coast having a ranking of +21.3, and -12 due to playing in Perth.

One main difference between my calculations and those of other analysts is that the latter do not account for the fact that higher-ranked teams do not have to play themselves. For example, Geelong gets a big boost in the ease of their fixture just by virtue of the fact that they don't have to play Geelong and other teams do.

Here are the rankings for the difficulty of each team's draw in 2012 - a higher ranking means that your draw is easier. Most team's rankings for difficulty of draw are positive since ranking points for AFL teams are skewed to the negative side, particularly with the introduction of Greater Western Sydney.

Adelaide 169.3
North Melbourne 164.4
Gold Coast 152.9
Fremantle 149.2
Melbourne 144.9
Richmond 115.1
St. Kilda 114.2
Geelong 100.7
Hawthorn 93.7
Brisbane 93.1
Greater Western Sydney 92.7
West Coast 86.6
Carlton 79.7
Port Adelaide 65.5
Essendon 46.9
Sydney 46.8
Western Bulldogs 4.5
Collingwood -9.0

So according to this system Adelaide has the easiest draw, and Collingwood has the hardest, which has pretty much been the consensus among commentators. However, Greater Western Sydney's draw is not as easy as some people think, partly because they don't get to play themselves, and partly becuase they don't have many home games at their own stadium. On the other hand, Geelong and Hawthorn have easier draws than most people think, partly because they don't have to play themselves, and in Hawthorn's case because they only travel three times to "hostile" territory.

But does it make a lot of difference? Essentially, Adelaide's draw advantages them by 178 points compared to Collingwood, so Collingwood needs to be a 178-point better team than the Crows to be expected to perform as well as them. Over the course of a season, that's about eight points per week. It's not negligible, but really if your team finds itself in 14th spot you shouldn't be blaming the fixture.

Monster Is REM's Acthung Baby

Why does REM's 'Monster' get so little love? OK, it gets some here, but it's generally seen as the tipping point for REM's popularity. Certainly it's a departure from their previous records, but that shouldn't be seen as a bad thing. I bought it on CD recently and listening to it again after the space of over 15 years I was struck by how much it reminded me of U2's 'Achtung Baby', widely considered one of the Irish band's most accomplished and inventive records. Both have a sense of irony and dirtiness after the clean, earnest soft rock of the albums that took them into the big leagues ('Automatic For The People' and 'The Joshua Tree') yet both still contain some of their band's most moving tracks ('Strange Currencies' and 'Let Me In' on 'Monster' and of course 'One' on 'Achtung Baby'). And the distorted glam-rock guitars on 'What's The Frequency Kenneth?' and 'Crush With Eyeliner' can match it with those on U2's 'Even Better Than The Real Thing' and 'The Fly'. Yet 'Achtung Baby' sold bucketloads and 'Monster' ended in the bargain bins. Maybe 'Monster' was just a little too murky for to garner mass appeal, or maybe it just didn't have the hits that U2's album had. Nevertheless it deserves to be remembered in a better light than REM's retreat from the mainstream.

Book Review: Supergods

Grant Morrison's 'Supergods' is uneven, a little bloated, but still weirdly compelling. A central thesis to the book is lacking, if not non-existent. Morrison's tome is part-history of the development of the American superhero, part memoir (although being one of the superhero genre's greatest writers, the two parts are almost necessarily intertwined), part-philosophical discussion. Contradictions abound - Morrison seems to both revere and mock the movement to make superheroes more realistic within the space of a few paragraphs, as well as his own efforts to bring high art to the long underwear characters. But then why should a unified view be imposed on multiplicity - shouldn't the author, like everyone else, have the freedom to change his mind? And really what does it matter when the best of Morrison's observations are so good, perfectly describing characters and tropes for which you thought everything had been said, such as his noting of the Sub-Mariner's 'fuck you snarl' and how the Silver Surfer was the first emo superhero. Morrison's choice of titles to focus upon mixes the accepted classics with his personal favourites (many of which seem like they might be better enjoyed on an acid trip), but in all cases his enthusiasm for the source materials makes the reader want to seek them out. Like much of Morrison's work, 'Supergods' could have benefitted from a little more discipline in the writing, but its best parts are pretty much as good as anything that has been said about this seventy odd year old genre.

Pop Will Eat Itself

I read a review today of a book called 'Retromania' which, according to the reviewer, essentially argued that pop music in the 2000s had lost its forward momentum and more than any other decade was obsessed with the past.

A few observations about this:

1) Every musical era is backward-looking to some degree. (I think the author would agree with this.)
2) Having said that, I think a case can be made that the 2000s have been more backward-looking than previous decades. It is possible this may in part be associated with the massive increase in access to pop's history through digital libraries.
3) Doing it first is not necessarily doing it better.

Is pop's time of innovation over and we are now doomed to an endless recycling of past tunes? Maybe, but you never know where the next wave of innovation will come from. A book I once read suggested that the next mega innovative pop star will come from outside the US and probably from one of the less developed nations. That was over ten years ago and we're still waiting, but hey don't count it out yet.

Why On Earth Does France Play Rugby?

Apparently it was introduced there by British residents in the 1870s - surely one of the crowning achievements of British 'colonization'. And apparently it's the most popular sport in the southern part of France.

AFL Power Rankings - Post-Finals 2011

Here are the final rankings for 2011. Collingwood were the No.1 team for much of the year, but that changed when Geelong beat them easily in Round 22, and the gap widened even further over September.

1 Geelong 51.2
2 Collingwood 35.9
3 Hawthorn 28.4
4 Carlton 27.0
5 West Coast 21.3
6 St. Kilda 16.3
7 Sydney 10.3
8 North Melbourne 1.3
9 Western Bulldogs -4.6
10 Essendon -11.1
11 Richmond -16.2
12 Brisbane -17.8
13 Fremantle -23.3
14 Adelaide -23.7
15 Melbourne -29.9
16 Port Adelaide -46.2
17 Gold Coast -46.9