Tackles versus Ball Recoveries
Today’s match was a fantastic example of the importance of ball recoveries in understanding the role of the defensive midfielder in football.
First, a ball recovery is, well, it’s when a team on defense regains possession. There are some more, technical, components to the way that Opta collect that data but the best way to think about ball recoveries is as the moment a defensive player turns defense back into attack.
A player can get a ball recovery by picking up a loose ball, by the opponent passing the ball to him, or even by collecting the ball after a teammate makes a tackle. The important thing is that the defender regains possession and keeps possession for his team. This is the essence of defense: to win the ball back and re-start the attack.
Tackles, on the other hand, are not what you think. Last week I got a bunch of questions as to how Coquelin could go 6/6 in tackles, get a yellow card for a bad tackle, and foul at least one more time to send the punditocracy into foaming mouth fever over how he should have been sent off. The answer is simple: tackles are by the Opta definition always a successful event.
When a player goes to ground to win the ball, in order for Opta to count that as a tackle, he must win the ball. If the tackled ball goes to a teammate or to the tackler, he get a successful tackle. If the player attempts a tackle and the ball goes back to the attacking team that’s a failed tackle. And if the player goes to ground and scythes someone’s shin in two, that is counted as a foul and not a tackle at all.
This runs counter to the way that broadcasters and the fans talk about football. When we see Ryan Shawcross attempt to break someone’s leg, we say “that’s a bad tackle” but Opta says “that’s a foul”. So, when you see me say “Coquelin was 6/6 tackling” what I mean is that he went to ground six times, won six tackles without fouling, and Arsenal recovered the ball six times.
Back to ball recoveries. If the point of defense is to regain possession then I think we need to have a stat like “Possessions Gained” which would include ball recoveries, tackles, and interceptions. In Coquelin’s case, he had 5 tackles (led both teams), 6 interceptions (led both teams, tied with Bellerin), and 15 ball recoveries. That’s 26 times that he recovered possession for Arsenal.
The entire team made 80 ball recoveries, 23 tackles, and 28 interceptions. That’s 131 recovered possessions. Since he had 26, that means that Coquelin accounted for 20% of Arsenal’s recovered possessions today and he only played 81 minutes. In the time that he was on the pitch, Arsenal only had 68 ball recoveries. That means he had 22% of Arsenal’s ball recoveries! And accounted for 23% of Arsenal’s total regained possessions for the first 81 minutes. That’s simply amazing.
Cazorla was second on the team and had 20 total recovered possessions (14 ball recoveries, 2 tackles, and 4 interceptions).
Interestingly, if you add Coquelin, Cazorla, and Gabriel together, they accounted for half (40) of Arsenal’s 80 ball recoveries.
One last thing I want to mention which I think tips the books in favor of Coquelin for man of the match: he had two inch-perfect tackles in the box which prevented Liverpool from getting big chance shots on Arsenal’s goal. In fact, despite the ooohing and aaahing of the announcers and the crowd, Liverpool only had one big chance in that game. I put that down, largely, to Coquelin. In that first half, he single-handedly kept Arsenal ticking over: winning the ball back, starting the attacks, and mopping up after the rather disjointed play of Arsenal’s makeshift center back pairing.
Oh yeah, and Coquelin led all players with 91% passing (Walcott went 1/1, that doesn’t count) on 76 attempts.
Cech
Cech had 8 saves today and it’s safe to say he hasn’t been worked that hard in years. In fact, the last time Cech made 8 saves for his club was when Chelsea beat Basel 3-1 in May 2013. He also had a 10 save match when the Czech Republic played Turkey in October of 2014. And, the last time he had at least 8 saves and kept a clean sheet was against Birmingham in the FA Cup in March of 2012.
But not all saves are made the same. Coutinho had a shot from 25 yards straight at Cech, that counts as a save, just like his finger-tip save that pushed the ball into the upright. I think we need a new stat: “big saves”! How many “big saves” did Cech make today? I counted three. You?
A tale of two halves
What a weird game. I tallied up the passes, tackles, and whatnot from the first half and compared them to the second half and it looks to me like both teams had exactly opposite halves.
Chambers and Gabriel
When I saw that Chambers and Gabriel were starting today I literally gulped. Not figuratively, literally. This is a huge game and Brendan Rodgers is a pretty astute manager. He would look to put pressure on Chambers in possession and exploit the 20 year old’s lack of experience.
It nearly paid off. Chambers didn’t turn the ball over constantly (Alexis led both teams in that stat with EIGHT total turnovers and dispossessed) but he did make a lot of poor passes and looked flustered throughout the match. He finished the game with just 81.6% completed passes. But more than the percentage (he only misplaced 9 total passes) it was where he coughed the ball up: 4 of his poor passes were in and around the Arsenal 18 yard box and 3 of those were in the first half. He also had a failed attempted dribble out of his own defensive third. The center back needs to be the rock in the team and can’t be running off like a Vermaelen with his head cut off or kicking the ball to the opposition when flustered.
But he is only 20 and he will learn.
@7amkickoff
All stats via Opta through Whoscored.com and the 442 Stats Zone App.
Got a question? @ me on twitter.
Want to learn more about Opta’s event definitions? Read this.
Great work as always, 7amkickoff! Could we get Gabriel’s defensive stats? Tackles, clearances, interceptions, and so on?
Thanks!
Gabriel was just behind Cazorla: 11 ball recoveries, 4/4 tackles, but just 2 interceptions.
I think one of those interceptions was in injury time inside our 18 yard box and saved us from a late against goal.
Speaking of Cazorla his shots on goal keep increasing in attitude with each game. Whats up with that?
Ox, our wildcard should start. Seems our sharpest shooter (even though he’s one of our worst passers of the ball)
Yes agreed, great work. Could we also get a reality check on points dropped to a not-great Liverpool team?
A shaky defense was saved by Cech, but great build-up play on the other end was wasted…again. We don’t have a great finisher in this squad.
Also surprised nothing was made of the number of stray passes today. It’s been a long time since I’ve seen that many passes simply passed out of touch or miss a teammate. Surely that would be worth mentioning.
50 misplaced in the 1st half is shocking by our standards, worst thing was where the misplaced passes were on the pitch! Seemed like old Arsenal yesterday, tippy tappy no penetration and getting caught out by silly mistakes deep in our own half
I’d love to see some stats on Goalkeepers playing behind a shaky/poor defence vs playing behind a solid back line.
How come Cech played so well today with such a poor defensive showing (firs half) when we were told all last season that was the reason why SZCZ was playing poorly and conversely, why Ospina managed to play well in the 2nd half of the season, he had a solid back four in front of him.
Ospina looked good because the team in front of him was playing so well, he didnt have much to do and unlike szez didnt come charging off his line randomly. Szez had at least one mad moment every game usually charging out of his area or what not. Cech looked better than them both because he made wonder saves. The shot that he tipped onto the post was almost unbelievable, the one low down from benteke at point blank range was just as good. I dont think ospina or szez would save one of those let alone both. So to… Read more »
He was also guilty of some very poor distribution and poor decision making on kicks when under pressure. Overall, I thought he just did his job and don’t see that he adds anything over Ospina particularly.
the irony is that Cech had his best game when our defence was worst.
The defending at WHU begs to differ.
No, the defending vs Livertool was worse than West Ham by miles.. How many saves did he have to make vs West Ham?
This is my point, it isn’t irony, it’s simply a myth that a keeper looks worse behind a bad defence. Look at De Gea last season, he was making 2-3 world class saves per game, despite the sieve of a defence in front of him.
That’s total bollocks. How does having less saves to make and making them make you look better than having more saves to make and making them? Ospina was saving a higher % of shots faced than SZCZ was, if he was having to make more or less saves that % wouldn’t have changed and he’d look no better or worse. To say I don’t think Ospina or SZCZ would have saved either of them is a non point. I wouldn’t have expected Cech, Ospina or SZCZ to save either of them and if Cech faced the same shots this weekend… Read more »
what a great breakdown. You always make numbers interesting. More than your funny claim of proving anything by the numbers, you should know you turn stats into proper narratives.
Well said. We’re lucky to have this kind of analysis for our team. I honestly don’t see it much elsewhere. My only real peeve with this site is how many would rather thumb inane comments up and down in the match reports and ratings rather than contribute to the discussion on the numbers.
Go to tims own website for his narrative take on The Gunners
http://www.7amkickoff.com
It’s always a good read
Some really encouraging things in there after analyzing the game. Initial gut reaction, especially at the end of the 1st half, was “this is awful” and “so much for the title race.” Maybe I’m rationalizing but a peek under the hood provides a rosier view of events. When considering the offensive performance, we failed to score but it was not because we couldn’t create chances. Expected goals estimates for us varied between 1.7-2.6, which means we should have scored at least 1, more like two goals based on what we created, and that’s not counting the disallowed goal. The Liverpool… Read more »
Interesting perspective. I think you are right in many ways, though the way we played for large parts of the first half was indeed shocking. In particular, I think our performance against the ball was perhaps not quite as good as you say. We started pressing Liverpool well, but once we had a few shocking moments and insecurity set in, that translated not only into an inability to make good passes and keep the ball for an organised attack, it also meant the team basically gave up on closing down the opponent. In a sense, our second half performance was… Read more »
There are a couple of accounts I follow on twitter who posts ExG stats; @MC_of_A and @SteMc74. The former a spurs fan and the latter and liverpool fan, but both stats geeks. I enjoy seeing their posts. @SimonGleave is also a bit of an authority on the matter. I agree with you, our defensive insecurity was a direct result of cheap giveaways in our own half. I don’t know how many of their shots came from those but it’s fair to say we’d have faced fewer chances and made more of our own if we did not give it away… Read more »
Do you have data that allows you to differentiate between games played in the rain and games played in the dry air? Also, was the turnaround in the second half due to anything other than Liverpool not being as physically able as Arsenal, and so tiring out? Or put another way…if you spend more energy chasing possession than possessing, shouldn’t a match where the possession balance is very one-sided, i.e. 2-1, mean that everyone knows the second half will be a one-way affair?
Absolutely. Liverpool decided to settle for the draw in the 2nd half. Things like that are why possession % on its own is pretty meaningless.
Nope. Sorry.
That was un ugly drawn game…COYG
Big plus today was Gabriel, showing us that he was born ready. Coquelin was great but the numbers confirmed why it was hard to enjoy his performance. He was forced into tidying up after his teammates and did it brilliantly. Dr. Gooner mentioned we missed Per, and while watching Vincent Kompany back to his best this weekend, it occurred to me that he doesn’t stride forward with the ball the same way but Per Mertesacker is really accomplished in his ability to draw pressing opposition players to him and play sharp and accurate passes to our midfielders. Calum has time… Read more »
I suspect this may be because at the age of 26 he has not mastered controlling a football. He has one attribute- speed. He is lacking in control and awareness of where he should be and thus in a team relying on “automatisms” he stands out like a spare prick at a wedding.
You are correct. The things you say about Theo are disturbing and they summarize the situation he is in. Coming in as a late substitution simply because there was no one else to throw in. Not the type of player you want in such game and his contribution was even less than expected. However, he did sign a new deal, which even worries me most, simply because I don’t see him fit in the team. I really hope he can out-coq his performance and score some goals for us.
it’s a real shame that we have one of the best no. 10’s in the world and refuse to give him a proper striker to convert the chances he creates. honestly, how can arsene think that there’s no other striker that, when put in front of our midfield, would convert more than giroud. it’s awful having to watch this guy run around at a snails pace on the field. he can’t dribble, no pace, hardly converts the sitters served to him on a plate…it’s criminal. you can bounce a ball of the guy – wow. hard to imagine winning the… Read more »
Harsh, but I agree that Giroud was static too often. Walcott does make runs a lot but he’s rarely involved in the buildup play. I think Wenger is tinkering and trying to figure things out
i thought that’s what preseason is for? it’s too bad wenger throws points away early in the season as he tries to figure things out.
Ozil is crap. Not sure why he doesn’t get any stock but Giroud does. He does fuck all and falls over at the slightest touch.
Izzy there muppet
Was at the match and for those that watched on TV it looks very different live as most of you will know I’m sure. My biggest frustration is that wenger continues to deploy Ramsey out side right (or even left in second half against west ham) he drifts in side as wenger has told him to do so not his fault. They pressed us in first half and played a high line, because we had no pace with Ramsey and Giroud we never threatened in break. Second half we dominated but it was all so congested on our left side.… Read more »
Wenger will continue to play the same starting 11 till December when the obvious sill click in and he’ll play Ramsey in the middle and Ox wide.
Liverpool showed what Arsenal fans feared- Liverpool have caught up on Arsenal with great summer buys, while Arsenal only got a new keeper in. Top 4 will be tight. Top two will definitely be City and Chelsea. That leaves two places for either Arsenal, Liverpool or United. I haven’t seen much between the 3 teams.
As much as it hurts John, you are right. I do believe this squad is well good enough to beat both West Ham and Liverpool at home, though not when our manager continues to put square pegs in round holes. I don’t blame wenger for every defeat like some but I feel he has cost us 5 points already with his selections and set up.
I don’t see it that way. I think Liverpool if anything were surprised by how tentative we became after a few mistakes at the back and then capitalized on that. If you want to know what they really think of us, watch the second half; BR had no intention of trying to go for a win despite how they barraged us at the end of the 1st. We are the better team and they knew it.
A save is a save. Got to give credits to the positioning & the focus of the Goalkeeper as well.
Based on what you stated about a tackle falling to the tackler or a teammate, wouldn’t a tackle which falls to a teammate be deemed both a tackle and a recovery, causing the stat to be counted twice?
This match made me wish for a 11/12 van persie striker in this team
There’s just no room for a striker – Giroud gets a hard time. I don’t know if a new striker would make enough of a difference. The midfield is loaded and pass the ball all the way to the box, squeezing the defense into a tightly packed area. Then they spend 5 minutes trying to thread the perfect ball to Giroud amongst 5-6 opposition players, where he must finish with a first touch as they’ve left him no room to maneuver.
The midfield do that because there’s no options ahead of them! Giroud tries but he just hasn’t got the mobility or skill to make quick incisive runs that can be found when we aren’t camped around the penalty area.
Has there been a chance this season yet where someone has had a through ball played to them and only then had the keeper to beat? I can’t recall one
is there Any stat on our win ratio when Chamberlain starts and wen he comes on or doesn’t play at all?
nicw work Tim… “punditocracy”
What I don’t get is that Ramsey started on the right, while Liverpool played with an 18 year old left-back. You would think Ox could have roasted him with power and pace..
Ramsey played wellicht, Gomez was never really tested, since Ramsey isn’t bound to the right side.
This was a problem. And with nobody ahead of Bellerin, Coutinho and their LB were free to pin him back to help Chambers out. There were several of our attacks where Cazorla, Ramsey, Ozil, Alexis, and Giroud were all left of the spot with nothing but open field to the right of it, until Ox was on and Coutinho off.
I kept shouting and pointing the space out to the lads on the screen, but none of them paid any mind.
I experienced the game the same way, really frustrating to see that the left side of lpool wasn’t tested more. I was shouting and pointing at the screen as well.. But I guess many more like us would love to be in the dug out during the game!
Well guys, I guess now we have all woken up to reality and see that, ONCE AGAIN we will only play to finish top four. This team, and especially this manager, are not on the same level as Chelsea, Man City and not even Man United. The only real player we have is Sanchez and only at times Ozil.
Despite our lack of goals, and one disallowed, we are hitting targets. And are creating bucket loads of chances. ‘Big chances’ as you call it, may be a little less, but are getting those too. So goals will come. May be in a game or two. But when it starts falling to place…..Some teams are in for a hiding this season. And talking of Chambers, was this his Bellerin moment??…I think it is. But one curious thing here is the amount of freedom Rambo has. His goal came from where usually Alexis would’ve appeared. And I feel only Ozil had… Read more »
Of course we do, Walcott.
7amkickoff, isit a tackle when a player makes a ‘tackle’, then the ball goes loose, then 2 opposing players struggle for the ball, then it goes loose again only for an opposing player to pick it up. Is it also a tackle is a player makes a ‘tackle’ and the ball goes out of play?
You my friend have asked a question for two.
Now over to you, Tim!
I’m not Tim obviously, but if I read it correctly the stat is entirely possession based – that is, if you initiate a tackle and your team gains possession, it’s a successful tackle. If the opposition gains possession, it’s a failed tackle. If nobody gains possession it’s not a tackle at all.
So in both cases, if I were the Opta, it would be No Tackle. But the first one would be a ball recovery for the opposition, the second an opportunity for Hazard to kick a cheating ball boy.
I would like to see how much distance does Bellerin cover compared to others. When Ramsey is played on the right, Bellerin is the Rightback, Wingback and Attacking midfielder. He has taken over a lots of responsibility.
Great breakdown, very interesting to read! Tough task to assess keeper stats! Just showing saves is pretty arbitrary, as you point out there’s no indication of how difficult the save was. A keeper can only save what is thrown at him so assessing how convincing the save was might be a better way show good/bad performances. You could go for some kind of likert scale for any action where the keeper is involved, excluding distribution (different kettle of fish). Maybe rank 1-5 where 1 is a convincing save that keeps the ball away from danger and 5 is an error… Read more »