This is the first of what will be a bi-monthly look at the Reds from a statistical viewpoint. In it, Oliver Anderson and I use the kind of analysis premiered in The Red Review to monitor progress and identify where things are going wrong.
As with the book, which covers Benítez's first two seasons, the aim is to use statistics as a way of analysing the game, but without ever believing they are the be all and end all. We're trying to do something a little different, without claiming it should replace more traditional football analysis. As the disclaimer we use says, we don't think stats ever
prove anything, but they do hint at trends and show patterns that are not always blindingly obvious. This is an edited version of the September report, which can be read in full at
www.paultomkins.com/redreview_SEPT06.html , and which includes the full tables for the categories discussed.
August and September provided an undeniably mixed start to the season for the Reds. The team progressed well in Europe, topping Champions League group C after holding on for a 3-2 win against Galatasaray, having beaten Macabbi Haifa in qualifying, and showed some good home form in the Premiership, winning the first three games played at Anfield.
It was in the early league encounters away from home where results were poor. Liverpool have so far managed one draw and three defeats on their travels this season, for one miserly league point. Are the Reds really returning to their away league form of 2004/05 in the league where only three teams (Southampton, Norwich and Portsmouth) had more away defeats, or are there other possible reasons for this tough start? Of course, with the European form looking good, similar success in Europe would be welcome recompense if the league form doesn't kick into gear.
Some may blame rotation (we’ll get to that later), some may blame the international breaks and others may look at strength of the teams and fixtures Liverpool have played this season. Then there's bad finishing, mistakes at the back, and poor refereeing. In truth it’s a combination of a whole host of factors, and we’ll look at as many as possible. The latter point is where we will start, with a look at how strength of fixtures can affect league results.
League FormWhen you look at Liverpool’s home league form, and consider who they have played this season, perhaps it is easy to see why they have won the first three fixtures. As of the weekend games ending 2nd October, the three teams Liverpool had played at home – West Ham, Newcastle, and Tottenham – were in 16th, 13th and 14th positions in the league respectively.
All should end up proving much better teams than those positions suggest, as they are all very capable sides (after all, all qualified for the Uefa Cup this season), but their seven-game form had them in those positions. Similarly, if you consider who the Reds have played away from Anfield this season it again gives us a clue behind the results. Liverpool have faced Sheffield United (18th place), Everton (5th place), Chelsea (2nd place) and most recently Bolton (3rd place), where the Reds are yet to win under Benítez. Even including Sheffield United, Liverpool have so far faced the 3rd-toughest away fixtures in the league this season based on the league positions of their opponents, behind only Spurs and Man City.
Even before the season started you would have to say Liverpool’s five toughest away games would include three already played this season, with the addition of Arsenal at the Emirates and Man United at Old Trafford. When you consider that Liverpool will even get those two fixtures out of the way in the first third of the season (Man United 22nd October, Arsenal 12th November) there will be 26 games left in the league season to put together long unbeaten streaks without extremely tricky away ties.
There will be difficult away ties after that point, but none as difficult on paper as the five that will out of the way by the 12th league game of the season. Has this ever happened before?
These type of extremely difficult away ties can easily turn into momentum killers. You just have to look at the fixtures that ended Liverpool’s impressive runs over the last two seasons to see what the impact of removing these fixtures early in the season might have on potential long runs of successive victories.
Last year Liverpool went on an eight game clean sheet run in the league between 29th October and 26th December, winning all eight games. The next game, which Liverpool won away at Everton 3-1, ended the clean sheet run but at least continued the 100% record for a 9th straight game. The 100% record came to an end with a 2-2 draw against Bolton at the Reebok, but the unbeaten run was still alive. That run finally came to an end two games later as Liverpool went to Old Trafford and Rio Ferdinand stole the points in the last minute.
The Reds’ next long run of impressive league form (An 11 game unbeaten streak) stretched from the end of last season, starting with the 5-1 win against Fulham at Anfield on 9th March, until 9th September, when the Reds where undone by Everton at Goodson Park.
It is important to keep the league form in perspective early in the season; teams do not play an even fixture list until the 38th and final league games have been played. While the Reds have also contributed to their own downfall, with mistakes at both ends of the pitch, the start of the season can be affected by the luck of the draw.
Due to this fact one might get a better perspective from looking at an adjusted (or fixture-standardised) league table to determine who is playing the best, irrespective of the opposition they have played. The true league table itself will always remain the only measure that matters, of course, and it’s where you need to be putting points on the board; but something like this can help gauge the true form of teams, and point out any underlying reasons behind a slow start.
The strength of fixtures any team has faced is based on their opponent’s league position (as of Oct 2nd), with adjustments made for current form and whether they played the fixture home or away. After all, playing Chelsea at Stamford Bridge, where they haven’t lost in the Premiership since the 2003/04 season, is much tougher than playing Chelsea away from Stamford Bridge, where they lost five times last season and have already succumbed to Middlesbrough this season.
The fixture strengths have then been adjusted to the league average and given a baseline of 100. In other words, 100 is the average fixture strength so far this season, with anything falling below 100 being easier and anything above 100 being tougher. For example, Charlton’s fixture strength of 145.5 meant that they had played fixtures 45.5% harder than the league average so far this season.
Up until October 2nd Liverpool had experienced a fixture strength of 105.1, meaning the Reds had played fixtures roughly 5% tougher than the league average so far this season. Many of the teams in the top half of the actual league table have played most of their fixtures against teams in the bottom half.
For example Bolton, who had 28.5% easier fixtures than the league average, had only played two teams from the top half of the table (Portsmouth 3rd and Liverpool 10th), whilst also playing the bottom two teams in the whole league: Watford 19th and Charlton 20th. By comparison, the Reds' three main rivals – Chelsea, Arsenal and Manchester United – have all had fixture list strengths of 85% or lower.
When fixture strength was accounted for, Liverpool moved up three places, from 10th to 7th. Still not particularly impressive, but it somehow sounds about right, given that the Reds had a fairly tough first seven games, without making the most of the chances created.
In the altered league, Chelsea move back to the top, knocking Manchester United into second. Good starters such as Aston Villa and Reading also move up.
Another way to look at the figures would be to say that Liverpool were three ‘adjusted’ points behind Chelsea on October 2nd, rather than six. This means that if all teams continued in that form, Liverpool would close three points on Chelsea in the league simply through playing an easier fixture list the rest of the season. This didn't subsequently happen with the Blackburn game, but logic suggests that the Reds will fare better in fixtures that are easier on paper.
Rotation Liverpool come out top in terms of the most changes made between league games. However, if you remove the very first fixture, where a “weaker” line-up was determined by the crucial, delicately-balanced Champions League qualifier three days later (something only one other Premiership club had to contend with), you get a different picture.
Remove the six changes made between the Sheffield United and West Ham games, and there were 15 changes in six games: 2.50 per game, and below Manchester United. And of course, this is not just down to rotation; Liverpool have also suffered lots of short-to-medium term injuries this season. Both Carragher and Riise were stretchered off in the first game, so that was immediately two enforced changes, while injuries between the Bolton and the Blackburn games meant two more automatic changes. (Going into this weekend’s game against United, both teams have now made the exact same number of personnel changes in Premiership games. No wonder Alex Ferguson defended rotation, but it never gets the blame when United have bad results.)
A manager has to rotate to some degree if his team are to play 60+ games a season. Liverpool have played 122 games the previous two seasons, compared with the 88 games of a team like Tottenham. That’s 34 more matches; or just four short of another full Premiership season. Add the World Cup, and other international fixtures, and you’re adding 20 further games to the footballing calendar of the top Liverpool players in the last two years.
Rafa has to be careful with his most valuable players, as he wants them around for the whole season. It's easy to say Steven Gerrard should play every game, as some pundits have, but when he picks up a muscle strain and misses five weeks, or suffers an alarming dip in form and sharpness, people will then moan about overplaying him.
So, has Rafa got the balance right with his rotation? Is he playing his best players, or is he fielding too many of the less effective squad members?
Rotation, assessed using Goal Value Goal value is a way of comparing any player to another; it's not a foolproof method, but it does tend to highlight who are the more effective players in the league. Every time a player takes to the field, he in some way affects how many goals his team scores and how many they concede. Some attacking players leave their team's defence less protected; some defenders actually help their team to score at the other end. Then there are those players whose inclusion seems to do only harm at both ends of the pitch, and have a negative value for each.
Goal Value to the attack compares a player’s goals and assists per 90 minutes with that of the average of their playing position (out of the entire Premiership), adds a team effect (explained later), and multiplies this by their playing time. In other words, it is their attacking goal value per 90 minutes multiplied by the amount of time they have played, to equal how many goals that player adds or subtracts compared with the average of his position.
Similarly, Goal Value to the defence compares the goals conceded per 90 minutes with that of the average of their playing position, adds a team effect, and multiplies that by their playing time. The total Goal Value is the two combined. A kind of individual ‘player goal difference’ if you will.
Eg., Daniel Agger has been worth 3.35 goals to the Liverpool attack compared with the average centre-back in the league. His inclusion in the team has also seen 3.89 less goals conceded than the average centre-back. Therefore his total goal value is 7.248 goals.
The ‘team effects’ to the attack and defence have been standardized by the quality of each player’s team-mates, and the fixtures he has faced.
346 minutes is the average time played by Liverpool players in the Premiership this season. The players who have played below the average amount have put up a combined total Goal Value of 0.86, whereas those who have played more than average have put up a combined total Goal Value of 5.42, which is clearly much more impressive.
Based on these results, it does look like Rafa is playing the right people more often than not, with the exception of Luis Garcia. At Valencia Benítez used Pablo Aimar sparingly until the second half of a season, saving him for his impact at that stage. Perhaps he has that in mind with Garcia, who is a similar kind of ‘lightweight’ flair player?
It’s worth nothing that Riise, Carragher and Gonzalez would probably have played more if not for illness and injury.
Total Goal ValueAugust’s leader in this category was Ryan Giggs, who was also voted Player of the Month, and who sits 3rd overall. This month’s leader (Goal Value in September alone) was Liverpool’s Daniel Agger, who was also the PFA Player of the Month. Agger currently sits 2nd out of all the Premiership's players.
Agger has been terrific on an individual basis, but more importantly the team has benefited massively from his inclusion – and that's the aim of this particular statistic. It’s interesting to note that Luis Garcia is also in the top five, but no other Red appears in the top 20. Whereas Agger adds a fairly equal amount to the defence and to the attack, Garcia's figures are made up with average defensive numbers and hugely impressive attacking averages, and this follows on from his 2005/06 figures.
Team Stats The average Time Of Possession for Liverpool is down 1.6% on last season, but this is based on four away games as opposed to three home, and difficult away games at that. Shots On Goal is down slightly, and Opposition Shots On Goal is up by a small amount, but not significantly. Opponents’ Corners is way down, less than halved from last season’s already impressive average, which shows that in many ways the Reds are defending better; unless, of course, this season the Reds are conceding goals in situations where last year a corner would have been the most damage sustained. Liverpool’s own Offsides are way up, probably due to Bellamy looking to get in behind teams.
Liverpool are 4th in the league of Shots On Goal; 2nd to Chelsea in fewest Opponents’ Shots On Goal; 2nd in Corners won; and 1st in least Opposition Corners allowed: Watford are 2nd, at three per game, but Liverpool top the list at a miserly 1.57 per game (a rate that will be lower still after the Blackburn game, where just one corner was conceded). Liverpool have racked up the 6th-most Offsides and sit 10th in the most Opponents’ Offsides. Time In Possession puts the Reds 5th in the league.
The number of saves by Liverpool goalkeepers is 20th; or in other words, Reina has had the least saves to make. Another way of looking at it is that he has not made them when he had to. (His personal save percentage is heavily down on last season, when he was joint top with Petr Cech, saving 87% of all shots. By October 2nd his 2006/07 rate was just 73%). Opponents’ saves are high, with only two teams forcing the opposition keepers into more action.
It’s interesting to note that the stats actually improved pretty much across the board away from home (even though three of the four were very tough away fixtures). It was just the results that weren’t good; lots of near misses at one end of the pitch, followed by clinical finishing by the opposition, or almighty cock-ups by the Reds at the back. The Chelsea game was a perfect microcosm of this: far better performance than in recent seasons, with numerous chances created, but none were despatched.
Shots On Goal was up, Opposition Shots On Goal was down, Corners were up, Opposition Corners down, and Time Of Possession was up on last year’s average. The key at Old Trafford on Sunday is to maintain all of these statistics, but to be that bit more clinical in front of goal; and if a Liverpool player has to hit the post, it prompty goes 'in off'.
© Paul Tomkins and Oliver AndersonThe Red Review is available from shops and online stores from November 1st, and signed pre-release copies are still available from www.paultomkins.com, including as part of package deals with my other books.
Also available to pre-order is An Anfield Anthology – the best of my articles since 2000/01, plus new articles and essays, and "Golden Past, Red Future Revisited", which examines what I got right and what I got wrong, as well as how things have changed in the interim.