Can you predict 1.01 defeats in Tennis?
Surface problems aside, the main news from Wimbledon this week was the ‘shock’ defeat of Rafa Nadal, shortly followed by Federer. Nadal was a shock, apart from the fact that last year he crashed out in similar circumstances, it was just a round later. Coincidence, predictable or a shock?
A couple of factors play into the hands of the Nadal shock this time. The first is that Rafa’s game leads him to get injured more often than other players, he’s getting older as well. Much as Murray can’t cope on clay, Rafa doesn’t particularly like grass. But probably the main clue was he had no warm up grass tournament ahead of Wimbledon. For Federer, I’ve been opposing him for ages, much longer and more costly than I care to mention and this is down to his age. The average age for retirement of a professional Mens Tennis player is 32, Federer is 32. So as players age I like to look for upsets. Overall it’s not a strategy that suits many because it involves taking a lot, really a lot, of small losses offset by some big wins.
But of course things are very easy in hindsight. But that said, short priced players are easily opposed when trading, more on that in a second . The thing that is surprising in these cases is to see somebody who traded at 1.01 before the off losing. That’s rare, but certainly not impossible.
1.01 Losers in Tennis
I shuffled through my database of matches and found 442 matches where the favourite started at 1.09 or lower and in that sample 15 players lost which was about right. One thing you can see when you look at most stats is that the market is remarkably efficient. Very often when people present stats that appear to show a bias, it could easily be just an aberration in that data set, sometimes if you are looking for something, you may actually find it! When analysing data you need to carefully clean out every bias, including your own. The rule of thumb is that unless you can explain why the bias occurred, you may just be viewing an aberration in your sample set. Data mining can be statistical deja vu.
If we look at players around 2’s we found 1835 and the lose to win numbers were 921/914. So if a player starts with a implied chance of winning of 50%, un-remarkably, they tend to win around 50% of the time.
Trading vs Traditional betting
This is were people get confused, mis-understand or fail to see the appeal of trading. If you trade rather than place an outright bet, your win percentage should shoot up. Should, because you can actually get the underlying trading process wrong or fall into various psychological traps. Traps where an invisible hand will help you pull defeat out of the jaws of victory.
With this in mind, if you revisit the ‘shorties’ on the Tennis you quickly see a pattern develop. Sure they may win, but before they do it’s not uncommon for them to drift a bit, sometimes even lose. If you look at players trading 1.09 or less then they come in a bit, but there are few ticks left to profit in these cases. However, there is plenty of room above them. Therefore, on average, they drift around 14 ticks away from their start price. If your player is starts at 1.20 then it’s an evens chance they will drift to 1.50 at some point during the game.
It’s for this reason I’m always looking for a set up that benefits from a favourite under-performing or the price of the outsider to come in. While it’s difficult to predict a very short priced loser, it’s much easier to trade. While they may not lose, it’s quite likely they will drift at some point.
Wimbledon is going much better than expected at the moment.
Federer broken in the fourth set
Category: Tennis, Trading strategies
