Ever since I fell in love with women’s soccer, and had more of a view into the leagues of the world (first and foremost thanks to this great website), I wanted to compare the major leagues with some kind of an appraisal. There are players who decide to play overseas in hope they can experience and learn something not easily available at home. Some are successful. Others are less so. If we have a better view of which league has what characteristics we may be able to foretell if a certain international transfer will work as hoped. Moreover, I often feel uneasy with the election of a global MVP (or “Ballon d’Or”) especially in a year, like 2013, with no major global competition, therefore players are appraised by their league performances. But comparing players playing in different leagues without assessing their respective leagues could easily lead to emotional popular vote not very different from, say, an Oscar Award in the movie industry. It is a big question, of course, if any objective and fair assessment is possible. But I think the key for an assessment deemed fair by majority is to do it “scientifically,” which in this case means judging everything purely by numbers, or “statistics” as it is called. I know there are souls who dispute it, insisting soccer is a richer endeavor of human beings than judged by mere numbers. I agree with them. However, we also know numbers are practically the only argument which precludes any serious disagreement. If we start arguing “who is the better dribbler than whom?” “which team is more aggressive than others?” “which league plays most fairly?” with only words and without solid numbers, then a hundred pundits make a hundred arguments. But there is no one who says 3 goals is less than 2 goals. Numbers are something indisputable, and this is why we divide the winners and losers mercilessly only by their number of goals, regardless of how technically, how dedicatedly, how beautifully each played. Because that is the way of science!
And this is my first attempt at a “scientific” appraisal of footballs and footballers of the world. Warning! This article uses a lot of data, and I’m afraid so many people would find it only boring, so, please do not trouble yourself unless you think you are very comfortable with mathematics!
This time I picked six leagues—Frauenbundesliga, D1 Feminine, NWSL, FAWSL, Damallsvenskan, and Nadeshiko to acquire data for my analysis. Limiting the specimen to these six is purely my personal limit and any attempt to add more is most welcome. Data of German and French leagues are from 2012/13 season. Others are from 2013. Here is the raw data. (Data of Nadeshiko League are obtained from Nadeshiko Official. Data of other 5 leagues are from Soccerway.)
Note: The points within D1 Feminine are re-calculated using commonly used system: win=3, draw=1, loss=0, because this is a key measure to compare leagues.
DIVIDE WITHIN LEAGUE
To begin with, I would try to ascertain which league is “divided” and which is not. With “divided” league I mean one which contains a wide variety in the strength of teams and often sees one-sided games, while in “undivided” league teams of equal strength compete neck and neck. In order to quantify this property, I employ the statistical index called “standard deviation” and apply it to teams’ points. To put it simply, standard deviation is a measure to tell how widely a set of data is distributed on both sides of mean value. The larger this index, the wider the variance of data—in this particular case, strong is stronger and weak is weaker. (Here I use the “STDEV” function of Microsoft Excel to calculate. For example, the standard deviation of points in the FAWSL is STDEV(36,33,31,18,15,10,10,6).) The result is:
D1 Feminine 19.07
Nadeshiko League 12.80
There is no major surprise here because D1 Feminine is a league of 1 mega-giant, 3 middle-giants, and 8 dwarves while NWSL drafted players to create equal teams. Some may be surprised to see the Nadeshiko, with a dominant INAC, is yet less “divided” than the German or Swedish league. However, this initial result is misleading because different leagues play different numbers of games. Generally, more games bring wider variance. (This, in fact, is the Second Law of Thermodynamics and the same reason why dropped ink spreads wide in a water tank as time passes.) For example, the value for the Nadeshiko League after a full 18 games is 12.80 as shown above. But the league was as shown below after the first 9 games and the standard deviation of points was 7.29.
The fact the final value is less than double the value at the mid-point indicates that teams were more equal in the second half of season (surely, Kobe lost twice but Urawa was revived) even if the absolute value of deviation still increased. For this reason, I apply a factor of compensation to compare leagues. Teams of the four leagues—Frauenbundesliga, D1 Feminine, Damallsvenskan, NWSL—play 22 games in a season. Nadeshikos play 18 and FAWSL 14. Therefore, the values of “22-game” leagues stay as they are while the value of the Nadeshiko is multiplied by 22/18 = 1.22 and the value of the FAWSL is multiplied by 22/14 = 1.57. The new result is:
D1 Feminine 19.07
Nadeshiko League 15.64
Note: On purely mathematic ground we also have to compensate by the number of teams because more teams increase the data volume thus its variance. But I won’t, because this is the point I am discussing. If you bring in more teams you will make your league more divided (strong is stronger and weak is weaker.) More teams is a REASON for the divide while more games lets it “looks” divided. Putting it other way, small number of games finishes a season before true scale of divide shows itself. (But anyone is free to apply this second compensation and see what happens.)
In the compensated result, D1 Feminine is still the most divided, but next comes the FAWSL, then followed by the Nadeshiko League. Such teams as Liverpool, Arsenal, Kobe and NTV make their impact felt. In contrast, the draft system of NWSL works well to level the league.
One interesting analysis is to check the standard deviation for goal surplus (“+/?” in tables), again compensated same way.
D1 Feminine 59.47
Nadeshiko League 30.43
Again, it’s no wonder D1 Feminine comes top with its super variance from Lyon’s 127 surplus to Toulouse’s 64 deficit. But FAWSL who shows the second largest divide in points comes fifth in this list, and it means their top teams win constantly but by smaller score leads. Same can be said for the Nadeshiko which falls way below Frauenbundesliga almost to par with Damallsvenskan. I believe this hints these two leagues—English and Japanese—are defense-oriented. Coming to the opposite end is Frauenbundesliga. Their upper teams win by large score leads, but find it difficult to keep winning. And NWSL is the most “equal” league by this measure too.
Apart from the issue of equality, this result may give some guidance, if a conjecture, to the national teams. For example, if the English and Japanese players are really used to winning by defense then playing too aggressively at international competition may result in them being upset. Is this a part of reason why the Nadeshikos are struggling in 2013? How about England in UEFA Euro?
WHO IS THE BEST?
Now the season to elect Ballon d’Or, or global MVP, is come. However, even if I have a full-hearted respect to elected players, I am also skeptical to the very idea of saying officially who is the best. How we can make the election process objective, comprehensive, scientific? Is it possible with soccer? In fact, soccer is in sharp contrast with another sport I closely follow, baseball, when it comes to the issue of who is better than whom. Baseball sets multiple measures, all quantified, to appraise individual players. Average, hits, homeruns, RBIs, SLG, walks for a batter; ERA, wins, holds, strikeouts, complete games for a pitcher; stolen bases for a runner; fielding average for a fielder, and so on. Soccer counts goals, maybe assists if necessary, concedes (for GK), and … anything else? So many players are uncovered by any quantified measure. It’s a sport to appraise a team as a whole, not individual players. If I understand cultures of the world, these two sports, baseball and football, are distinctive representatives of American and European cultures.
And yet, we elect MVP, not only within a particular league or tournament but a global MVP, with or without major global tournament. If so, isn’t it a minimum requirement to compare players playing in different leagues on level ground? And this is my personal attempt, incomplete but best I can manage. Used measure is number of goals. For my analysis, I picked three most scoring players of each league. Here is the raw data to give player’s name, goals in a season, total playing minutes, and her goals against each opponent.
Note: 3 additional players are taken from the French league for the reason to be stated later.
First of all, I calculate, from each player’s goals and playing minutes, how many she “could have scored” had she played 2000 minutes, to compensate for the difference in minutes. (For example, Ogimi scored 18 goals in 1848 minutes. Therefore, her hypothetical scores in 2000 minutes is 18 ÷ 1848 × 2000 = 19.48.) The results are:
#1: Conny Pohlers, 16 goals in 1614 minutes ? 19.83
#2: Yuki Ogimi, 18 goals in 1848 minutes ? 19.48
#3: Mandy Islacker, 15 goals in 1679 minutes ? 17.87
Note: Pohlers tops Ogimi even if she scored 2 less goals because she produced them in considerably shorter minutes.
[D1 Feminine 2012/13]
#1: Lotta Schelin, 24 goals in 1095 minutes ? 43.84
#2: Eugenie Le Sommer, 20 goals in 1220 minutes ? 32.79
#3: Camille Abily, 20 goals in 1640 minutes ? 24.39
#1: Lauren Cheney, 12 goals in 1740 minutes ? 13.79
#2: Sydney Leroux, 11 goals in 1694 minutes ? 12.99
#3: Abby Wambach, 11 goals in 1873 minutes ? 11.75
#1: Natasha Dowie, 12 goals in 1237 minutes ? 19.40
#2: Nicole Rolser, 10 goals in 1217 minutes ? 16.43
#3: Toni Duggan, 9 goals in 1135 minutes ? 15.86
#1: Christen Press, 23 goals in 1770 minutes ? 25.99
#2: Pernille Mosegaard-Harder, 17 goals in 1871 minutes ? 18.17
#3: Anja Mittag, 13 goals in 1802 minutes ? 14.43
[Nadeshiko League 2013]
#1: Beverly Goebel-Yanez, 15 goals in 1203 minutes ? 24.94
#2: Nahomi Kawasumi, 12 goals in 1620 minutes ? 14.81
#3: Ji Soyun, 9 goals in 1440 minutes ? 12.50
Seeing these results laid out, everyone believes no further argument is needed. Lotta Schelin is by far the most productive goal-scorer not only of the French league but the whole world—almost an advent from another dimension.
However, we must keep in mind two reasons why things are not so simple—(1) goal-scoring opportunity varies a lot depending on which team she plays for, and (2) some leagues are easier environment than others to score goals. Don’t mix them up in confusion. The first tells Schelin is in more beneficial position than her NT fellow Asllani. The second tells I will win the Golden Boot if I joined a league of kindergarten kids.
Since these two factors can be viewed as “independent” (neither is influenced by the other) I apply their compensation methods separately. Let’s begin with (1) because this deals with balance within a league therefore a domestic issue. I explain my method using Toni Duggan as example.
Duggan played for Everton and scored these 9 goals in 2013.
4 against Chelsea, 3 against Bristol, 1 each against Doncaster and Arsenal.
However, it is assumed scoring against Doncaster who conceded 42 goals in a season is easier than scoring against Arsenal who conceded 11. How easier? Based on the result alone, the reasonable number is 42/11 = 3.82. We assume scoring against Doncaster is 3.82 times easier than scoring against Arsenal. Conversely, a score in a Doncaster game has 11/42 = 0.26 (26%) “value” of a score in Arsenal game. We apply this principle to compute “value” of a goal against each team. In FAWSL, 2013, Arsenal conceded least goals, 11. If we define the “value” of a goal in Arsenal game as 1.00 then we can determine the value of a goal against each team.
Against Arsenal: 11/11 = 1.00
Against Lincoln: 11/15 = 0.73
Against Liverpool: 11/19 = 0.58
Against Bristol: 11/20 = 0.55
Against Birmingham: 11/21 = 0.52
Against Chelsea: 11/27 = 0.41
Against Everton: 11/30 = 0.37
Against Doncaster: 11/42 = 0.26.
Now we can calculate the overall “value” of 1 goal by Duggan.
(value of Chelsea game goal × goals in Chelsea games
+ value of Bristol game goal × goals in Bristol games
+ value of Doncaster game goal × goals in Doncaster games
+ value of Arsenal game goal × goals in Arsenal games)
÷ total goals
= ( 0.41 × 4 + 0.55 × 3 + 0.26 × 1 + 1.00 × 1 ) ÷ 9 = 0.51
Employing the same method for Natasha Dowie and Nicole Rolser, we get
Natasha Dowie: 0.41 (Doncaster 4, Everton 4, Chelsea 1, Birmingham 1, Lincoln 2)
Nicole Rolser: 0.46 (Birmingham 5, Chelsea 2, Bristol 1, Everton 1, Doncaster 1)
Applying these “value” factors to each player’s goals in “2000 minutes” computed above, we get
#1: Toni Duggan, 15.86 × 0.51 = 8.00
#2: Natasha Dowie, 19.40 × 0.41 = 7.94
#3: Nicole Rolser, 16.43 × 0.46 = 7.58
We conclude, then, that Toni Duggan who scored the third most goals is actually the “best” goal-scorer of FAWSL in 2013. Obviously, her valued goal against Arsenal matters.
Now, we apply the same method to all leagues. First, we compute value of a goal against each team for all six leagues.
[Frauenbundesliga] where value of a goal against
Wolfsburg: 16/16 = 1.00, Potsdam: 16/16 = 1.00, Munchen: 16/24 = 0.67,
Frankfurt: 16/26 = 0.62, Essen: 16/30 = 0.53, Freiburg: 16/31 = 0.52,
Neuenahr: 16/29 = 0.55, Leverkusen: 16/40 = 0.40, Duisburg: 16/47 = 0.34,
Jena: 16/47 = 0.34, Gutersloh: 16/72 = 0.22, Sindelfingen: 16/73 = 0.22
[D1 Feminine] where value of a goal against
Lyon: 5/5 = 1.00, PSG: 5/10 = 0.50, Juvisy: 5/16 = 0.31,
Montpellier: 5/23 = 0.22, Saint Etienne: 5/35 = 0.14, Guingamp: 5/43 = 0.12,
Yzeure Allier: 5/44 = 0.11, Rodez: 5/49 = 0.10, Vendenheim: 5/74 = 0.07,
Issy: 5/74 = 0.07, Arras: 5/76 = 0.07, Toulouse: 5/81 = 0.06
[NWSL] where value of a goal against
New York: 20/20 = 1.00, Kansas City: 20/22 = 0.91, Portland: 20/25 = 0.80,
Sky Blue: 20/26 = 0.77, Boston: 20/34 = 0.59, Chicago: 20/36 = 0.56,
Seattle: 20/36 = 0.56, Washington: 20/39 = 0.51
[Damallsvenskan] where value of a goal against
Malmo: 14/14 = 1.00, Tyreso: 14/24 = 0.58, Linkoping: 14/25 = 0.56,
Umea: 14/29 = 0.48, Goteborg: 14/31 = 0.45, Pitea: 14/32 = 0.44,
Vittsjo: 14/34 = 0.41, Orebro: 14/35 = 0.40, Jitex: 14/37 = 0.38,
Mallbacken: 14/40 = 0.35, Kristianstad: 14/43 = 0.33, Sunnana: 14/82 = 0.17
[Nadeshiko League] where value of a goal against
Kobe: 13/13 = 1.00, NTV: 13/15 = 0.87, Iga: 13/17 = 0.76,
Okayama: 13/19 = 0.68, Chiba: 13/20 = 0.65, Sendai: 13/24 = 0.54,
Urawa: 13/26 = 0.50, Niigata: 13/32 = 0.41, Osaka: 13/44 = 0.30,
Takahashi: 13/49 = 0.27
With these value factors, we compute the “value-compensated” goals of scorers.
#1: Yuki Ogimi, 19.48 × 0.41 = 7.94
#2: Conny Pohlers, 19.83 × 0.35 = 6.97
#3: Mandy Islacker, 17.87 × 0.38 = 6.77
Note: Pohlers falls below Ogimi again because 12 of her 16 goals are against the lower 4 teams while Ogimi scored 10 out of 18 against the upper 7 (not 8, becuse she couldn’t score against Potsdam.)
#1: Lotta Schelin, 43.84 × 0.17 = 7.56
#2: Camille Abily, 24.39 × 0.11 = 2.57
#3: Eugenie Le Sommer, 32.79 × 0.07 = 2.38
Note: See the low “values” of a goal by Lyon players! It clearly indicates they are scoring easy goals against easy opponents. Even with this “devaluing” factor, Schelin still rules with her towering 7.56. But the second of the league is neither Abily nor Le Sommer. It is Hoda Lattaf mainly thanks to her “valued” 1 each goal against Lyon and PSG. (I included the 3 additional players for D1 Feminine to make it sure truly “valued” players are not missed.)
Hoda Lattaf, 20.87 × 0.16 = 3.25
#1: Lauren Cheney, 13.79 × 0.71 = 9.76
#2: Sydney Leroux, 12.99 × 0.71 = 9.19
#3: Abby Wambach, 11.75 × 0.69 = 8.07
#1: Christen Press, 25.99 × 0.32 = 8.39
#2: Pernille Mosegaard-Harder, 18.17 × 0.40 = 7.36
#3: Anja Mittag, 14.43 × 0.43 = 6.17
#1: Beverly Goebel-Yanez, 24.94 × 0.58 = 14.43
#2: Nahomi Kawasumi, 14.82 × 0.49 = 7.32
#3: Ji Soyun, 12.50 × 0.44 = 5.54
Here we already see a result which is surprising yet convincing if we believe in numbers. The best goal-scorer in the world is (as I claimed over and over and over and OVER!) Bev, with almost double point of Schelin! Bev’s point is so high because she scores against difficult opponents (four against Okayama but none against Osaka, for example) as if she feels excited when challenging the strong and rich but is reluctant to bully the weak and poor.
But we are not finished yet. All the data used so far is obtained within each league therefore it’s insufficient to compare different leagues. We defined value of a goal against Wolfsburg, Lyon, New York, Arsenal, Malmo and Kobe all as 1.00, therefore equal. Of course, we have done nothing to support such basic assumption. What we have revealed is that Bev is by far the greatest goal-getter of the Nadeshiko League. If we want to see what status she, or anyone, occupies in the world, we have to take the factor (2) into account. Which is the league where goal-scoring is difficult, and who is scoring there?
Unfortunately, there is no scientifically solid way to compare leagues based purely on each league’s internal data. We can’t tell which of these two cars—one sold for $20,000 and the other for €15,000—is more pricey unless we know the currency exchange rate. We need a reference. But what can be it, if there is any? Europe has its UEFA Champion League, but is there anything global?
How about national teams’ most recent performance in major international tournaments?
[2011 Women’s World Cup]
Japan conceded 6 goals
USA conceded 7 goals
Sweden conceded 6 goals
France conceded 10 goals, all in 6 games
USA conceded 6 goals
Japan conceded 4 goals
Canada conceded 8 goals
France conceded 8 goals, all in 6 games
If we rely on the data above we can assume, at least tentatively, that the Nadeshikos are more difficult than others to score against, and it adds even more value to Bev’s record. (In fact, this is what I personally believe.) However, the national team is not necessarily the right representative of a country’s top league. The presence of top-rated foreign players in such leagues—Bev, Soyun, Schelin, Ogimi, Press, Mittag, Anonma, Marta, Rapinoe to name but a few—makes it a tricky task to equate a national league with a national team. Also, we know that a national team is often less than 100% serious in some friendly matches. As Norio Sasaki’s “experiments” demonstrated in 2013, a national team plays a friendly match more to test than to win. And, in any case, matches played by a national team are too few to be a reliable source. On top of that, different national teams play different opponents which makes what the scientists call “controlled” data acquisition impossible.
What if we use a history of players who transferred from a league to another as such reference? For example, Nicole Rolser scored 2 goals in 11 matches, or 1 goal per 5.5 matches, for Neuenahr in 2012/13 season, then joined Liverpool and scored 10 goals in 14 matches, or 1 goal per 1.4 matches, in 2013 season. Does it indicate FAWSL is 5.5/1.4 = 3.93 times easier to score than Frauenbundesliga as environment? But wait! Yuki Ogimi scored 18 goals in 21 matches for Potsdam, then joined Chelsea and is yet to score her first goal after 5 games. The history of the two players point 2 directions 180 degrees apart. Ogimi, in fact, scored 7 goals in 20 matches, or 1 goal per 2.86 matches, for NTV in 2009 season, then joined Potsdam and scored 6 goals in 10 matches, or 1 goal per 1.67 matches, in 2009/10 season. Does it mean Frauenbundesliga is 2.86/1.67 = 1.71 times easier environment than Nadeshiko League? But wait! Kozue Ando scored 18 goals in 21 matches for Urawa, then joined Duisburg and scored 6 goals in 10 matches, i.e. 1.16 and 1.67 matches per each goal respectively. Her history indicates Nadeshiko is the easier of the two. And remember, Ogimi and Ando moved to Germany in a same year.
I am afraid history of individual players is no more reliable than history of national teams, because it depends on too various factors against which we have too small pool to sample our data to average it out.
Let me repeat what I stated at the beginning. I am skeptical about the very idea of electing global MVP, especially in a year without major global competition. Even baseball, with so many quantifiable measures, would not dare elect global MVP (even if the Americans call their domestic championship “World Series” and elect “World Series MVP.”) Still, Ballon d’Or is elected, with or without a World Cup or Olympics. Let us see then what scientific measure, or something close, is attainable.
First of all, we must understand the chain binding us. As long as we have no scientific measure to compare leagues, we have no choice but to assume all leagues are essentially equal. No particular one is higher than any other. This is like saying 1 dollar is always equal to 1 euro. Silly? Perhaps, but again, what other choice have we if we don’t know the exchange rate? Either we live with it, or we dump the Ballon d’Or. That is our choice.
Note: And, still, we need some limit set by common sense. It is crazy to assume the Frauenbundesliga and a league of Tokyo high schools are equal, simply because there is no “exchange rate” between them. Perhaps we must limit our specimen to leagues among which players can transfer without finding themselves suddenly in an “another dimension.”
Even if we assume there is no overall difference in the level among these leagues, I hope we can yet discern one from another by its propensity—some leagues are good at attacking while others are better in defense. Attack-oriented leagues have relatively good attackers but poor defenders. Defense-oriented leagues have it the other way around. The general rule is, of course, a score in a league with good defenders is more valued for the same reason a score against Arsenal is more valued than a score against Doncaster. I already mentioned the possibility that FAWSL and Nadeshiko are more defense-oriented than Frauenbundesliga. Let us delve into this problem.
Naturally, the most reasonable measure to judge how easy or how difficult to score in a certain league is the league’s score count itself. The aggregate score counts in each league are:
D1 Feminine: 530
This time again, these counts are to be compensated by number of games. This time, however, the number of games means the overall number of games in the league, not how many games each team plays. Therefore, each league’s number of goals per game is:
D1 Feminine: 530 ÷ (22 × 12 ÷ 2) = 4.02
Frauenbundesliga: 451 ÷ (22 × 12 ÷ 2) = 3.42
FAWSL: 185 ÷ (14 × 8 ÷ 2) = 3.30
Damallsvenskan: 426 ÷ (22 × 12 ÷ 2) = 3.23
Nadeshiko: 259 ÷ (18 × 10 ÷ 2) = 2.88
NWSL: 238 ÷ (22 × 8 ÷ 2) = 2.71
Therefore, if we define the value of a goal in the least scoring NWSL as 1.00, the value of a goal in other leagues are:
NWSL: 2.71/2.71 = 1.00
Nadeshiko: 2.71/2.88 = 0.94
Damallsvenskan: 2.71/3.23 = 0.84
FAWSL: 2.71/3.30 = 0.82
Frauenbundesliga: 2.71/3.42 = 0.79
D1 Feminine: 2.71/4.02 = 0.67
We now recalculate our “value-compensated” goals of our players applying these factors.
#1: Yuki Ogimi, 7.94 × 0.79 = 6.27
#2: Conny Pohlers, 6.97 × 0.79 = 5.51
#3: Mandy Islacker, 6.77 × 0.79 = 5.35
#1: Lotta Schelin, 7.56 × 0.67 = 5.07
#2: Hoda Lattaf, 3.25 × 0.67 = 2.18
#3: Camille Abily, 2.57 × 0.67 = 1.72
#4: Eugenie Le Sommer, 2.38 × 0.67 = 1.59
#1: Lauren Cheney, 9.76 × 1.00 = 9.76
#2: Sydney Leroux, 9.19 × 1.00 = 9.19
#3: Abby Wambach, 8.07 × 1.00 = 8.07
#1: Toni Duggan, 8.00 × 0.82 = 6.56
#2: Natasha Dowie, 7.94 × 0.82 = 6.51
#3: Nicole Rolser, 7.58 × 0.82 = 6.22
#1: Christen Press, 8.39 × 0.84 = 7.05
#2: Pernille Mosegaard-Harder, 7.36 × 0.84 = 6.18
#3: Anja Mittag, 6.17 × 0.84 = 5.18
#1: Beverly Goebel-Yanez, 14.43 × 0.94 = 13.56
#2: Nahomi Kawasumi, 7.32 × 0.94 = 6.88
#3: Ji Soyun, 5.54 × 0.94 = 5.21
Now we have reached our conclusion.
We have counted in the two factors—(1) influence of a player’s team and (2) influence of her league—to know who is the “best” goal-scorer in the world of women’s soccer, and end up with the top five as given below.
#1: Beverly Goebel-Yanez 13.56
#2: Lauren Cheney 9.76
#3: Sydney Leroux 9.19
#4: Abby Wambach 8.07
#5: Christen Press 7.05
All Americans. No wonder USA is the absolute #1 in the FIFA rankings. (And what a crime their NT keeps ignoring Bev!)
Note: In fact, there may be some between Wambach and Press whom we might find if we check with more players, because I listed only 3 from NWSL.
Let me add one thing. I guess some people are upset or even outraged to see particular players they love are low-rated here by cold, heartless numbers. They might even find my assessment provocative. Please keep in mind, though, electing a global MVP (or, maybe, any MVP) in a sport like football is provocative in the first place. Either we do this by Mr. Spock-style cold logic or we don’t do it at all.
Thank you, Dennis Barton, for proofreading!