Stats And Stuff – 30/12/15

It is quite staggering just how much statistical information about football teams and individual players is now freely available within the public domain. Statistical analysis is now an accepted and growing part of the game and given the quality and depth of the data that I was able to unearth free of charge on the internet I can only wonder at the level of information that is gathered and provided privately to the clubs themselves.

I generally go to a wonderful website, which provides a treasure trove of easily accessible data that can be understood even by a mathematical dunce like me.

What I find so fascinating about using the data before I write anything about Brentford FC is that it makes me question my judgement about pretty much anything that I have seen unfurl on the pitch in front of me when I watch the team play.

Watching Brentford play can be a veritable roller coaster ride with so many highs and lows as your spirits and emotions are taken to the heights and then plummet to the depths all within the course of a ninety minute match. Judgements can be clouded by what you think that you have seen rather than what actually took place out on the pitch.

We also all have prejudices and preconceived views about every player. For example if you spoke to a Brentford supporter today and asked for an opinion on John Swift, our talented midfield player currently on loan from Chelsea, there would probably be some purring comments of appreciation about his quality on the ball, eye for a pass, ability to glide sinuously and effortlessly past opponents and also make late runs into the penalty area but these would probably be interspersed with some grudging mention of his supposed defensive weaknesses, as to the naked eye he does not always appear to track back, press and support his defenders as much as you would like or is needed.

Is John Swift a defensive liability and a luxury player? Fact or fiction? Does his offensive contribution more than make up for his supposed defensive shortcomings? In order to come to some sort of conclusion I consulted the oracle and Stats God at and here are the stark, objective facts, untainted by any bias or rose tinted spectacles.

I looked first at his defensive statistics and they were telling. Swift makes 0.9 tackles per game, comfortably the least of any first team regular, apart from Lasse Vibe. Yennaris and Colin make the most (2.5) and all of his midfield colleagues attempt more tackles than Swift. He also makes less interceptions than any of his team mates and he has yet to block a shot. These stats would therefore appear to bear out the suggestion that defending is not yet a strong part of Swift’s game. Tellingly in a description and profile of his overall game rates his defensive contribution as weak.

Where things begin to look much better for him however is when you look at his offensive statistics. John has scored three goals and made four assists in his fourteen appearances to date. He also takes 1.3 shots on goal every game and makes 1.4 key passes per game, more than anybody else in the team apart from Alan Judge. He also attempts more dribbles than all his team mates apart from Max Colin.

I could break his game down even more, but hopefully the message is coming through loud and clear that John Swift is making an exceptionally effective offensive contribution to the team that more than justifies his starting position, even if he needs to pay more attention to the defensive side of his game, as it is what you do without the ball that can often be just as important as being a Fancy Dan when in possession.

I thought it might be interesting to delve a bit deeper into the Brentford team analysis on and see if there were any trends emerging after the first half of the season. According to the figures our style of play is typified by the following:

  • Possession football
  • Attacking down the right
  • Play with width
  • Short passes
  • Playing in their own half
  • Opponents play aggressively against them
  • Aggressive
  • Consistent first eleven

Our strengths are:

  • Counter attacks
  • Finishing scoring chances
  • Shooting from direct free kicks
  • Creating chances using through balls
  • Creating chances through individual skill
  • Coming back from losing positions

Whereas we are deemed to be weak at the following:

  • Defending against attacks down the wings
  • Aerial duels
  • Defending counter attacks
  • Defending set pieces
  • Stopping opponents from creating chances
  • Avoiding fouling in dangerous areas

These all look pretty much spot on to me and it is reassuring that the figures in this instance back up and totally substantiate the subjective opinion I had already come to after watching the overwhelming majority of our twenty-four Championship games to date.

Our top six performing players given an analysis of all aspects of their game have been Alan Judge, James Tarkowski, John Swift, Harlee Dean, Jake Bidwell and Nico Yennaris, again, no surprises there, and interestingly enough, of the regular players, Toumani Diagouraga and Konstantin Kerschbaumer rate the lowest. Judge and Tarkowski are also rated as the top and fourth best player in the entire Championship to date – a wonderful achievement by the pair of them.

According to the best eleven players in the Championship over the entire first half of the season were as follows:

Martinez (Wolves)

Onuoha (QPR)

Duffy (Blackburn)

Tarkowski (Brentford)

Friend (Middlesbrough)

Gallagher (Preston)

Norwood (Reading)

Stephens (Brighton)

Judge (Brentford)

Forestieri (Sheffield Wednesday)

McCormack (Fulham)

Not too many surprises there, in my opinion.

Of our thirty-six goals to date, one of the highest totals in the league, an eye-opening nine have come from set pieces, including two penalty kicks and two have come from counterattacks. That is a massive improvement on last season.

We attempt just under five hundred passes per game with a seventy-seven per cent accuracy rate. Eighty per cent of our passes are short, but we also hit nineteen crosses every match.

In that respect I only wish we could find out the average number of attacking players we had in the opposition penalty area every time we hit a cross as I am pretty sure that is an area where improvement is still needed.

I suspect that our analysis department might have a few words to say if they saw this article and would draw my attention to all sorts of facts and figures that have escaped my attention or that I have misinterpreted, and I am sure that I have barely scratched the surface of what is a fascinating subject that will become more and more important as the years progress.

Statistics have certainly changed the way that I look at matches and I have found them an invaluable tool in terms of helping me write more sensibly, rationally and objectively about players and matches and avoid going off on an unsubstantiated and ignorant rant.

Most importantly, what they show quite clearly is just how well we are performing as a team and also on an individual basis too.

Maybe It’s Time – 3/1/15

statsI wrote just the other day expressing some mild frustration and concern about the times when Brentford are guilty of overplaying, our long spells of possession come to nothing and we frustratingly lose games where we have had the lion’s share of the ball.

I need to preface those remarks by stating that I remain massively supportive and appreciative of the overall style of play we have adopted, as well as our customary 4-1-4-1 or 4-2-3-1 formation, and am happy to accept our occasional defensive shortcomings as a natural result of our total commitment to exciting, positive and vibrant attacking football.

I was therefore delighted to receive the following thoughtful and detailed comment from another fervent Brentford supporter in Paul Grimes. His view is rather different, and perhaps far more pragmatic than mine, but rather than prejudge, I would prefer that you all make up your own mind once you have read what Paul has to say, and very interesting it is too:

Maybe its time…..

I wondered what exactly the title of this guest piece should be and I looked at ” To try something different,” “To play all three centre halves,” “To surprise our opposition” and, more controversially, “For Mark Warbuton to take some of his own advice.” I settled for “Maybe it’s time to think outside of the box?”

So what am I advocating exactly and are there any clues in the other prospective titles that ended up in the waste paper basket?

analysisWell, yes, in a nutshell I think now is the perfect opportunity to play all three centre halves together and to tinker a little with our possession and footballing philosophies that have got us this far. 

Why change, you might ask? But for me its evolving not changing, for me its learning something now that we will definitely try next season if this season ends up in disappointment. By “disappointment,” of course, I am not talking about relegation, or finishing in the bottom half or even missing out via the play offs, all of which would have been viewed as much more likely than…. disappointed that we did not get promoted to the Premiership having given ourselves such a good chance!

It’s madness to even talk about it but the fact remains that little over a fifty per cent win ratio from our last twenty-two league games would see us in the Playoffs from the position in the league where we are now.

So if we have done so well to get here why am I advocating using a different system? Well first of all it’s more technical than just that change, because I am actually saying let’s use it only for specific games starting on Saturday with Brighton in the FA Cup. A team rejuvenated under an interim coach over their last three matches and with a new manager that all of their players will be looking to impress. They will have had us watched and will also remember us beating them earlier in the season using our current system. But they have their tails up and so we should beware.

Throughout this wonderful season, Mark Warbuton has regularly repeated his mantra of “defeat is okay as long as the players learn from it” so maybe it is time also for Mark to learn from our experiences. For instance our home shots tally is just twelve higher than the total number of shots taken by our opponents in the first twelve matches. Without wishing to be stating the obvious but that is only one shot per game more than our opposition! Well it gets worse because Brentford and our opponents have managed to hit the target an equal amount of times, sixty-six in total.

So are the forwards misfiring, or are we too open and allowing away teams to come on to us a little too much? Well, no, not really, because we have scored eight more goals than our opponents. Twenty-three to fifteen.

Is it due to our preferred style of possession football or tiki-taka or tippi-tappa as my Dad calls it? I don’t think so or more to the point the stats do not back that argument up, as we have only won once with possession above fifty-eight per cent, and that was against Fulham in a charged atmosphere in front of a near sell out crowd at home in a local derby.

No, it’s about when we are NOT in possession, because despite strong stats in our favour in those twelve home games, we have only three times been on the wrong side of the dominant possession line and each of those three occasions have resulted in victories against Brighton, Leeds and Reading.

So instead of trying to work out how we can create more chances or score more goals, or increase the percentage of goals from those shots, maybe it’s time to start thinking outside the box or outside of both penalty boxes if you like? But particularly outside our own box where we have in my humble opinion been a little too open and, if we want to take full advantage of our excellent start, this is where I feel we need to tighten things up. We need to be more defensively minded in the second half of the season, to concede less shots and less goals and to earn the right to win ugly at times.

The player who drops out initially would be Toumani Diagouraga despite some excellent performances so far this season. Both wing backs will have even more freedom to bomb forward, Tarks will have licence to step in front and push JD up, who will in turn push Judge that bit further forward in support of Jota and Pritchard. On both flanks we would have options as Bidwell could tuck in at times and encourage Pritchard to go outside, Jota would no longer be as easy to read as he steps inside, as a marauding Moses up on his outside, or making an angled dash for the inside line would all be additional options, and all would pose problems for the opposition to cover and force them onto the back foot a bit more than has so far been the norm.

tarksAt the back we need defenders who can defend first and foremost, and the defensive positional cover play of each of our defenders needs to be paramount in their minds before they start thinking about where our next goal is coming from.

The defensive line should be curved and resemble half a revolving door. When the right edge is furthermost forward, the left edge is furthermost back, and vice versa, and the lines between our defenders or channels need to be narrower to stop teams getting in behind us, which is where we are vulnerable due to a lack of speed, height and strength.

We do not need to buy another centre half, rather we need to play another centre half. Basically, its time to grind out a few one-nil wins at home and to go away from home and come away with a clean sheet, something we have only achieved twice each this season home and away.

I have read Paul’s comments carefully and can appreciate his main point, that we need to tighten up at the back. If we carry on at our current rate we will finish the season conceding seventy goals and it is hard to envisage us achieving much if that turns out to be the case. In Tarkowski we have a decent defender but also a gifted footballer in every sense of the term, and Paul’s belief is that he could also act as the first line of attack as well as defence.

deanIt is illuminating to look at the arcane world of statistical analysis in order to discover interesting insights into the game, and Mike Forde, formerly Chelsea’s Performance Director, hit the nail on the head when he stated:

If you look at ten years in the Premier League, there is a stronger correlation between clean sheets and where you finish, than goals scored and where you finish.

Chris Anderson is a well known author who specialises in football statistics and analysis and he totally agrees with Forde

Forde is right on the money. Clean sheets on average produce almost two and a half points points per match. And even only one goal allowed still gives a team slightly more than one and a half points on average. But by the time we get to two goals allowed, the point value rapidly declines. And once the other team scores three or four times, the point value declines to zero.

Compare this with offensive production. You might think that scoring at least one goal will help you as much as not letting one in. You’d be wrong. Scoring one goal only gives you about one point from a match, on average. Compare that to one and a half points for allowing one goal only.

craSo the point value of one goal allowed is fifty per cent greater than the point value of one goal scored. Another way to think about this is to ask how many goals a team needs to score to produce the points produced by a clean sheet. The answer for the 2009-10 Premier League season is slightly greater than three.

So a clean sheet produces about as many points for a team as scoring three goals. Keeping a clean sheet significantly increases your chances of winning a match and going from not having a clean sheet to having one, increases the average team’s odds of coming away with three points from .20 to .72 – a staggering difference in the probability of winning of .52.

So maybe Paul Grimes is onto something. Given our own dedicated team of analysts, I am certain that Mark Warburton is well aware of the figures quoted above and it will be interesting to see if the second half brings about a minor shift in emphasis or approach, or if we simply concentrate on executing Plan A better.

Your comments on this highly contentious subject would be greatly welcomed.