So let me introduce you to Katwala's first law of political recovery:
No party which loses a General Election should elect its next leader within the first six months following the defeat.
Hence my answer when asked, among many others, by the New Statesman to identify the Labour party's next leader. It is much to soon to tell, hence my mildly counter-intuitive scenario:
More important than who should lead is how Labour could get its next leadership election right. Let's hope it takes place in government, a few years from now. But if Labour loses this spring, the first decision could be its most important in opposition: whether to have an immediate leadership contest or leave it until after the first autumn conference. If Michael Howard had not done the latter in 2005, the Tory leader would be David Davis, not David Cameron.
If Labour goes straight for a contest, we would miss the debate we need. So let's see the merits in playing it long. If Labour loses, Gordon must stay! If he prefers to make a quick exit, the National Executive Committee should appoint a caretaker and schedule an autumn contest. It could make a real difference to the chances of being out for one term, or three. Front-runners - such as David Miliband, Ed Balls or Harriet Harman - should not fear a longer debate with more ideas. Those who might not run - perhaps James Purnell and Jon Cruddas - might sharpen debates.
Without hearing their arguments, it is too early to set the field. So why not imagine that Ed Miliband and Yvette Cooper could surprise us by emerging in front, with Cooper edging home as the party's first permanent woman leader?
Let's have a closer look at the general case for parties taking that time to get questions of leadership right.
The Michael Howard precedent of 2005 is well known. The Tories took six weeks to elect William Hague after being crushed in 1997; and began balloting MPs within a month in 2001, with IDS crowned by September.
Dig back a bit further: it is striking just how well the law holds up. I admit there is also a fair dollop of coincidence in the fact that none of the serial election winners of post-war election history - Wilson (1963), Thatcher (1975) and Blair (1994), with ten victories out of eleven between them - first became leader of the opposition in a contest beginning straight after an election defeat.
In fact, relatively few party leaders did so: I think just George Lansbury (1931), Hugh Gaitskell (1955), Neil Kinnock (1983), John Smith (1992), William Hague (1997) and Iain Duncan Smith (2001). Note that none of them ever made the premiership though.
It isn't a magic formula. More time to talk and think doesn't guarantee it will be used well. James Callaghan (reluctantly) agreed to delay his departure until 1980. But the party factions were almost entirely fixated on a clash over the rule-book, with Old Labour's arch-fixer proving singularly unsuccessful in managing the outcome, before Michael Foot was elected under the old rules, though very much in anticipation of the new. Similarly, more time to talk wouldn't have seen the Conservatives bounce straight back after 1997. But they could well have begun a necessary conversation in the party that they didn't really begin for another eight years.
And that's much less likely if the very first question after an election isn't about what happened, or what it means for a party's future direction, but rather narrows down to 'who's nominating who'.
Is even mentioning this now a potential distraction? Well, this shouldn't be shouted from the rooftops. But there may be no time at all to do so if the worse happened.
In general, if parties generally followed this approach there would be pre-election benefits too.
Right now, the political media (still some months from the election) have finally been robbed of two of their hardy perennials for endless speculation: the potential for an anti-Brown coup, and the timing of the election. (Of course, we told you so; on both fronts). The next leader will be the next thing.
Columnists will no doubt be very happy to host a coded, off-the-record and mostly private debate, which might deliver an untested consensus on who should emerge.
Party members want leading figures to concentrate on the General Election now. So overt proxy leadership manouvering should backfire with those whose votes will matter. That pressure would be stronger too if parties always knew that there would be plenty of time to have a proper debate, out in the open, at the right time.
Hypotheticals; hypotheticals. It may well not be a law we need to recall for many years to come. Still, let's make sure we remember it if and when we ever do need to.