Savannah McCaskill

As promised in my post about notable players that I would not include in my protected IX, here is an entire post dedicated to Savannah McCaskill. There is probably no player on Angel City that makes me pull out my hair more than her. For most of the past two seasons, she has driven me crazy. Yet commentators love her, the coaches seem to love her, and lots and lots of Angel City fans love her. But plenty of fans are also in my camp. So what’s the truth?

For starters, there is a lot backing her up as a good player. She’s Angel City’s Golden Boot Winner in 2022 and 2023. She won NWSL Player of the Month three times this season and she was on the NWSL Best XI Second Team. When I started looking at the statistics for the season, there were also a lot of positives, and a lot that surprised me. All stats are from FBref.com. She led the league in progressive passes and was second in through balls behind Jaedyn Shaw. She finished tied for sixth in the league in passes into the opposing penalty area, and seventh in passes into the final third. She led Angel City in non-penalty xG + xAG with 10.2 in all competitions, translating to 6g+3a. So far, she sounds like an excellent attacking midfielder.

And it’s true, she definitely gets on the scoresheet, and she provides some real highlight moments as well. But the thing that always bothered me was how much she slowed down the game. I felt like there were countless times when she would receive the ball, she’d wait, letting the defense recover, and then often play a safe pass sideways or backwards. So I was pretty shocked to see that she actually led the whole league in progressive passes. But was I wrong? Her 142 progressive passes represent 14.4% of her total passes. So that means 85.6% of the time she’s not making a progressive pass, but actually her rate is pretty high. It’s the highest amongst the top 6 NWSL players in progressive passes. But what’s odd here is that of that top 6, there are four defenders, and one other midfielder, Sam Coffey, who I would definitely say has a deeper role. So I don’t know that we gain too much from that comparison. I’m not really surprised that Carson Pickett is making a lot of sideways passes, right? But if McCaskill is playing a different role then everyone else in this category, then that begs the question, what exaactly is McCaskill’s role?

I think this is actually really hard to define. I would say that she plays as No 10, for the short answer, but that she has the license to drop deeper, or go wide as she sees fit. Out of possession, it’s always seemed to me like Angel City goes to a 4-4-2, with the wingers dropping deeper to help out on defense and McCaskill and whoever is playing as the No 9 staying high. Sofascore has a season heat map for McCaskill that is almost the entire field, but with a little more on the right side of the field, and with the most common position being in the opposition half, but not in the opposition box.

What this says to me, and what my perception has been, is that when Angel City win the ball back in their own defensive third, McCaskill is one of the furthest players forward. In the buildup, she will drop deeper to receive the ball, wait, and once the rest of the team has advanced, then she will move the ball on. And I think that the stats bear this out. McCaskill is the only player in the entire NWSL who is amongst the league leaders in Progressive Passes Made (#1) and Progressive Passes Received (#5). Nobody else is doing both to a high degree. Either players play the ball forward, (e.g. Sam Coffey) or they receive the ball in attack (e.g. Morgan Weaver). McCaskill is in the unique position of typically being in front of her teammates, but also, typically being behind them.

I think that it’s important when looking at her role to also factor in the sheer volume of it. McCaskill played 94.2% of all NWSL minutes available. The next closest attacker on Angel City is Emslie at 64.2%. Now, it is a testament to her that she is able to consistently play for 90 minutes, and it shows the trust the coaching staff has in her as well. And she’s not just on the field, she’s involved. I’d argue that the attack is run through McCaskill. Her 982 pass attempts were second on Angel City behind Gordon, and her 632 completions were third behind Gordon and Nielsen. You expect your centerbacks to have the most passes, so this looks pretty good, right? It breaks down, though when you look at her passing completion percentage. McCaskill’s 64.4% completion percentage in NWSL games ranks 14th amongst outfield players on Angel City who have played at least 360 minutes. But, I generally think that attacking players should get some leeway in this stat, as you should be making riskier passes and the final third is usually more crowded. But actually, comparing her to other attacking midfielders in the NWSL, McCaskill is only in the 43rd percentile in passing completion. For comparison, Debinha had 70.2% and Savannah Demelo had 67.6%.

I think that things also break down when you look at McCaskill’s stats on a per 90 basis. She leads Angel City in xG, but is 7th in xG/90, tied with Le Bihan. She leads the team in xAG, but is third in xAG/90. First in progressive passes, but third in progressive passes/90. Second in key passes, but fifth in key passes/90. She leads Angel City in shots and shots on target, but is 5th in goals per shot and goals per shot on target (this is fifth amongst attackers to eliminate the randomly high percentages of Riley, Ertz, Hammond, and Spencer who each scored one goal off two shots on target). So if someone else got to play McCaskill’s role with McCaskill’s minutes, would they also be able to replicate McCaskill’s goal and assist numbers? We don’t really know, in the end, because there are only 5.8% of the minutes available that McCaskill wasn’t playing. I do think that Endo would do great in this role. Her technical skill and ability to receive the ball under pressure, coupled with a closer proximity to teammates for one-twos, make this role really intriguing. And I also think that Le Bihan deserves a shot, with her league leading xAG/90 and second place shot creating actions per 90 (SCA/90) this year. In fact, her SCA/90 of 5.95 combined with Angel City’s team shooting rate of .09 goals per shot, would give Le Bihan 11.08 goal contributions if she played the same minutes as McCaskill, compared to McCaskill’s 7 actual goal contributions (4g+3a). And I don’t have any stats, just a feeling, that Leroux could also thrive there at this stage in her career. But since we never saw any of them in that role, it’s just a hunch on my part.

But back to McCaskill, there are two other things that I wanted to look at, and those are some of the attributes that other fans have mentioned as McCaskill’s strengths. First is dribbling. I think I once heard McCaskill describe nutmegs as her superpower. She certainly has some memorable ones, and often leading to some big time goals. However, McCaskill was only fourth on the team in take-ons attempted and in successful take-ons. And her success rate of 45% actually ranks 11th amongst ACFC players (minimum 10 take-ons). She is 2nd on the team in carries, yet only fourth in progressive carries, and sixth in progressive distance. And again, three of the players ahead of her here played fewer minutes. So she’s not usually driving at goal with her dribbles either. She also led Angel City in miscontrols and number of times being dispossessed. Again, that’s going to be skewed higher by the fact that she plays so many more minutes than any other attacker, but is still not a good endorsement of her dribbling prowess. And this, to me, epitomizes why we can’t just trust our eyes. We only remember the big time nutmegs that lead to goals, not the many times that she lost possession.

The second thing I hear from people is that they love that she’s always running. I don’t dispute this at all. To my knowledge, there is no public data on how much distance players cover in the NWSL, or things like sprints or top speed. I’m sure that Angel City keeps track of it, though, and I would be fascinated to see it. If this data is available to the public, then please let me know. We do have data, though on what she accomplishes with that running, namely, is she winning the ball back? In 22 games, she has 12 tackle attempts in the middle third and 6 attempts in the attacking third, so .82 attempts per game and her general tackle rate is 77%, so only .63 successful tackles per game in the middle or attacking third. But these tackles can happen in all kinds of situations. When we look at tackle attempts on dribblers anywhere on the field, she only has five successful tackles on 16 attempts. In fact, she only ranks in the 17th percentile in tackles compared to other attacking midfielders in the NWSL. So, yes, she does run all game, but I don’t think that it really leads to anything. Robbie Keane has a lesson to offer on this, I think. Most of the time, out of possession, he’d just walk around, often offsides, seemingly disinterested. Sometimes a centerback would have the ball only fifteen feet away and Keane would just ignore him. But usually once or twice a game, he’d dart in when their guard is down and win the ball. In my book, it’s better to not run and win the ball, then run all game and never disrupt your opponent.

So I’ve covered a lot of ground here, but I’m not sure that I actually have any clearer ideas about McCaskill then when I started. She’s a unique player and it does become kind of a chicken and the egg situation if McCaskill looks good because of the system and the minutes, or if the system is in place because of her skills. My personal feelings are that Angel City’s tactics and formation are built around getting the most out of McCaskill. Or at least that she’s the player that really thrives in the current setup. But I think that’s coming at the cost of not getting the best versions of Emslie, Endo, or Thompson. Or Press either, if we see her return. The slower buildup means that opposing defenses have an opportunity to get organized, and when our wingers are doing defensive work instead of McCaskill, it negates their threat even more. Thompson is especially a threat in transition, and we should be emphasizing that.

I did think that towards the end of the season, as Henry became more integrated into the squad, she and Vignola handled more of the ball progression duties, which freed up McCaskill. And consequently, she scored a goal in each of the last two games, both times arriving as a late runner. So I think that McCaskill could even benefit if she were less involved in the buildup. As I write this, she is a free agent, but if she is signed to a new contract, I would be interested in seeing her play as a false nine. Currently, when she drops deep, she doesn’t pull any defenders away, but moving her into that role could use her natural tendencies to create more space behind for the wingers. And we’ve never figured out the No 9 position anyway. However, if I was the GM of North Carolina, I’d be doing everything to sign McCaskill. On the Opta Analyst website, in a comparison of team’s playing styles, North Carolina ranked as the most slow and intricate team by a huge margin. Every other team in the NWSL has between 2.4 and 2.9 passes per sequence, while North Carolina has 3.6, as well as the slowest direct speed of any team. I think she’d be a great fit in their system. But my concern is for Angel City, and I believe that Endo, Le Bihan, or Leroux could do a better job than McCaskill if they were given the number 10 role, and comparable minutes, and that it’s in our best interest to move on from McCaskill. We’ll see what the offseason brings.