FA Cup Blessing In Disguise?
Huddersfield Town will play 51 games this season. It's the first time in a few we've known before Christmas how many that will be, and is 10 fewer than last season, and 4 fewer than the season before - obviously, it may increase to 53 if they end up in the playoffs, but we have some interesting numbers to work with here.
I was disappointed to see Town go out of the cup on Saturday, but more disappointed with the way it came about. My mind, instantly, wondered whether or not we had any reason to think that 'concentrating on the league' would be a viable option, and end up successfully. I've done a post previous to this about whether or not there's any trends in the JPT winners, and that concluded that it was a pretty random process there, but did not affect league form one way or the other.
Luckily, I had the data almost to hand and have worked out what history teaches us. I've taken the end of season top three at the end of the last 24 seasons (the years inclusive of playoffs) which was an arbitrary figure - I did consider starting at the first year of the Full Members' Cup, but settled on the playoff years being more arbitrary and compared league records at the end of the season to league records on FA Cup First Round day.
So what do we see in the numbers? On the surface of the dataset, I can only congratulate Mrs Marco's Gillingham team of 1999/2000 for having the longest FA Cup run of that group (8 games, culminating in a Quarter Final thrashing at Chelsea) and the Stockport County team of 1996/97 for playing a frankly astonishing 21 games extra to their 46 league games - they were semi-finalists in the League Cup.
They're just points of interest, though. There's some more nitty gritty to go with. Gillingham, at the top of the figures, saw their league form increase from 1.571 per game to 1.848 per game (an 18% increase) so their cup run clearly didn't hamper them. The 10 teams who have been eliminated in Round 1 have seen their average ppg fall from 1.969 to 1.850 - that's down 5%, but includes within that 10, a rise of 12% and a fall of 21% - so certainly there's form for teams struggling after an FA Cup First Round defeat, but no guarantees - indeed Leeds in 2009/10 (you'll remember they won at Old Trafford) saw their early season form dip 22% - from a frankly unsustainable 2.400 ppg to a far more reasonable 1.870.
This trendline (black) indicates that average points per game ratio rising more the further teams get in the FA Cup; the red line is the specific ratio rise and a fall (in ppg) so if Town end up following the average level, they'll be looking at their points per game dropping around 0.05 of a point per game - which might not sound a lot, but would be the difference between 95 and 92 points from current form.
Those same ratios would indicate that any team who gets past the first round, even only as far as the second, averages either the same or better than their PPG up to that point, so Sheffield Wednesday and Charlton would be looking at - from what we've seen before - improving their tallies from 108 (Charlton, and 'Wow') and 89 (Sheffield Weds) - Wednesday would need to increase their ppg by 4.2% to get up to the 92 that is projected for Town. I hope that makes sense; basically, history points to the fact that Town's form is unlikely to improve, but Charlton and Wednesday's is - on average.
The FA Cup isn't the only competition tiring hearts and minds of League One's finest and Paul Thirlwell throughout the Spring, though. As Thirlwell himself will know, everyone in England's third tier is afforded four opportunities to get to Wembley - the FA Cup (as before), the League Cup (held by a team only a division above), the playoffs (last season's trip to Old Trafford notwithstanding) and the JPT (held by Carlisle). A team winning all four would play a maximum of 32 extra games (that would be a mammoth season - 78 games - the closest we've come to that is - and you probably wouldn't guess this - Grimsby Town in 1996/97) and open themselves to more the season after should the enter the Europa League.
That chart, though, looks very, very similar to the FA Cup one. A wild and wavy line of specifics overlooked by a gradual dip on the trendline - good Tier 3 teams, then, don't seem to tire as the season runs on; something which is confirmed by the average ppg changes by position; those who win League One average a 0.005% dip between now and the end of the season; (that keeps their average points at 92, so small a dip is it); the second and third place teams both increase their points per game tallies, by 0.052% and 0.073% respectively - (from 84 to 86, and from 79 to 83). Recent recollections would suggest there's an element of that that is down to title winners taking their foot off the gas - Brighton's defeat against Town last year, and Norwich losing at home on the last day of the season before spring to mind.
Omens, then, are mixed. I did look to see if there was any correlation between playoff games and total games, but there is nothing that can be worked into a pattern that would indicate anything being more like to happen than not.
10 teams, as I mentioned, have been knocked out of the First Round of the FA Cup and gone on to finish in the Top 3 of Tier 3 since 1987/88. 3 were Champions, 3 finished second and four finished third. In the interest of balance, if nothing else, its only fair if it goes 3/4/4 or 4/3/4. Only time will tell, of course. (Any Addicks or Owls reading this - aim to get knocked out after either 2 or 7 games - there's a 5/4/2 split on 2 FA Cup games and an even better 3/2/0 split)
Older/Newer
« EPPP In Numbers Addendum | What A Difference A Day (Or A Year) Makes »






Good article ruined by a lack of axis titles on the grpahs - I can't work out what they're telling me.
Thanks. I'll label up the graphs properly tomorrow - they're percentage points per game differentials of teams ranked firstly by number of FA Cup games, then by total games played in the season.
Hopefully that makes a little more sense; next time I post I'll keep the graphs with axes as I get them in Excel - I'd changed a lot of the data (trying to work out a few other things that came to nought) so I had to add them to the images of the graphs I had saved.