Wow, I was unsure of whether the prediction table from last week’s column would become a regular part of “By The Numbers” but I think it’s going to have to be because a: It correctly predicted 8 out of 10 results last weekend – I’ve been doing this for many years and that truly an amazing result and b: the level of positive feedback I’ve received has been brilliant – it would appear you all like it as much as I do! So, without further ado, here’s this week’s predictions (I’ve sorted it so that the strongest home performers are at the top).
You can see that there are a number of teams that you would probably expect to see at the top as this week is far less balanced than last week. To put it in perspective, the odds of Southampton, City, Everton and Arsenal all winning their matches is 7.5% compared to just 1.1% if they were completely equal matches.
I think that there’s a couple of matches that stand out as really interesting – West Brom as favourites at home to Man Utd and Newcastle moving back into favourites in a match. Manchester United lost away to Leicester recently and West Brom are far more solid defensively I think. This prediction could almost ‘feel’ right however I’m concerned that United are a team finally in the ascendency and by incorporating data from six weeks ago it might not be giving a fair crack of the whip to them. Newcastle are a one man band at the moment thanks to Cisse and I think that he could be the “difference that makes the difference” if you’ll pardon the Cliché.
Last week I almost had a nightmare between the sticks when Courtois went off early scoring just 5 points, luckily I have him at 5.94 so he very nearly returned his cost. This week, due to the less balanced match-ups there’s four keepers who seem to offer the best chance of returning their cost. Here’s the graph:
The four players suggested this week are Courtois (7.05m), Szczesny (10.28m), Forster (5.58m) and Howard (7.29m). It appears that Courtois will be back and that makes him a great choice again but many of you will be shocked that Howard is on the list. I know, his defence is extremely leaky but the bookmakers believe that they will win without conceding in better that one in every three matches.
A different view
I thought I’d also have a look this week at sorting the data in a slightly different way. Whilst the normal approach is great at choosing those who will keep a clean sheet, the reverse is equally true and this graph shows (starting from the left) those teams the bookmakers believe are least likely to keep a clean sheet.
It stands to reason that choosing forwards from teams playing against those on the left should give you a stronger probability of goals. The teams? Arsenal, Man City, Chelsea, Everton, Liverpool, Man Utd and Sunderland. This nearly matches those above but they have far more faith in Manchester United.
It’s an international weekend so these predictions have to be taken with a pinch of salt if there are any significant injuries. Ordinarily though Goalkeepers are pretty safe so it shouldn’t be a problem.
I’ll be back next Tuesday with my first column covering the Poisson distribution, which creates a model that suggests not only who will be successful but also the most likely score line (pretty good huh?).