Franchise Mode Simulation & AI Issues (Detailed Feedback for NHL 27)

Simulation

The current simulation does not produce realistic statistical outputs.

This video by 2bc highlights several of the issues with production totals across different scoring and shot frequency settings:

https://youtu.be/WgNTHOX-Uys?si=II_GlWHWtdtJufp7

• Elite forwards such as Connor McDavid, Nathan MacKinnon, and Nikita Kucherov often finish with only 80–90 points. In reality, these players are capable of producing 100–130 point seasons (and up to 150 in McDavid’s case) over a full 82-game season.
• Assist totals are too low due to a lack of secondary assists:
• 80-assist seasons for forwards rarely occur
• This lowers overall point production
• Star players should elevate teammates’ scoring while still maintaining 30–40 goal seasons
• Mid-tier defensemen (84–86 overall) will sometimes randomly produce 40–50 assists, which feels unbalanced relative to elite forwards.
• Shot volume is too high, which negatively impacts realism:
• This contributes to inflated goal totals but poor shooting percentages
• It also impacts goalie stats and overall balance
• Forwards do not block enough shots:
• Players like Elias Pettersson and Auston Matthews already have high block totals in real life
• In-game, forwards rarely exceed ~25 blocked shots in a full season
• Increasing the shot blocking attribute does not seem to affect this
• Power play percentages are too low across the league:
• Attempt volume is reasonable, but conversion rates are not
• More teams should realistically exceed a 20% success rate
• (Minor suggestions):
• Show assists during “live sim” in the event tab
• Show shot type during live sim
• Reduce unrealistic goal types (e.g., low-percentage shots from the boards going in too often)
• More net front goals off rebounds and tip plays

Attributes and X-Factors

Attributes are generally inflated, reducing the gap between player tiers.

• Too many players fall in the 85–90 overall range
• There should be a clearer distinction between elite players and middle-six forwards
• Player weaknesses are not accurately reflected:
• For example, Quinn Hughes has relatively high physical attributes despite physicality being a known limitation of his game. 
• Offensive and defensive awareness appear to be weighted too heavily compared to skating and physical attributes
• This also makes 5v5 practice the most optimal preseason focus choice
• Shot blocking ratings do not appear to impact forwards, even when set extremely high
• Suggested addition:
• A Hustle/Motor attribute to better represent high-energy players
• Useful for distinguishing effort level, forechecking, and defensive engagement
• Could also differentiate highly skilled but lower-motor players like William Nylander

This video also demonstrates improved attribute tuning compared to the default roster:

https://youtu.be/it4Bj6zly2o?si=5ymt0OZ48UboevHA

• X-Factors should have a greater impact on simulation:
• Only a few (e.g., tape-to-tape, quick release) noticeably affect results
• Others (e.g., backhand beauty, big rig) seem to have little to no impact outside gameplay
•  
• Example:
• Leon Draisaitl has quick release, tape to tape and backhand beauty. Only quick release and tape to tape seem to inflate production which makes the last slot in backhand beauty a waste during simulation.
• If attributes remain inflated:
• X-Factors should be the primary way to differentiate elite players
• Allow more than 3 X-Factors per player in offline modes
• A good skater like McDavid having wheels and elite edges only allows room to have 1 other skill based x-factor

Coach / GM AI

AI decision-making often leads to unrealistic player usage and roster choices.

• Coaches over-prioritize scheme fit:
  
  • Example: Minnesota Wild deploying Quinn Hughes on the second pairing while Jonas Brodin plays on the first
  • This results in unrealistic ice time distribution
• AI teams may overvalue X-Factors and bench 85 overall young rookie over an 80 overall with x-factors
• Coach attributes have too much impact on team performance:
  
  • Weak teams can overperform unrealistically with high-rated coaches
  • Coach attributes should be editable in roster customization
• Trade logic needs improvement:
  
  • Contenders should not act as sellers
  • Non-playoff teams with poor records should not act as buyers
  • Division standings should influence team behavior more strongly

Contracts / Player Conversations

• Improve menu navigation:
• Allow switching between players using triggers instead of exiting each conversation
• Contract logic can be inconsistent:
• Example: winning a Stanley Cup, then having players refuse extensions due to “wanting to play for a contender”
• Player retirement age:
• Players often play too long (into their 40s)
• Retirement age should be lowered for realism

This doesn't cover everything but are the biggest issues that come to my mind.