This is the final piece of a series I started around Christmas. It wasn’t intended as a series, it just turned out that way. Idle speculation about the importance of goal difference in the relegation dogfight led to the first post, in which I wrote “I’m going to stick my neck out and say that the five (from SAFC, the Mags, Villa, Wigan, Southampton and Fulham) whose goal differences show the greatest improvement in the second half of the season will avoid relegation, irrespective of their points on Boxing Day, and a consistent decline will point to the doomed team”.
Proving such fighting (or foolish, take your pick) talk meant I had follow it through. Goal differences themselves were simple enough to display so I began with Sunderland’s as an introduction. Here it is for the season:
When I started our goal difference was heading downwards. This was bad news but not necessarily lethal as long as other teams were dropping more quickly. What was required was a chart containing the goal differences for all of the teams in the drop zone to show how we compared, so that came next. I’ve repeated this format with the chart below, to which I’ve added the final positions for each team. I couldn’t do that before as positions often changed between writing and posting:
However, although this chart gave an idea of what was going on it didn’t really meet my needs. Changes after Christmas began in the middle of the graph, when GDs were all over the place and difficult to compare. I had to reset each team’s goal difference to zero at mid-point and start from there. This produced my next chart:
As the season progressed this chart changed on a weekly basis. The lines went up and down as teams’ fortunes waxed and waned or waned and waxed, and while most of the goal differences did decline at some point none did so consistently. Villa came in for some hammerings, putting them in the relegation frame, while we had a couple of good results. Newcastle started poorly then their Christmas signings provided a revival, as did Southampton’s managerial change and QPR’s signings. We resumed our downward path as Fulham trod water, then Villa picked up, QPR faltered and the Mags slipped as Wigan began their customary late charge. My theory was in trouble well before April, when wide margins in some six-pointers (we won one 0-3 and lost another 6-1, you’ll remember) and also in other games (who lost 6-0?) put it beyond recovery.
Before then, though, I had tried to clarify things by using trendlines and R² values. This wasn’t the most mathematically sound procedure but it did produce a simpler chart and the gradients of the clubs’ trendlines became my key indicator. In February Villa’s line had by far the steepest gradient and I predicted they would go down with QPR, with Wigan joining them if Reading managed to maintain a revival.
In March more results supposedly gave greater reliability and I revised my prediction according to the new gradients. Wigan would join QPR and Reading, and the next three teams would be Villa, Southampton and SAFC in that order. Fulham (never seriously in danger) and Newcastle were safe.
In April, with only five or so games left, the trendlines pointed at Reading, with Southampton looking safe. Wigan remained my favourite for the third relegation spot unless they beat Villa, who might then replace them, but NUFC and ourselves were still not out of trouble.
Newcastle made themselves safe with a game still to go, by which time Southampton’s relegation would have required a miracle, but it wasn’t until Wigan went down at Arsenal that the final game for Villa and ourselves became irrelevant. And with those final games came the end of the season and the complete chart:
And this puts the final nail in the coffin of my theory. Villa and Southampton, who finished in 14th and 15th place, are the only teams in positive territory. Fulham, who finished above them, have an iffy gradient, then there’s us, slightly better than Wigan, who are slightly better than QPR, who are slightly better than Newcastle, with only Reading out of it. There’s no single line pointing to a third doomed team. As I always had Wigan in the frame I could say my predictions were not too far from what happened but it would be like saying “… I never meant to punch that horse, officer”.
Building the charts and generating dodgy lines provided an interesting diversion but was ultimately as useful as a nosebag in the Strawberry; looking at points and places would have served me just as well. To see what I mean have a look at the table below. It gives points and positions at the half-way stage and at the end of the season. Who would you have put your money on at Christmas?
|19 games||points||38 games||points|
As my method is busted I won’t be drawing graphs and plotting trendlines next season. This doesn’t mean I won ’t be making predictions. As Baldrick might say, I have a cunning plan. Join me next Christmas to find out what it is.
See all Salut! Sunderland’s articles recalling May 5 1973 and the run that took SAFC to FA Cup glory: https://safc.blog/category/fa-cup/may-5-1973/
Enter the great Salut! Sunderland Two Million hits competition. Read about the first prize – £200 towards a holiday from Sunderland-based Hays Travel at https://safc.blog/2013/06/attention-safc-newcastle-middlesbrough-even-scarborough-supporters-2m-hits-competition-prize-announced/ BUT POST YOUR ENTRY IN THE COMMENTS SECTION ONLY AT https://safc.blog/2013/06/salut-sunderlands-two-million-hit-party-even-newcastle-are-invited/