we are 7th on Xpoints, 5 Xpoints behind Bournemouth in 3rd, 11 Xpoints behind Liverpool top, level with Chelsea, AVilla and Fulham.
Our results have underperformed our XG, so on that basis, no we are worse off than XG stats say we should be.
-
The Fighting Cock is a forum for fans of Tottenham Hotspur Football Club. Here you can discuss Spurs latest matches, our squad, tactics and any transfer news surrounding the club. Registration gives you access to all our forums (including 'Off Topic' discussion) and removes most of the adverts (you can remove them all via an account upgrade). You're here now, you might as well...
Tactics TFC's Tactical Autopsy Thread
- Thread starter bus-conductor
- Start date
Latest Spurs videos from Sky Sports
we are 7th on Xpoints, 5 Xpoints behind Bournemouth in 3rd, 11 Xpoints behind Liverpool top, level with Chelsea, AVilla and Fulham.
Our results have underperformed our XG, so on that basis, no we are worse off than XG stats say we should be.
I remember when we were 2nd on xPoints. Those were the days…
"You'll never sing that!"
BC, a more nuanced look at these stats would A) consider the defensive side (xGA has us 9th); and B) control for the many multiple goal wins we've had to get a more granular look at our performance.
Of the games we've lost only 1 was by multiple goals and that was last week's loss to Liverpool. But we've had 7 wins and they've all been by multiple goals...feast or famine! And that is what is clouding your understanding/perspective...you think we're unlucky/underperforming but the truth is we are just flat track bullies with the only creditable (inexplicable) win vs Villa...
7 wins - team, score, min scored
- Everton 4-0 (14)
- Brentford 3-1 (8*)
- ManU 0-3 (3)
- WHam 4-1 (36*)
- Villa 4-1 (49*)
- City 0-4 (13)
- Soton 0-5 (12)
And beyond that, if I was looking at a team and saw that ALL (prior to Liverpool) their wins were blowouts and ALL their loses were by the odd goal then I would worry about their setup balance.
Longish post looking at our goal creators...
First a graph of xG involvement vs minutes played:
This is taking the xG of each of our shots and sharing that between the players involved in creating that shot, generally 3 players (the shooter, the assist and the secondary assist). For penalties the credit goes to the player who gets fouled and whoever passed to them, not to the penalty taker.
You can see that all of our "front 5" players are in a fairly narrow band, with Son, Madders and Dom leading the way. Kulu and Brennan have accelerated a bit in the last few games while Son and Madders have flattened off. I've included Sarr in the "front 5" role here but we'll see later that he's a bit special.
If we add in shooting performance (difference between goals scored and xG) to the xG involvement:
It gets a bit noisier because a goal from a low xG chance, or a missed tap-in, give big deviations. But generally, Madders, Son and Kulu pull ahead (at least until Son's missed Penalty which is why he has nosedived), while Johnson, Solanke, Sarr and Timo are behind their xG. Johnson is a bit surprising but this is EPL only and I think he's ahead of his xG in EL and Carabao.
I'm going to take the shooting performance back out of it for the rest of this post because it is pretty noisy. So just xG involvement from here.
The above graphs were per minute played. And that's fair enough, because the clock is ticking. But it's also meaningful to look at chance creation per possession turnover, because you can create more chances by taking bigger risks, but at the cost of giving the ball back to the opposition more often, which is, well, not good (insert Trump meme here).
So I have calculated "turnovers" as the sum of fbref stats for incomplete passes, unsuccessful take-ons, getting tackled, mis-controlling the ball, aerial duels lost, getting caught off-side, errors leading to a shot for the opposition, and shots off target. Here's the graph of xG involvement vs turnovers:
So Kulu really stands out in terms of number of time he turns the ball over relative to the chances he helps create.
That was just number of times possession lost, and not taking into account how often you win the ball back. So the next graph uses "Net Turnovers" instead, which is the difference between turnovers (incomplete passes etc) and winning back possession (tackles, intercepts, loose ball recoveries and aerial duels won). Here's the graph of xG involvement vs Net Turnovers:
So this starts to open up a bit of a gap back to Timo and Kulu, who apparently aren't winning the ball back as often as Dom, Son and Madders.
Also we start to see how Sarr is, as alluded to earlier, a bit special. He's creating chances at a similar rate to the rest of the front 5, but is not turning the ball over particularly often in the process, and is winning the ball back just as often so he's basically vertical on the above graph.
If we look at the "back 5" players (and I've included Sarr here too):
Note I've included Bergvall as "back 5" because his stats so far look more like a "6" than anything else.
So generally most of the back 5 are pretty vertical, except for Porro, Dragusin and Bissouma who are leaning to the right. Porro in particular is a problem: he has turned the ball over 214 more times than he has won it back. Yes he generates some attacking stats in the process but not nearly enough to justify the defensive cost. Kulu has conceded a similar net turnover (226) but helped generate 2.6 times as many chances. Maddison has only conceded 74% as many net turnovers but generated 2.8 times as many chances.
If you look at the back 5 on a xG involvement per minute basis...
... most of them are on a similar trajectory. Bisouma and Radu are the two lagging behind creatively, which is a worry, because they are also losing the ball more often than they win it back.
Sarr is... well, special.
My takeaways from this:
bus-conductor
feel free to heart emoji this.
First a graph of xG involvement vs minutes played:
This is taking the xG of each of our shots and sharing that between the players involved in creating that shot, generally 3 players (the shooter, the assist and the secondary assist). For penalties the credit goes to the player who gets fouled and whoever passed to them, not to the penalty taker.
You can see that all of our "front 5" players are in a fairly narrow band, with Son, Madders and Dom leading the way. Kulu and Brennan have accelerated a bit in the last few games while Son and Madders have flattened off. I've included Sarr in the "front 5" role here but we'll see later that he's a bit special.
If we add in shooting performance (difference between goals scored and xG) to the xG involvement:
It gets a bit noisier because a goal from a low xG chance, or a missed tap-in, give big deviations. But generally, Madders, Son and Kulu pull ahead (at least until Son's missed Penalty which is why he has nosedived), while Johnson, Solanke, Sarr and Timo are behind their xG. Johnson is a bit surprising but this is EPL only and I think he's ahead of his xG in EL and Carabao.
I'm going to take the shooting performance back out of it for the rest of this post because it is pretty noisy. So just xG involvement from here.
The above graphs were per minute played. And that's fair enough, because the clock is ticking. But it's also meaningful to look at chance creation per possession turnover, because you can create more chances by taking bigger risks, but at the cost of giving the ball back to the opposition more often, which is, well, not good (insert Trump meme here).
So I have calculated "turnovers" as the sum of fbref stats for incomplete passes, unsuccessful take-ons, getting tackled, mis-controlling the ball, aerial duels lost, getting caught off-side, errors leading to a shot for the opposition, and shots off target. Here's the graph of xG involvement vs turnovers:
So Kulu really stands out in terms of number of time he turns the ball over relative to the chances he helps create.
That was just number of times possession lost, and not taking into account how often you win the ball back. So the next graph uses "Net Turnovers" instead, which is the difference between turnovers (incomplete passes etc) and winning back possession (tackles, intercepts, loose ball recoveries and aerial duels won). Here's the graph of xG involvement vs Net Turnovers:
So this starts to open up a bit of a gap back to Timo and Kulu, who apparently aren't winning the ball back as often as Dom, Son and Madders.
Also we start to see how Sarr is, as alluded to earlier, a bit special. He's creating chances at a similar rate to the rest of the front 5, but is not turning the ball over particularly often in the process, and is winning the ball back just as often so he's basically vertical on the above graph.
If we look at the "back 5" players (and I've included Sarr here too):
Note I've included Bergvall as "back 5" because his stats so far look more like a "6" than anything else.
So generally most of the back 5 are pretty vertical, except for Porro, Dragusin and Bissouma who are leaning to the right. Porro in particular is a problem: he has turned the ball over 214 more times than he has won it back. Yes he generates some attacking stats in the process but not nearly enough to justify the defensive cost. Kulu has conceded a similar net turnover (226) but helped generate 2.6 times as many chances. Maddison has only conceded 74% as many net turnovers but generated 2.8 times as many chances.
If you look at the back 5 on a xG involvement per minute basis...
... most of them are on a similar trajectory. Bisouma and Radu are the two lagging behind creatively, which is a worry, because they are also losing the ball more often than they win it back.
Sarr is... well, special.
My takeaways from this:
- Porro is a weak link. Maybe it's partly lack of support around him (ie Johnson / Kulu / Sarr not giving him good options compared to Maddison on Udogie's side), but I don't think there's enough there to let him off the hook. Combined with eye-test defensive performance, I think there's a big issue there. Which kind of goes against the gut instinct that he's a great guy and always gives 100%, but the stats are pretty solid.
- Spence is outperforming Porro both offensively (xG per minute) and possission-wise (net turnovers). Combined with eye-test defensive performance it's fast becoming a no-brainer.
- Kulu is exciting / unpredictable and I'd hate to be marking him - I'd almost prefer to be marking Maddison and that's no easy target. But Kulu has gone too far to into risk-taking style and it's costing us too much possession. He needs to dial it down. Which kind of goes against the gut instinct that he's a great guy and always gives 100%, but the stats are pretty solid.
- Maddison and Son have plateaued a bit lately but have been our best performers creatively while not being wasteful with possession (surprised me a bit re Son tbh).
- Sarr is just blossoming and is our biggest hope at the moment. He's matching the front 5 creatively while matching the back 5 defensively. When I have a bit more time I'm going to compare him against the rest of the EPL and see if anyone else comes close in terms of chances created vs net turnovers.
- Bentancur is well ahead of Bissouma, with Bergvall looking likely to overtake Bissouma into backup #6 role sooner rather than later (note: Gray untested in that role to date).
- If we can replace (or redeem) Porro, Dragusin, Bissouma, Werner and (yes) Kulu then we are going to seriously move up a gear. We're currently conceding 50 net turnovers per 90 minutes between those 5, which should be halved if they come back in line with how their team-mates are performing. That's 25 less opportunities per game for the opponents to test our defence's resolve and hamstrings.
bus-conductor
feel free to heart emoji this.Longish post looking at our goal creators...
First a graph of xG involvement vs minutes played:
This is taking the xG of each of our shots and sharing that between the players involved in creating that shot, generally 3 players (the shooter, the assist and the secondary assist). For penalties the credit goes to the player who gets fouled and whoever passed to them, not to the penalty taker.
You can see that all of our "front 5" players are in a fairly narrow band, with Son, Madders and Dom leading the way. Kulu and Brennan have accelerated a bit in the last few games while Son and Madders have flattened off. I've included Sarr in the "front 5" role here but we'll see later that he's a bit special.
If we add in shooting performance (difference between goals scored and xG) to the xG involvement:
It gets a bit noisier because a goal from a low xG chance, or a missed tap-in, give big deviations. But generally, Madders, Son and Kulu pull ahead (at least until Son's missed Penalty which is why he has nosedived), while Johnson, Solanke, Sarr and Timo are behind their xG. Johnson is a bit surprising but this is EPL only and I think he's ahead of his xG in EL and Carabao.
I'm going to take the shooting performance back out of it for the rest of this post because it is pretty noisy. So just xG involvement from here.
The above graphs were per minute played. And that's fair enough, because the clock is ticking. But it's also meaningful to look at chance creation per possession turnover, because you can create more chances by taking bigger risks, but at the cost of giving the ball back to the opposition more often, which is, well, not good (insert Trump meme here).
So I have calculated "turnovers" as the sum of fbref stats for incomplete passes, unsuccessful take-ons, getting tackled, mis-controlling the ball, aerial duels lost, getting caught off-side, errors leading to a shot for the opposition, and shots off target. Here's the graph of xG involvement vs turnovers:
So Kulu really stands out in terms of number of time he turns the ball over relative to the chances he helps create.
That was just number of times possession lost, and not taking into account how often you win the ball back. So the next graph uses "Net Turnovers" instead, which is the difference between turnovers (incomplete passes etc) and winning back possession (tackles, intercepts, loose ball recoveries and aerial duels won). Here's the graph of xG involvement vs Net Turnovers:
So this starts to open up a bit of a gap back to Timo and Kulu, who apparently aren't winning the ball back as often as Dom, Son and Madders.
Also we start to see how Sarr is, as alluded to earlier, a bit special. He's creating chances at a similar rate to the rest of the front 5, but is not turning the ball over particularly often in the process, and is winning the ball back just as often so he's basically vertical on the above graph.
If we look at the "back 5" players (and I've included Sarr here too):
Note I've included Bergvall as "back 5" because his stats so far look more like a "6" than anything else.
So generally most of the back 5 are pretty vertical, except for Porro, Dragusin and Bissouma who are leaning to the right. Porro in particular is a problem: he has turned the ball over 214 more times than he has won it back. Yes he generates some attacking stats in the process but not nearly enough to justify the defensive cost. Kulu has conceded a similar net turnover (226) but helped generate 2.6 times as many chances. Maddison has only conceded 74% as many net turnovers but generated 2.8 times as many chances.
If you look at the back 5 on a xG involvement per minute basis...
... most of them are on a similar trajectory. Bisouma and Radu are the two lagging behind creatively, which is a worry, because they are also losing the ball more often than they win it back.
Sarr is... well, special.
My takeaways from this:
- Porro is a weak link. Maybe it's partly lack of support around him (ie Johnson / Kulu / Sarr not giving him good options compared to Maddison on Udogie's side), but I don't think there's enough there to let him off the hook. Combined with eye-test defensive performance, I think there's a big issue there. Which kind of goes against the gut instinct that he's a great guy and always gives 100%, but the stats are pretty solid.
- Spence is outperforming Porro both offensively (xG per minute) and possission-wise (net turnovers). Combined with eye-test defensive performance it's fast becoming a no-brainer.
- Kulu is exciting / unpredictable and I'd hate to be marking him - I'd almost prefer to be marking Maddison and that's no easy target. But Kulu has gone too far to into risk-taking style and it's costing us too much possession. He needs to dial it down. Which kind of goes against the gut instinct that he's a great guy and always gives 100%, but the stats are pretty solid.
- Maddison and Son have plateaued a bit lately but have been our best performers creatively while not being wasteful with possession (surprised me a bit re Son tbh).
- Sarr is just blossoming and is our biggest hope at the moment. He's matching the front 5 creatively while matching the back 5 defensively. When I have a bit more time I'm going to compare him against the rest of the EPL and see if anyone else comes close in terms of chances created vs net turnovers.
- Bentancur is well ahead of Bissouma, with Bergvall looking likely to overtake Bissouma into backup #6 role sooner rather than later (note: Gray untested in that role to date).
- If we can replace (or redeem) Porro, Dragusin, Bissouma, Werner and (yes) Kulu then we are going to seriously move up a gear. We're currently conceding 50 net turnovers per 90 minutes between those 5, which should be halved if they come back in line with how their team-mates are performing. That's 25 less opportunities per game for the opponents to test our defence's resolve and hamstrings.
Some tip top work as ever.
Hey BC...you miss this counter-programming?
...I mean, otherwise you might as well move this circle jerk to the EvAngelist thread.
Creativity vs Profligacy:
EPL 2024/25 to date (R19), minimum 360 minutes played, at least 0.15 xG contribution per 90.
Onana (Villa), Sarr (Spurs), and Max Allister (Dippers) are the stand-outs in that combination of winning / retaining possession while still helping create chances.
Nkunku (Chelsea) looks amazing but only 400 minutes played so far and most of that was against Ipswich, Southampton, Man City and West Ham so not really representative
Salah (Dippers), Luis Diaz (Dippers), Evanilson (Bournemouth) and Szoboszlai (Dippers) are the stand-outs as highly creative yet efficient with possession.
A lot of very wasteful players at Man Utd, Rashford, Hojlund and Zirkzee in particular, (although Diallo and Casemiro look good)
EPL 2024/25 to date (R19), minimum 360 minutes played, at least 0.15 xG contribution per 90.
Onana (Villa), Sarr (Spurs), and Max Allister (Dippers) are the stand-outs in that combination of winning / retaining possession while still helping create chances.
Nkunku (Chelsea) looks amazing but only 400 minutes played so far and most of that was against Ipswich, Southampton, Man City and West Ham so not really representative
Salah (Dippers), Luis Diaz (Dippers), Evanilson (Bournemouth) and Szoboszlai (Dippers) are the stand-outs as highly creative yet efficient with possession.
A lot of very wasteful players at Man Utd, Rashford, Hojlund and Zirkzee in particular, (although Diallo and Casemiro look good)
Also none of top of the league by a mile Liverpool's.
However, Forest appear very high. It's almost as if playing to the strength of your squad is a benefit to results.
Or…recruiting to match your style is a benefit.
It doesn’t really make sense to alter your whole philosophy because of one positional weakness.
We have been playing, more often than not possibly, with a very changed backline. Recently with only 1 of our first choice back 5 I think. Saying 1 positional weakness is perhaps underplaying the situation a smidgeon?
But yes, either way, play to suit your availbility or make resources avaiable to play to your desired style.
Why the pundits are wrong about Ange Postecoglou: the high line
Stop the presses! Spurs like to get high.
cartilagefreecaptain.sbnation.com
Why the pundits are wrong about Ange Postecoglou: the high line
Stop the presses! Spurs like to get high.
Apart from showing that our line has got deeper since we lost our Cb’s it also shows the difference losing our first choice cb’s has made:
Apart from showing that our line has got deeper since we lost our Cb’s it also shows the difference losing our first choice cb’s has made:
That's good stats but it's just looking at team performance when they are in or out of the team. You can break it down further and look at the individual stats for them and their cover (2024/25 EPL averages to R20):
| Player | Net Turnovers per 90 | xG Contribution per 90* |
| Romero | 0.1 | 0.10 |
| Dragusin (cover) | 3.9 | 0.03 |
| GAP | +3.8 | -0.07 |
| van de Ven | 0.3 | 0.09 |
| Gray (cover) | 2.7 | 0.00 |
| GAP | +2.4 | -0.09 |
That's an extra 6.2 turnovers to the opposition per 90 minutes, and 0.16 less xG created directly by the back 2.
If you look at the key "defensive spine" combination of GK/RCB/LCB/DM it's even more telling:
| Player | Net Turnovers* | xG Contribution* | Backup | Net Turnovers* | xG Contribution* | Turnover Gap* | xG Gap* |
| Vicario | 1.4 | 0.01 | Forster | 4.6 | 0.00 | +3.2 | -0.01 |
| Romero | 0.1 | 0.10 | Dragusin | 3.9 | 0.03 | +3.8 | -0.07 |
| van de Ven | 0.3 | 0.09 | Gray | 2.7 | 0.00 | +2.4 | -0.09 |
| Bentancur | -0.9 | 0.13 | Bissouma | 1.9 | 0.04 | +2.8 | -0.09 |
| TOTAL | 0.9 | 0.33 | 13.1 | 0.07 | +12.2 | -0.26 |
Our best defensive spine combination is directly involved in creating 0.33 xG per game (on average), compared to just 0.07 for the backup combination. But even more significantly, our backup defensive spine is turning the ball over to the opposition 12 more times per game compared to our best 4. That means 12 more opportunities for opposition to score, and 12 less times for us to score. That's huge. And reflected in the team xG against stats when our first choice CB's areent playing, as tabled by MattyFlat in his very good article.
btw I don't think Gray having worse stats than van de Ven above should reflect badly on Gray, I think most of the gap is due to having Forster, Dragusin and Bissouma in support rather than Vicario, Romero and Bentancur.
Last edited:
I have that article open on a tab to read.Apart from showing that our line has got deeper since we lost our Cb’s it also shows the difference losing our first choice cb’s has made:
We had bad results with Romero and VDV but the performances and underlying numbers when they play suggest we’d be much better off right now if they weren’t injured.
Of course we would.
If we had just recruited better the last two windows, and made sure we had two CB's that fit the system, it wouldn't;'t cure all of the flaws and risks of Ange's system, but we'd certainly be more able to operate the system, and we were so much more efficient when we did.
So fucking annoying
We don’t need 6 CBs, we just need 4 that can play and rotate without a massive drop off.
Worst thing is we’ve never really done it. Even under Poch, Davinson was our 3rd CB and was nowhere near as good or skilled as either of them.