The Celts were there at Windsor Park as the 2017/18 season started with a win over Linfield in the Second Qualifying Round of the Champions League. The Northern Irish Champions are one of those clubs fond of World Records, and 24 domestic doubles is undoubtedly impressive. If there is a World Record for having more trophies than supporters, their name would be on that one too.
“Are You Albion Rovers in Disguise?”
In fairness to the Blues, they are a part time team, something Celtic don’t come up against too often. In the three seasons I have been capturing detailed data, I don’t believe there has been a more one sided contest, by many criteria. I can only think of the Scottish Cup tie away at Albion Rovers as a suitable reference point.
There is a case for Linfield being the worst team Celtic have faced, certainly in the last four seasons. Celtic racked up 703 passes, completing 92% of all passes. 28 shots were attempted but 10 were blocked. Whilst 12 found the target, the veteran Carroll saved 10 of them in a Man of the Match performance. Celtic’s 91% possession is unprecedented in my record taking. Linfield went direct with every possession and seemed content to avoid a drubbing as a priority.
Celtic out performed their xG massively last season, scoring 50% more goals than the simple xG model I use would predict, whilst conceding almost exactly the predicted number of goals. Some of this might be down to a cadre of excellent long-range finishers such as Armstrong, Roberts (*sobs*) and Griffiths. One would expect that this profile would regress to the mean over time. That is: you may get a run of luck due to speculative shots finding the top corner and defensive errors being punished, but over time, you’ll revert to the average such that, for example, only 3% of shots outside the box will result in goals. In this match Celtic underperformed their xG by 1.27. The Hoops were accurate with 43% of efforts yet only converted 7% of all shots. This is despite managing to take 15 of 28 shots from inside the box. Finally, Celtic managed 36 possessions in the box compared to 5 by the home side. Early season rustiness, I’m sure (*I hope*).
However, whilst it is not that difficult to get eleven professional players to simply fill space and be compact, and no more, Linfield achieved that limited objective with admirable doggedness and application. The three centre backs never left the width of their box. I mean, never. Such was the depth of defending, the centre forward, Waterworth, was often found half way inside his own half starting the low block.
Staunch Resistance – all ten outfield players are in shot.
An excellent and detailed tactical analysis of how Celtic failed to break down this deep blanket defence can be found in Spielverlagerung. I shall be keeping an eye out for the “Horseshoe Circulation” pattern of death over the season. But credit is due to the Healynaccio. I wonder where he learnt it?
Welcome the Whizz
Welcome to Celtic, Jonny Hayes. The Irishman made his bow at Windsor Park, and packed a fair bit into his 20 minutes. He completed 19 passes, including 2 set pieces, from 24 attempts, creating 1 chance. However, it was his pace and direct running that caught the attention. A hard-working and direct player, Hayes completed as many successful dribbles (3) as any other Celtic player despite being a sub. But it was the impact of those runs that stood out. Or “Impect”.
Time to introduce “Packing” and “Impect”. I have no idea the genesis behind the use of those terms but they are explained very well at Bundesliga.com. It also has the potential to ruin my life. I have undergone a radical widening of the data points I cover per match and I will do an article to explain. Most time consuming is logging “packing”, “pack rate” and “impect”. If I explain what they are this will become clear:
| Packing | A completed pass that, on receipt and control, has taken 1 or more opposition players out of the game. That is, they are now behind the ball relative to their own goal. |
| Pack Rate | The total number of players taken out of the play by a completed “pack”. |
| Impect | A value given to the players taken out of the play:
|
The same principle applies to dribbling with the ball, except there is only the dribbler to credit as opposed to having a passer and receiver for a pass. All other passes that do not result in an opposition player being taken out the game get rewarded with – a pass completion of 1! So now all passes have a value AND we can see the relationships between players – but that is a teaser for another article.
An example. Lustig passes to Forrest on the right, bypassing the centre forward and the wide midfielder:
- Lustig gets 1 Packing Pass for the completed pass (rewards the vision and passing)
- Forrest gets 1 Packing Receive for receiving the pass (rewards the movement and control)
- Lustig and Forrest get Pack Rate of 2 for bypassing 2 opposition players
- Lustig and Forrest get 3 Impect points for bypassing 1 x forward and 1 x midfielder (they have likely exposed the full back, so this is more effective than a pass in front of the opposition).
23% of Celtic passes resulted in an opposition player being taken out the game (as Celtic completed 703 passes, it means 162 times I had to stop the video and work out the pack rates – I hope I get quicker! And Celtic will not get 703 passes in most games). That means that 77% of Celtic passes take no players out the game – for example Simunovic passes to Sviatchenkop square across the back and so on.
Hopefully you get the drift and I will explore this in more detail separately.
Where were we? Oh yes! Hayes! Take a look at the end positions after a Hayes dribble.
In added time, a glimpse of the art of the possible from the Irishman. Receiving the ball wide right with the whole Linfield team in front of him, a simple shift then burst of direct pace up the line resulted in Hayes “packing” the whole opposition team bar the centre forward. 9 players packed, and 21 Impect points, the highest score for a single move in the match. The next highest was when Rogic fed in the rapid Hayes for a pass behind the defence in the 71st minute resulting in 20 Impect points for both. The part-timers were no doubt tiring after their heroic defensive efforts, but still, promising as a “special teams” move.
The position above is clearly a very dangerous one as every single Linfield player has his eyes on Hayes and not what is behind them, which is advancing Celtic players. A well-placed pass gives the attacking team a potential clear chance.
Sheer speed and determined running will be a useful addition to the Celtic armoury. Playing in a side where he is not the main threat, as was the case at Aberdeen, may also help him. I only hope he is the significant upgrade as 4th winger on Mackay-Steven, and not the Roberts replacement (*sigh*).
One might think Forrest can also play a similar role but Hayes looks quicker. This is not a slight at Forrest, who has his critics. Note that Forrest scored 154 Impect points for pass receipts, easily the highest of the game for Celtic (Sinclair with 137 was next). He completed 3 dribbles, joint highest. Forrest also led the starting team with an average of 3.18 players take out the play by his pack actions. Only one player had a higher score – Hayes at 3.89 over his 20 minutes.
A final aside, Linfield as a team achieved a “pack” pass 26% of the time, higher than Celtic. If you constantly bang high long balls, the odds are one or two will stick and you will bypass many opposition players. Also, Celtic were not as compact in defence, favouring a high press which is vulnerable to quick long accurate passes. Much more to come on this.
The Griffalo
No behaviour was modified in the making of these pictures.
There is probably no curing Griffiths of his Griffithness. Both as a player and as a study in eye catching behaviour. A search down the back of players’ socks should raise enough to pay the inevitable UEFA fine on its way, but what of Griffiths the player?
He did not manage 10 completed passes from open play. He had the lowest PEI by miles, keeping possession with only 63% of on field actions. Griffiths did not complete a single defensive action in the match. Only Gordon, who did not make a single save, had a lower Usage Rate than Griffiths’ 3% (he was involved in only 3% of all Celtic possession events – the highest was Brown and Armstrong at 15%). In Dembele’s 19 minutes, he had more touches in the box (4-3), completed more passes (10-9), had a far higher Impect score (45-27) and as many shots on target (1).
And yet.
There was no player in Scotland* that matched Griffths for Scoring Contribution (Non-Penalty Goals and Assists per 90m). He averages 0.84, significantly ahead of the next highest – Dembele with 0.74. Both Celtic goals came from Griffiths corners. The first indirectly, as Forrest recovered Simunovic’s flick at Griffiths corner and returned it for Casement (all the papers and UEFA said Haughey but I have watched it, like, 10 times – it was the blond number 18!) to score an own goal. More directly his clever corner found Rogic to striker the second. Griffiths finished with an Expected Scoring Contribution of 0.32 – Rogic led with 0.66. His 3 possessions in the box resulted in 2 shots, 1 on target. He created 4 chances and had 3 key passes from his total of 13 (9 outfield and 4 set pieces completed passes). 31% of his passes created a chance! He led the team with 0.32 xA.
As noted in We Need To Talk About Leigh, Griffiths be what Griffiths be doing. A squad with two senior central strikers can do without losing one to silly suspensions. But oooh, that produvtivity!
(* This season, there are (so far) two new additions to the Premier League in Scotland with better career Scoring Contributions than Griffiths. Simon Murray at Hibernian is averaging 0.92 NPG and Assists per 90m and the Colombian Morelo at The Rangers has managed 1.05 in Finnish football.)
