Angeball: Defensive Analaysis

I was talking with a friend of mine who is an Arsenal fan (yes, I do actually consider some people who support Arsenal as friends), and he felt like Angeball was unsustainable, and that Big Ange Postecoglou was going to be sacked by midseason next year at the earliest. Obviously, he’s looking at this with Gunner tinted glasses, while I, clearly, don’t want that to be true. Results aside, I’ve really enjoyed watching Spurs play under Ange. I remember a game under Mourinho, when Spurs beat Man City 2-0, but probably only took two shots all game, while allowing Kevin De Bruyne to put what felt like about 25 balls into the box that were never quite converted. I remember after that game just feeling like what was the point of getting up at 6am to watch that game? I could have seen every single good Spurs moment in the shortest available highlight reel. Yes, we won, but I thought we were lucky, and frankly, it’s just not fun watching the other team control the entire match. To me, there is more to fandom then results. If wins were all that I cared about, I’d be a PSG fan or something. So, I’m admitting my bias up front. I want Angeball to stay, but I do wonder if there is any merit in my friend’s argument. I’d like to look at two factors: are we giving up too many chances, and is this style contributing to our injury problems.

To try to compare chances created, I’m going to look at the last four seasons for Spurs on a per 90 basis, as the current season obviously hasn’t completed 38 games. Stats are from footystats.org, which I’m using because the provide xGA for Scotland and Japan as well. However, take it with a grain of salt as their xGA doesn’t perfectly align with other sites. This season, Tottenham has an xGA/90 (expected goals allowed per 90) of 1.46, while actually allowing 1.45 GA/90 (goals allowed per 90). Last year, with mostly Conte and a whole fiasco to end the year, it was 1.64 xGA/90 and 1.66 GA/90. In ‘21-’22, with the best of Conte, it was 1.57 xGA/90 and 1.05 GA/90. In ‘20-’21, mostly with Mourinho, it was xGA/90 of 1.60 with an GA/90 of 1.18.

So, this doesn’t look too bad, actually, although this season’s xGA is the one that I consider the most suspect from footystats. Still, even if we say that the xGA should be higher under Ange, it doesn’t actually look like we’re giving up significantly more chances than we did under previous defense first, park the bus managers. I think that part of this can be attributed to a difference of stats vs. eye test. I don’t know how many times we’ve seen an opposing attacker get in behind the high line only to get run down by van de Ven or Udogie. We see that there was an opportunity for the opposition, but it didn’t actually translate into a shot that would generate xG. And I think that Ange is comfortable with all of this. In his mind, the system is working if defenders are making big plays and the keeper is making athletic saves. There are questions of sustainability, but I think that was also true under Conte and Mourinho allowing the opposition to control the game. Under either system, the opposition will get some chances, and the defense will have to deal with them. The other thing that I get from looking at the graph above is that dramatic difference in xGA and actual goals allowed in ‘21-’22. My take away from that and this current season, is that I think that teams often will overperform their xGA when the whole team buys into the system. Initially under Conte, they did, and had good results. The difference, I think, is that Conte was always publicly complaining about how bad the team was, and there was no path forward for so many players. Eventually players grew disillusioned and stopped buying in. Ange seems to be doing a better job at keeping the whole team on board. Also, Ange has empowered his players to try to make a difference, while Conte was all about giving the power to opposition and forcing Spurs players to be reactive. Anyway, as long as they are buying in, I think that they can continue to overpeform. My one really big concern here would be if there is a long term injury to Vicario, who has been outstanding. He currently leads the league in Post Shot Expected Goals - Goals Allowed (PSxG-GA) with 4.7. But in the next couple transfer windows, I think that we will be able to sign a backup keeper that would be more able to fill in. Which is another important point: right now, at least half the players in Tottenham’s squad are not good fits for Angeball. It should get better, as more players are brought in that fit that mold.

We can also look at Postecoglou’s previous teams for some historical context. In ‘22-’23 with Celtic, they had an xGA/90 of .89 and an actual GA/90 of .95, both best in the league. In ‘21-’22 Celtic had a .94 xGA/90 and a whopping .58 actual GA/90, again both best in the league. Now I know that Celtic tend to dominate the Scottish Premiership, but I think that it shows that Ange’s preferred style does not guarantee giving up a boatload of chances. I’d also like to point out that in both years, the squad actually overperformed their xGA. In his three seasons with Yokohama F. Marinos in Japan, Postecoglou had the following: 1.57xGA/90 and 1.65GA/90, 1.41xGA/90 and 1.12 GA/90, and 1.5xGA/90 with 1.74GA/90. In 2021, Ange left halfway through the season for Celtic, but they had a 1.32xGA/90 and a .92GA/90. Obviously that second season is the best, and that’s the one that they won the title. The third season is a bit of a concern to me, though the correction in the fourth season is reassuring. I think that it’s also important that there was a bigger oscillation in total goals allowed, but much smaller changes in the chances allowed, which I think is more illustrative of the manager’s ability. The point, though, is that given more time, Postecoglou’s defenses tend to improve, and that historically, just because they have fullbacks flying up the pitch, they’ve still found ways to avoid conceding too many goals. Something I heard a lot when he was first hired was that it takes times for his teams to get used to the style. Tottenham actually hit the ground running, but I think that’s still true, and consequently, I don’t expect a regression in defensive statistics in the near future. I would actually expect them to get even more solid, based on this evidence, coupled with the fact that Spurs can buy a higher caliber of player than any other club he’s managed.

The other concern about Angeball is injuries, and Spurs have had a number of those. Some of them clearly have nothing to do with the training, or the style of play (e.g. Sessegnon, Bentancur, and Perisic). But there are also a number of muscle injuries, specifically hamstring injuries to defenders. And in my experience, hamstrings get hurt when either the athlete hasn’t properly warmed up, or the muscle is fatigued and gets overexerted. Postecoglou’s defenders are probably asked to sprint more than the average defender and none of those hamstring injuries happened right at the start of a game. The thing that makes me think that Ange’s preferred style is not causing injuries is his managerial career to date. Tottenham have more resources than any other club that Ange has been at, and that includes training and medical facilities. If the argument is that the human body cannot keep up with this level of intensity, then I think that we would have seen break downs at his previous clubs, and frankly, he wouldn’t have adopted this style. Postecoglou is not a manager that has only worked at big clubs like Guardiola. He did not have the luxury of bringing in the very best athletes on the planet, and those clubs did not have the recovery facilities that Spurs have. One might argue the number of matches played, but as Spurs don’t have European football this year, that shouldn’t be a factor right now. Scotland also plays 38 matches, while Japan play 34 matches, and 30 in Australia. So a little less, but again, not enough for me to think that Angeball is beyond the limit of human endurance. I couldn’t find anywhere that tabulates total man-games lost to injury for Australia, Japan, or Scotland for the times when Ange was coaching there, but we can look at squads to see if there were any obvious injury epidemics.

I’m going to look at the first season Ange was with a team (to see if injuries were a problem as they adapted to his style) and see how many players were in all but one match, and how many games the player who finished 11th played in. Then I’ll compare Ange’s team to the other top teams (who ostensible had better than average luck with injuries to finish at the top of the table). So going back to 2010 with Brisbane, they had 6 players play at least 29 of 30 games, with their 11th highest player in 22 matches. Compared to Central Coast (who finished 2nd) with 4 players featuring in at least 29 games, and their 11th highest player in 24, Adelaide (in 3rd) with 3 players getting to 29 games and their 11th highest player at 23 matches, and Gold Coast (in 4th) with 2 players at 29 games and their 11th highest at 19 games. Brisbane doesn’t seem significantly different than the other teams, to me. Lots of players were able to handle the entire workload for the season, and the depth was comparable to what other teams had.

Now let’s look at Japan. In his first season with Yokohama F. Marinos, we find 2 players that played 33 of 34 and the 11th player at 22 matches. The champions that year were Kawasaki Frontale, who had 2 players that registered 33 of 34 and 26 matches for their 11th player. 2nd place was Sanfrecce Hiroshima, with 8(!) players in 33 of 34 and the 11th coming in at 31 matches. 3rd place was Kashima Antlers with no players getting at least 33 games, and their 11th player at 20 games. Again, nothing jumps out at me. Sanfrecce barely rotate, while Kashima seemed to have all sorts of issues. Postecoglou’s Marinos were squarely in the middle of that.

And finally, Scotland. In his first season with Celtic, Ange won the Scottish Premiership, had 0 players reach 37 games, and his 11th highest player finished with 24 league matches. Rangers finished 2nd, also with 0 players reaching 37, and their 11th highest player at 26 league matches. 3rd place went to Hearts, also with 0 players making 37 appearances, and their 11th highest at 27 matches. And finally, Dundee United were in 4th. Nicky Clark did actually play 37 times in the league for them, and their 11th player made 20 appearances. So, again, Ange’s team doesn’t seem out of place. Zero players reaching 37 games for any of the top 3 teams isn’t surprising to me, since they all also had to play European matches. But even just looking at the 11th man on the top 4 teams, Celtic were not the lowest.

So historically, when Postecoglou has taken over a team, there is no evidence that they are hit with a rash of injuries. Coupled with the idea that a club of Tottenham’s resources should be able to provide better strength and conditioning, better nutrition, better medical, and better recovery facilities, I’m not going to blame the current rash of injuries on the manager. Just one more comparison on this: Newcastle have as bad or worse an injury situation, and this is after Eddie Howe has been there for over a season and a half. It’s nothing he’s changed that brought it on, sometimes these things just happen all at once. So, I’m now going to confidently state, as free of bias as possible, that Angeball is sustainable. I do expect that Ange will make adjustments as he goes, and I would expect something similar to how Klopp brought “heavy metal football” to the Premier League, and eventually kept the spirit of that but slightly reworked to meet the needs in England. Once Ange finds a similar balance, I look forward to seeing Spurs finish above Arsenal again. If not this season, definitely next.