Empty Head To Head

Although my first ever bet was on the FA Cup, betting on cup football hasn't been my cup of tea for many years. It's hard enough to calculate probability for league games, never-mind matches between teams from different leagues, and even for games between teams from the same level, league form, for a number of reasons, doesn't carry over. Even those of us expecting a minor tremor somewhere in the Fourth Round must have been surprised by the number of shocks. 

At the time of writing, the highest placed team definitely through to the Round of 16 (which sounds much better than the Fifth Round) is Crystal Palace who, for the third match in a row, came from behind to win, and if Bolton Wanderers and Cambridge United can do the business in their respective replays, and Brighton and Hove Albion can beat holders Arsenal tomorrow (I can't believe I just wrote those words), in-form Palace will surely become hot favourites to win the FA Cup 25 years after their last final appearance(s).

On a more serious note, some of you may have seen the newly launched FootyResults site which unfortunately seems to have fallen into the trap of thinking that historical match-up data is actually meaningful:

If anyone can explain to me what relevance matches from more than five years ago have when determining the outcome of a match today, I'd love to hear it. It's absolute nonsense, as I have explained before. Clubs are not static. Managers come and go, tactics and formations change, players come and go, players age. The names of the clubs are the same, and while it may be a mildly interesting curiosity that Norwich City beat Brentford in League One five years ago, the events from that day have absolutely no value as a predictive factor now. Lazy newspapers have fed us this rubbish for years, (it's a lazy way of filling up some space with what looks like quality data), but surely no one with half a brain buys into this nonsense any more.