FA Cup Third Round: Predicting the Upsets

The FA Cup third round is often cited as one of the best weekends (as it has now become) of football in the season.

With the biggest teams in the country entering the knockout competition for the first time, it means the first possibility of any real big upsets.

I’m going to try my best to highlight some of those games that, statistically speaking at least, look like they have a decent chance at producing upsets.


The first third round game this weekend is (my beloved) Championship side Cardiff City hosting League One Colchester United.

On the face of it this would seem a fairly unlikely one for an upset – Cardiff are comfortably set in mid-table (for the moment), while Colchester are struggling to avoid relegation.


There is of course the vast mess that Cardiff finds itself in off the field.
But comparing shot stats there is every chance that tonight could be an embarrassment on and off the field for the Bluebirds.
Obviously there is a division separating the two teams but Colchester’s 23rd spot in League One is very heavily the result of poor PDO. Only Yeovil out of the top four leagues in England has a worse PDO score.


Both Cardiff and Colchester are averaging similar shots for per game: 12.71 and 12.35 respectively.
But here is where the statistics start to diverge. The visitors appear a stronger defensive side, with a much lower overall shots against average of just 11.78 per 90 minutes, compared to Cardiff giving up 15.13 shots per game.


So as a result Colchester has far a better Corsi/TSR score than Cardiff: 0.457 vs 0.512. It is a similar story with the shots on target share, although here the difference is slightly less: 0.47 vs 0.505
In raw numbers, Cardiff manages almost one shot on target less per game (3.96) than Colchester (4.71) although the Bluebirds do concede slightly fewer shots on target per game (4.46 vs 4.63).


The only statistic in which Cardiff has a significant advantage over Colchester is in PDO (the combined shooting % and save %, which also serves as a simplified measure for footballing “luck”).

Here its shooting % which is really killing Colchester as the visitors score from just 23.89% of all shots on target, compared to Cardiff’s success rate of 33.68.
The save % (percentage of shots on target saved) is closer, but again Cardiff has the edge: 67.3 vs 63.96.

Overall the PDO scores are 100.98 for Cardiff and just 87.85 for Colchester - second lowest in the football league.

Position	Team	Played	Shots for per game	Shots vs per game	Shots for total	Shots against total	Corsi/TSR	Shots on target for per game	Shots on target vs per game	shots on target for total	shots on target against total	Shots on target share	Shooting % For	Save %	PDO
12 (Championship)	Cardiff	24	12.71	15.13	305	363	0.457	3.96	4.46	95	107	0.47	33.68	67.3	100.98
23 (League One)	Colchester	23	12.35	11.78	284	271	0.512	4.71	4.63	113	111	0.505	23.89	63.96	87.85

So sure, a division is big gap to overcome, but given the turmoil at Cardiff City FC and the poor shot stats comparison, an FA Cup upset is a real possibility at the Cardiff City Stadium tonight – and it may be witnessed by barely a couple of thousand brave souls.

Ticket sales this week were looking poor at best (as indicated by @ViewFromTheNin) and with antipathy growing towards Vincent Tan and seeping in to additional criticism of Russell Slade, it could be a very cold quiet night in the Welsh capital.