Review Team Seasonal Performance Based on Implied Odds & xG Data.

Tool Methodology Methodology

Post-Match Expectation (xG)
The xG data for this site is sourced from understat.com. xGoals and xAssists data is compiled from understat and a quality multiplier is applied to adjust the data based on a players historical finishing ability (normalised based on sample size). xCleanSheet probability is determined for each player while he was on the pitch from the xGoals data. Own goals are removed as they are deemed to be random events & penalties are adjusted for the risk of being missed/saved/scored.

Yellow and Red Cards are not adjusted. A background xAssists probability is added to reflect the possibility of an innocuos FPL assist that is not considered an assist by other systems. Bonus points are determined following a probabilistic simulation influenced by the xG data and historic analysis of the BPS system. The end result is an expected points total based on situation each player found himself in during the match. This is a very strong performance indicator.

Pre-Match Expectation (Odds)
Implied Odds data for each GW is used in this tool with margins removed for Clean Sheets, Player Goals, Cards and Team Goals. Projected Assists are deduced from historic xG data for each player and the implied odds predicted scorelines for each match.

Own Goals and Saves are also forecast on a player specific basis. Bonus points are determined following a probabilistic simulation influenced by the Implied Odds data, xG data and historic analysis of the BPS system. All of these factors are adjusted for the time each player spent on the pitch to reflect a fair expectation of the points they could earn in the GW.

Which is Indicator Most Useful?
Implied Odds points have been found to be extremely predictive of future FPL performance, more so than xG points- this is because it is impacted by the massive variance of football to a lesser degree. FPL pts are a weak indicator of future performance relative to the other two. Check out this link for a detailed analysis on this topic

How to Use this Tool How to Use this Tool
  1. Enter the Team Code you wish to investigate prior to hitting the submit button.
  2. Review the results. A higher FPL points score than expected score indicates over-performance/luck. As many managers follow similar selection patterns your luck percentile score is more informative and indicates your relative luck compared to other managers. 0-30% = underperforming, 30%-70% = normal, 70-100% = overperforming

Input Team Code, this can be found on your gameweek history page in the URL. e.g. https://fantasy.premierleague.com/a/entry/Team_Code/history
Last Complete GW