30 November 2009

Introduction

Hello. After several years of doing sporadic work on cricket stats – occasionally posting bits and pieces here and there on various messageboards – I've decided it's time I started to collect my output in a slightly more structured way. This blog is an attempt to do that.

I think the posts are likely to fall into three broad categories:

The first type will use conventional cricket stats to answer simple questions (please feel free to ask me simple questions). These are the kinds of query you might normally put to your favourite online stats engine (e.g. Cricinfo's statsguru). Obviously, if one of those resources can give you your answer, then there's no point asking me, but there are questions that might be a bit more stubborn – perhaps because they don't quite fit into the parameters provided by those online engines, or perhaps because they relate to another kind of cricket (above all, I'm not aware of any online source that enables you to query FC records). I'm probably unlikely to come up with many of these posts on my own initiative, so how often they appear will depend on how frequently I get asked these kinds of questions. The kind of thing I have in mind is very much like the BBC used to do with Ask Bearders or Cricinfo still do with Ask Steven. Heck, let's call it Ask Gabe.

The second type of post will be one that breaks out some slightly more advanced stats to try and get beyond what normal cricket statistics can tell you. With very few exceptions (some are listed in the sidebar), cricket analysts have been dreadful at mobilising anything like the full range of tools at their disposal. Statisticians tend to make a distinction between descriptive statistics (those that simply present empirical data) and inferential statistics (those that seek to make sense of it). Cricket anaylsis is dominated by an awful lot of the former, and tends not to feature much of the latter. It is especially embarrassing to compare attempts at quantitative analysis of our game with those that innovative baseball statisticians have been producing for decades. I'm interested in applying a similar approach to cricket. I'm going to tag such posts as Going Deep (or, at least, I am unless I can think of something a bit better to call them).

The third type of post will take the second approach to its logical conclusion, to attempt to specify and answer some Big Questions. The kind of thing I have in mind is the sort of analysis that David Barry's really good at – an attempt to characterise and make transparent some of cricket's innermost dynamics. Tell the honest, I'm not absolutely certain I'll ever manage to get any of these together, 'cause they require... y'know... thinking really hard, and that. I do have a couple of questions at the back of my mind that may make it to the surface one of these days.

Some warnings:

  • The resources I've built for myself so far relate only to test, ODI, and first-class cricket (from 1850). I haven't, as yet, built anything to look at domestic limited-overs ("List A") or Twenty20 cricket, though it's on my mind to do so when I get a quiet moment (perhaps over Christmas).

  • Similarly, I don't haven't have access to any ball-by-ball stats. Much as we'd all love to get our hands on that kind of data, it's pretty hard to come by (and, of course, anything you might be able to derive only relates to the last decade or so).

  • Australia v. ICC World XI, 14–17 October 2005, wasn't a test match, and anyone who says otherwise is a liar. Similarly, these games ain't proper ODIs.

  • I'd love to promise frequent updates, but I'm known to be unreliable in such regards, so stuff will go up as and when I can find the time. Unless anyone wants to pay me a salary to do this shit.

No comments:

Post a Comment