Hopefully you’ve spotted that I include the number of good and bad passes in the Statshacks and maybe you find it interesting. Maybe you don’t. There’s a certain logic which suggests that the higher the number of passes, the more expansive the style of play. That’s crude and overly simplistic, but there is enough correlation that I thought I’d look at how we’ve been doing over the last few seasons.

One thing that’s immediately obvious when looking at this sort of data is the relativity. For instance, if Villa strung together 150 completed passes in a game it wouldn’t be overly impressive. However, if our opponent had only managed 75 it becomes far more impressive. Villa would have out-passed the other team by a ratio of 2:1. By the same token, 200 completed passes wouldn’t be so impressive when compared to 350 of our opponents.

This lead me to create a new metric I may find a slot for in the Statshacks in future which I’m calling ‘Completed Pass Share %’, or ‘CPS’ if you like your acronyms. It’s a simple concept, not dissimilar from possession. Basically it’s Villa’s completed passes expressed as a percentage of the total completed passes in a game. For instance, sticking with the first example above; if Villa made 150 good passes and our opponents 75 that would be 150 as a percentage of 150 + 75, which is 67%.

Quite arbitrarily, I also decided a CPS of 55% or greater would indicate that we had dominated the passing game, 45% to 55% would represent an even match and less than 45% would mean that our opponents had controlled the passing. Sure, that’s debatable, but that’s what I came up with and I’m sticking with it.

With that in mind I fed the passing stats since 2006 through some filters and came up with the following table, which I hope isn’t too convoluted:-

The left hand column of data shows the games which Aston Villa had more than 55% share of completed passes. If you take the 2006/07 season you’ll note that accounts for 12 games, which is 31.6% of the season. Below that I’ve split the 12 games into won, drawn and lost. So in this case, of the 12 games that we had more than 55% of the completed passes, we won 4, drew 7 and lost just the 1 and the percentages below that row relate to the 12 games. 4 games won is 33.3% of 12 for instance.

What I found surprising is that we apparently controlled the passing game more frequently during MON’s first season in charge and we finished 11th. That is if you’re willing to accept that the number of passes can tell you about the style of play. I think it’s a useful piece of information, but on it’s own, like most statistics if isolated, it only says so much.

As we always say; there’s only one statistic that matters and that’s winning. It would be nice to dominate games playing total football and win, but I’ll take smash and grab wins over sexy losses any day. However, trawling through the data from the last few years I found some interesting illustrations of how football can defy the stats.

Most Complete Passes

Sorting our 136 Premiership games under MON by number of completed passes, 4 of the top 5 are from the 2006/07 season and all were at Villa Park.

The most completed passes by Villa during this period was 373 against Charlton in a 2-0 win back in September 2006. Charlton managed 265 passes themselves, so Villa’s CPS was 58.5%.

Villa’s record for the top 5 games with most passes was 2 wins and 3 draws, scoring a total of 8 goals and conceding 5.  This isn’t exactly total domination.

Interestingly, the 6th overall most completed passes and highest placed away match was against Man Utd at Old Trafford in March 2008.  Stringing together 321 passes with an impressive 82% completion rate didn’t stop Man Utd romping to a 4-0 victory.  With Villa achieving a CPS of 46%, this game falls into the ‘even match’ category, yet that didn’t prevent a heavy loss.

Least Complete Passes

Unsurprisingly, the 5 lowest number of complete passes in a game all occurred away from home, but oddly enough contains a better record of 3 wins and 2 draws. 2 games were from last season and 3 from the ’07/08 season.

The lowest number of complete passes was 109 and you might be surprised to learn that it was during the the 4-4 draw with Tottenham. That’s a game that you might have in your mind as one we might have won if not for Marlon Harewood’s clumsy challenge to give away a penalty, but the home team’s 352 successful passes means that Villa were left with a CPS% of just 23.6%.

The second worst number of passes (112) coincidentally occurred during another 4-4 draw, this time at Stamford Bridge on Boxing Day ’07, a game that was the definition of a Thrilla from the Villa. Yet it also provided our worst CPS of 19.9%.

The next three games were away wins at Portsmouth, Middlesbrough and Everton, scoring 7 and only conceding 2 with an average successful pass rate of just 126 and an average CPS of 33.3%.

This Season

The most successful passes were made during the frustrating 0-0 draw with West Ham at Villa Park and we also enjoyed a CPS of 57.1%. However, the highest CPS%was 63.5% achieved in the 5-1 win over Bolton, also at home.

You might not be surprised to learn that the lowest number of passes was made during the dreadful opening day loss to Wigan – 155. You may be a little surprised to find out that the worst CPS of just 29.1% happened in the 2-1 win over Chelsea.

The average CPS so far this season is 45.26%, which is slightly better than the last two seasons; 44.55% and 44.38%, but a little poorer than the 47.92% set in ’06/07.

Average Pass Completion

And finally, to wrap up a numerically heavy post, let’s just check out the average pass rates over the course of the last few season with that of our competitors during the same matches.

I haven’t done anything sophisticated with the data here, just some basic grouping & simple sorting and I was easily able to see evidence of games where the passing statistics were apparently pretty decent, but Villa didn’t get the desired results and vice versa; where the passing was statistically poor, but we got the right results. That’s football.

I was most surprised to discover how well the 2006/07 compares to the last few seasons, at least when looking at passing. If you didn’t know which season we finished the lowest and had to pick it purely based on the statistics in this blog, you’d be crazy to pick that one.

However, this is one of the great things about working statistics without trying to just find numbers to cherry pick that lend weight to any preconceived ideas you might have; you actually learn something. I don’t know about you, but I like learning something and if there’s one thing to learn from this it’s that you can pass it around all day and get nothing, but sometimes you can be more direct and you’ll win games.

Arsene Wenger knows that, don’t let him tell you otherwise.