Obviously since MLS soccer is so unpredictable the amount of points a player scores can vary wildly from week to week. An analysis of the data from the 2015, 2016, and 2017 MLS seasons reveals the the amount of fantasy points a MLS player typically gets in certain situations. The data is remarkably consistant and does not vary much from season to season.
There are 5 main factors to predict points:
- Player’s Average PPG
- Player’s Position
- Opponent’s Offense
- Opponent’s Defense
- Where the game is played (Home or Away)
Supplemental criteria:
- Whether the player is going to take more set pieces than normal
(See these articles on the value of penalty kick taking and corner kick taking) - Whether the player is expected to start the game
- Whether the player and teams involved have played a large enough sample size of games
(for the beginning of the season, use last season’s results if available)
Notably, form has not been factored in. Sometimes players getting hot is just a result of good matchups. Also the rising points per game a player receives from playing well ends up accounting for good form. The model could be slightly improved to account for this.
Forward Point Projection
The data shows that forwards are the least consistent and most matchup dependent position. The best forwards really take advantage of home games against teams with bad defenses. However, forwards in bad matchups do not tend to do well, since the points they get depend largely on scoring or assisting.
| Player Average | Location / Home | Location / Away | Opp. Defense / 2gcpg | Opp. Defense / 1.5gcpg | Opp. Defense / 1gcpg |
|---|---|---|---|---|---|
| FW + 7.5 – | 1.25 | -1.25 | 1.5 | 0 | -1.5 |
| FW 7.0 – | 1.25 | -1.25 | 1.25 | 0 | -1.25 |
| FW 6.5 – | 1 | -1 | 1.25 | 0 | -1.25 |
| FW 6.0 – | 1 | -1 | 1 | 0 | -1 |
| FW 5.5 – | 0.75 | -0.75 | 1 | 0 | -1 |
| FW 5.0 – | 0.75 | -0.75 | 1 | 0 | -1 |
| FW 4.5 – | 0.5 | -0.5 | 1 | 0 | -1 |
| FW 4.0 – | 0.5 | -0.5 | 1 | 0 | -1 |
Offensive Midfielder Point Projection
Offensive midfielders are either wingers or play the number 10 position. Compared to forwards, the data shows that they are slightly less dependent on matchups, although matchups still matter a lot. Since offensive midfielders tend to get more bonus points than forwards, scoring or assisting is not as crucial to their total score.
| Player Average | Location / Home | Location / Away | Opp. Defense / 2gcpg | Opp. Defense / 1.5gcpg | Opp. Defense / 1gcpg |
|---|---|---|---|---|---|
| AM + 7.5 – | 1 | -1 | 1.75 | 0 | -1.75 |
| AM 7.0 – | 1 | -1 | 1.5 | 0 | -1.5 |
| AM 6.5 – | 1 | -1 | 1.25 | 0 | -1.25 |
| AM 6.0 – | 1 | -1 | 1 | 0 | -1 |
| AM 5.5 – | 0.75 | -0.75 | 1 | 0 | -1 |
| AM 5.0 – | 0.75 | -0.75 | 0.75 | 0 | -0.75 |
| AM 4.5 – | 0.75 | -0.75 | 0.75 | 0 | -0.75 |
| AM 4.0 – | 0.75 | -0.75 | 0.75 | 0 | -0.75 |
Defensive Midfielder Point Projection
Defensive midfielders play either the number 8 or number 6 role on a team. While they do not score or assist often, they can accumulate a lot of defensive bonus points and passing bonus points. As a result, their upside is not as high as forwards or offensive midfielders, but they can be extremely consistent in the amount of points they generate regardles of the matchup. For more detail, please see: Why defensive midfielders are the least matchup dependent fantasy plays
| Player Average | Location / Home | Location / Away | Opp. Defense / 2gcpg | Opp. Defense / 1.5gcpg | Opp. Defense / 1gcpg |
|---|---|---|---|---|---|
| 6M + 6.0 – | 0.25 | -0.25 | 0.25 | 0 | -0.25 |
| 6M 5.5 – | 0.25 | -0.25 | 0.25 | 0 | -0.25 |
| 6M 5.0 – | 0.25 | -0.25 | 0.25 | 0 | -0.25 |
| 6M 4.5 – | 0.25 | -0.25 | 0.5 | 0 | -0.5 |
| 6M 4.0 – | 0.25 | -0.25 | 0.75 | 0 | -0.75 |
Fullback Point Projection
The data shows that fullback point production is dependent on three main factors: the player’s ppg average, the opponent’s offense, and whether the game is home or away. Notably, the opponent’s defense is not a big factor. The data over the years shows that even when fullbacks play the worst defenses in the league, they only score .25 points more per game on average. For more detail, please see: Why picking defenders against weak defenses is a fallacy
| Player Average | Location / Home | Location / Away | Opp. Offense / 1gpg | Opp. Offense / 1.5gpg | Opp. Offense / 2gpg | Opp. Defense / 2gcpg | Opp. Defense / 1.5gcpg | Opp. Defense / 1gcpg |
|---|---|---|---|---|---|---|---|---|
| FB + 6.0 – | 1.25 | -1.25 | 1.25 | 0 | -1.25 | 0.25 | 0 | -0.25 |
| FB 5.5 – | 1 | -1 | 1.25 | 0 | -1.25 | 0.25 | 0 | -0.25 |
| FB 5.0 – | 0.75 | -0.75 | 1.25 | 0 | -1.25 | 0.25 | 0 | -0.25 |
| FB 4.5 – | 0.75 | -0.75 | 1 | 0 | -1 | 0.25 | 0 | -0.25 |
| FB 4.0 – | 0.75 | -0.75 | 0.75 | 0 | -0.75 | 0.25 | 0 | -0.25 |
Centerback Point Projection
Top centerbacks tend to be more steady point producers than top fullbacks. They typically get more defensive bonus points, especially in away games, when they are defending more. Also, centerbacks on teams with a lot of possession can get a lot of passing bonus points, while fullbacks rarely do.
| Player Average | Location / Home | Location / Away | Opp. Offense / 1gpg | Opp. Offense / 1.5gpg | Opp. Offense / 2gpg | Opp. Defense / 2gcpg | Opp. Defense / 1.5gcpg | Opp. Defense / 1gcpg |
|---|---|---|---|---|---|---|---|---|
| CB + 6.0 – | 0.75 | -0.75 | 0.5 | 0 | -0.5 | 0.5 | 0 | -0.5 |
| CB 5.5 – | 0.75 | -0.75 | 0.75 | 0 | -0.75 | 0.25 | 0 | -0.25 |
| CB 5.0 – | 0.75 | -0.75 | 0.75 | 0 | -0.75 | 0.25 | 0 | -0.25 |
| CB 4.5 – | 0.75 | -0.75 | 0.75 | 0 | -0.75 | 0 | 0 | 0 |
| CB 4.0 – | 0.75 | -0.75 | 0.75 | 0 | -0.75 | 0 | 0 | 0 |
Goalkeeper Point Projection
The data on goalkeepers is interesting. The matchup does not matter as much as in other positions, especially for the top goalkeepers. Away goalkeepers tend to get more defensive actions, so that somewhat offsets their lower chances at getting a clean sheet. Furthermore the upside is lower for goalkeepers since they very rarely score or assist, so their variance is lower.
| Player Average | Location / Home | Location / Away | Opp. Offense / 1gpg | Opp. Offense / 1.5gpg | Opp. Offense / 2gpg |
|---|---|---|---|---|---|
| GK + 6.0 – | 0.25 | -0.25 | 0.25 | 0 | -0.25 |
| GK 5.5 – | 0.25 | -0.25 | 0.5 | 0 | -0.5 |
| GK 5.0 – | 0.5 | -0.5 | 0.5 | 0 | -0.5 |
| GK 4.5 – | 0.5 | -0.5 | 0.75 | 0 | -0.75 |
| GK 4.0 – | 0.5 | -0.5 | 0.75 | 0 | -0.75 |
Notes
- gpg = goal per game, gcpg = goals conceded per game
- There are gradual increases / decreases in points based on the strength of the opponent’s offense and defense. For example some teams concede 1.75 goals instead of 2 goals per game, therefore the point projection would be only 50% of the 2 goals per game projection.
- These numbers are estimates based on regression curves. There is a larger data set on players with lower averages, so those tend to be slightly more accurate.
- All the numbers are rounded to the nearest quarter point, since finer data is largely irrelavent. When considering two players within .25 points then it is probably just better to go with your gut feeling than these average numbers.
- Sometimes players are a hybrid of two different positions. For example, Jan Gregus and Jack Price are both defensive midfielders, but they have had a lot of responsibility on set pieces, so their projected production is somewhere in between an offensive midfielder and a defensive midfielder. Another example would be Julian Gressel and Brooks Lennon. Both players have been classified as defenders but often played in the attack for their teams, so the math on these players is a little more complex.
- The players with the highest expected points projections each week tend to be highly correlated with the popular picks each week. This shows you do not necessarily need to do the math to know which players are best to add to your team.
