ACFC Goalkeepers

This is the first of a series of articles I’ll do that’s kind of an end of year analysis of each player on the Angel City squad. Each article will be broken up by position group. Obviously, some players played more, therefore there is more data to look at, but I’ll try to flesh out the season for everyone as best as I can.

Didi Haračić

Haračić actually started the season as the backup, but came in as the starter in Week Four, after Anderson had a particularly bad game against Kansas City, and we’ll talk about her next. Once Haračić became the starter, she held onto the spot for the rest of the season. I’ve been pretty clear throughout the season, that I don’t think that Haračić is the answer in goal, but I do want to try to to determine how good a season Haračić had. There were certainly moments in which she made some really good, athletic saves. There were also some in which she allowed goals on what looked like very stoppable shots.

One of the major difficulties in looking at Angel City keepers is that they were playing behind a very shaky defense. Angel City allowed the 6th most Shots in the league, and the 2nd most Shots on Target. They also allowed the 4th highest xGA and PSxG. But all of that would suggest that Angel City also allows the 4th most goals, but they actually allowed the 2nd most.

Shot Stopping

The most indicative stat, for me, of a goalkeeper’s shot stopping ability is PSxG+/-, because it’s only looking at the shots that are on target, and shows what the actual goals allowed versus what we might expect by the shot placements. Haračić had PSxG+/- of -2.4, which is the worst mark in the NWSL, but it needs a little bit of context. AD Franch had -2.3 and Lysianne Proulx had -2.4 as well, and both played fewer games (17 and 6 games respectively). When you look at this on a per 90 basis, Proulx stands out as much worse, with a PSxG +/- per 90 of -.4, compared to -.1 for Haračić. The big distinction is that Proulx and Franch both lost their jobs as starters, with Proulx even getting traded away to Juventus. Also, just to give Proulx her credit, she is only 24 and had only one professional season in the Australian A-League prior to this, so having her first season in the NWSL with an expansion team was throwing her into the deep end. Back to Haračić, though, when we compare her PSxG+/- on a per 90 basis with the rest of the NWSL, it still looks pretty bad, ranking in the 14th percentile. As I mentioned earlier, Haračić had a -.1 and the league average was .13. Other statistics, I think, are impacted by the defenders, as well, but nothing really jumps out. Haračić had a Save Percentage of 69.9% which was good for 42nd percentile, and only recorded a clean sheet in 17.4% of her matches, which was good for 25th percentile. There was no shot stopping category in which Haračić was not below average. And just for context, the best PSxG+/- in the league was Berger for Gotham with a +7.1. And not exactly shot stopping, but I also want to mention Crosses. Haračić stopped 5.6% of the Crosses that came in, which put her in the 31st percentile of the NWSL. The league average was 7.2%, and the leader was Berger again with 16.6%.

Distribution

Angel City tried a more possession based style and Haračić had to do more passing. Last year she had 31.1 Pass Attempts per 90 with 15% of them going for more than 40 yards. This year it went up to 37.6 Pass Attempts per 90 and down to 12.8% going for more than 40 yards. So more passes, and more of them are shorter. I don’t think that there’s much use comparing Haračić to all goalies, as everyone plays differently, so I’ll just compare her passing with the keepers of the Courage, Wave, Gotham, and the Spirit, as they make up to top teams in Possession, along with Angel City.

When we look at this group of five keepers, all of them are close to perfect on short passes, which you’d kind of expect. Amongst Medium Passes, or between 15 and 30 yards, is where we can see more of a difference. I think this difference is also important because of what those passes mean. Primarily, passing that distance, instead of taking the short option usually means that the other team is pressing high, so the keeper can’t pass to the closer defenders, but has to pass straight to the midfield, and break that first line of pressure. Haračić is at a 96% completion rate on these passes, going 501 of 522. That seems pretty good, but that is actually the worst rate of these five goalkeepers. Kailen Sheridan is the best, with 98.6% on 692 of 702. Haračić’s 11 more incompletions than Sheridan doesn’t sound like a lot, but of course any pass that doesn’t go to a teammate is going to be a turnover in a dangerous position. Best case scenario is that it’s a throw in, but worst case is that the other team is gaining the ball in their attacking third, and because your defenders are trying to keep themselves open for a pass, that means that there is probably nobody in position to challenge them. So if we look at it in terms of allowing an extra scoring opportunity every other game, those 11 extra turnovers end up being significant.

In that same vein, let’s look at Defensive Actions Outside the Penalty Area (#OPA), or basically how much of a sweeper keeper someone is. Haračić averaged .43 #OPA per 90, with an average distance of 11.3 yards from goal for all defensive actions. That puts her in the 25th percentile of #OPA per 90 and 42nd percentile of NWSL keepers in terms of average distance. I’m personally of the opinion that the keeper has to stop shots first and foremost, but I don’t think this should be overlooked. A keeper that can come out of goal and play a pass allows the rest of the team to stay higher, and keep their shape. And especially for a team that plays out of the back, I think this is a real help. I also think that the ability to win the ball back earlier and higher can prevent opposing shots from happening in the first place. I’m thinking of that last minute equalizer from North Carolina as I write this.

Final Thoughts

Haračić is a free agent at the conclusion of this season and my opinion is that Angel City should not re-sign her. Her shot stopping, distribution, and other defensive actions were all below the average of the NWSL. Her PSxG+/- was also -1.4 last year, so it’s not like this was one bad year, and at 32, there’s not a lot of reason to belive that she’s going to get any better. It’s important, though, that the NWSL is one of the best leagues in the world. Some of the other keepers here are Olympic medalists and World Cup winners. I think that there are a lot of clubs around the world that would be lucky to have a keeper as talented as Haračić, but if Angel City is going to compete in this league, I think they need a someone that is at an even higher level.

Angelina Anderson

The first thing to say about Anderson’s season is sample size. I was very frustrated that she wasn’t given any additional games at the end of the season when the playoffs were out of reach, because I certainly wanted to see more. She ended up with only three NWSL matches, and two Summer Cup matches. Those three NWSL matches didn’t go too well, with Angel City only getting 1 point from them, but it’s worth mentioning how poor the defense was. In those three games, Anderson faced 22 Shots on Target and conceded 6 of them. It took Haračić until her sixth game before she faced her 22nd Shot on Target, and she also conceded 6 of them. These aren’t necessarily apples-to-apples comparisons, but they both had 1 game each against Kansas City during this stretch. One interesting comparison is to look at how the rest of the team functioned. On a per 90 basis, Angel City was +.57 in Goal Differential when Haračić was on the field, compared to when she was not. Anderson, of course, was -.57. This is not to say that the team was actually good with Haračić, (Goal Differential was still -10) just that Goal Differential was even worse with Anderson. However, xG Differential is actually the opposite. Anderson is +.61, while Haračić is -.61. So there is evidence that the team played better with Anderson in goal, but just got worse results. This is where some additional game data at the end of the season really would have helped. Because Anderson only played the three games, and also because all three of those games were before the Henry and Nielsen trades when the team definitely had a different look, it’s hard to make any kind of accurate comparisons.

Distribution and Defensive Actions

#OPA is the stat that stands out the most for Anderson. Her #OPA per 90 is 2.0 (compared to .43 for Haračić), and her average distance from goal for those Defensive Actions was 18.6 yards (compared to 11.3 for Haračić). Anderson didn’t have enough minutes for her to qualify for FBRef’s comparion tools, but if we compare her at least to the four keepers we looked at with Haračić, who play for the other four most possession heavy teams, Anderson has more #OPA per 90 and plays further from goal than any of them.

Distribution does not look as good. Part of this, for me, is that I think Angel City was kind of a mess in those first three games. I distinctly remember a moment in the Kansas City game when we won the ball in midfield and it was passed straight back to Anderson, but half of the team started sprinting upfield to start a counterattack. So when Anderson received the ball, there were almost no passing options. Still, she was only 94.9% on Medium Passes (compared to Haračić’s 96%). I do think that it’s worth noting that last year, Anderson played 4 games, and had a 97.3% completion rate. What’s interesting, though, is that last year Anderson averaged 18.8 Medium Passes per 90, but this year it was 33 Medium Passes per 90. Haračić’s numbers also went up this year, averaging 22.7, up from 17.3. There was a clear emphasis to play out of the back, though after Zelem arrived, the goalkeeper became less a part of that. My interpretation of that, which is also based on my observations, is that Anderson is a pretty good passer, but that the nature of the style we were playing forced her into making a number of bad passes in the games in which she played. Again, some extra playing time would have helped us make a better assessment. But since Haračić and Anderson were both at similar levels last year, but Anderson’s rate of passing went up about three times as much, I think that the initial strategy was for the goalkeeper to have the ball a lot more, and consequently she had to attempt some passes that shouldn’t be made.

Shot Stopping

Anderson PSxG+/- was a -.7, which is more than twice as bad as Haračić on a per game basis. However that is all due to the awful game in Kansas City when she allowed 4 Goals off 3.2 PSxG. In the first two games Anderson was actually a +.1 PSxG and I don’t think she was at fault for either of the goals in those two matches. The KC game was brutal, and she faced 9 Shots on Target. She also did not play well, but that was one match. Anderson’s Save Percentage was 72.7%. On Crosses, Anderson was also 5.6%, same as Haračić. Again, sample size.

Final Thoughts

Anderson is under contract for next year, but Angel City needs to make a decision about her future. At the end of last season, she was starting a playoff game. Now she can’t even get playing time in games with no consequences at the end of the season. My impression is that Anderson looks like a proper modern goalkeeper, and I think that she is also a good shot stopper. The aspect that she needs to work on the most is communicating with and organizing her defenders, and that requires playing time. If Angel City don’t view her as the starter next year, I think that they should look at a loan move or transfer for her. She’ll only be 24 next season, with plenty of time to keep developing.

Hannah Stambaugh

We never saw Stambaugh play except for a brief cameo in the friendly against Juarez. The indication is that she will not be back for the 2025 season, as her contract had a team option for 2025 that Angel City declined. I also thought that she would have been a candidate to play in the that last game in Portland, just as a thank you. As I wrote about before, this deal never made a ton of sense to me. The No 3 goalkeeper should be the position that you’re spending the least amount of money on, if you’re worried about the salary cap, since that’s the position that is least likely to see the field. Even though Stambaugh’s transfer fee of $10,000 was very low, it also seemed like $10,000 that didn’t need to be spent. This was never stated, but I felt like it was at least possible that bringing in Stambaugh was a move to help Endo feel more comfortable. She’s never seemed particularly unsettled, but it does have to be hard living in another country. So it makes a lot of sense to me to bring in another Japanese player, except for the fact that they brought in a goalie. On the pitch last year, I thought that Endo was always looking for another midfielder that was as technically skilled as she is, and with whom she could combine on one-touch passes. If, in fact, the club felt like this was something that Endo needed to feel settled, I think that a Japanese midfielder would have filled more of the team needs, and softened the blow of Endo’s loss to injury. Again, that’s pure speculation on my part, so maybe this move had nothing to do with Endo. But then I can’t think of another reason to make the transfer for Stambaugh. She’s 25, so not exactly a prospect. She was a backup at Ōmiya Ardija, the 7th place team in a league that is not as strong as the NWSL, so it seems unlikely to me that there was an expectation that she would leapfrog Haračić and Anderson to become the starter. Stambaugh seems like a really nice person and a great teammate, but her signing, to me, is indicative of a lack of overall transfer strategy. In a vacuum, Stambaugh is someone that Endo played with at the youth national team level, so I can see why Angel City might have interest. But when we look at who is already on the roster, and what the other team needs are, it just doesn’t make sense.

Graphs

There were a lot of numbers in this article so I want to put a few of the key stats up where we can easily compare keepers. I’ll include the numbers for Haračić and Anderson, as well as the best and worst keepers in the NWSL in that stat for context.

PSxG+/- per

#OPA per 90

Conclusions

As I said at the top, the expectation from the shots that Angel City allowed was that we’d concede about 38 Goals (4th most), but we actually conceded 42 (2nd most). Defending wasn’t good, but the keepers didn’t do enough either. At a minimum, Angel City needs average goalkeeping, but to really compete, Angel City needs above average contributions. Of the top four teams in the NWSL, who are also the four semifnialists, three of them are in the top 5 of PSxG+/- for the season. The team that didn’t make it was KC Current, who had an underperforming keeper in AD Franch, and brought in Almuth Schult, who’s PSxG+/- per 90 of .37 has been amongst the league’s best this year. It still staggers me that she played for Angel City that first year and practically never saw the field. I just can’t believe that she could win an Olympic Gold medal for Germany, and then play terribly in training for Angel City, and now be back to elite levels.

But back to the current state of things, I think that Goalkeeper is a position that has to be addressed prior to next season. I don’t see any reason to think that Haračić is going to get any better at this point in her career. Anderson looks like she could take that step, but the team needs to commit to developing her, whether that’s away on loan or as the ACFC starter, though there probably will be some growing pains to keep her as the starter. My personal feeling is that this is the best course of action. I don’t think that Angel City will be in position to challenge for trophies next year, so I think there is time for Anderson to grow, and I think that the money could be better spent on other positions. But this absolutely means that Anderson has to be given the proper support to realize her potential, which means not benching her after one bad game. In the best case scenario, Anderson reaches her potential as an elite keeper. Worst case scenario, she doesn’t, but at least the club is making an informed decision about moving on. Regarding Stambaugh, I certainly wish her the best. Having brought her in, I think that it would have made sense to keep her for another season, unless she expressed a desire to leave. But it does seem like the fans like her and the squad likes her. We’ll just have to see what direction the front office decides to go in. It’s still possible that we’ll see all three of these keepers back with Angel City next year, or we won’t see any of them.

All stats in this article are from FBRef.