|
Quote: |
|
|
|
|
Originally Posted by Yazan Gable |
|
|
|
|
|
|
|
|
I thought it was obvious how to improve free agency: hide their exact ratings of free agents and instead provide a range like in FIFA. In fact, why not apply this to all rosters that are not your own? In this way you now have to scout for players mid-week (an addition to gameplanning) and help you discern whether a player was only successful within the system, circumstance, or is genuinely good. Brock Osweiler has a stupid contract because of the perception of him being a decent quarterback. In Madden as it is now we would know he's garbage and not sign him, but if we have a range and have to devote resources into getting a clear understanding of his abilities (reducing the range of the ratings, perhaps with the best position coaches straight up giving you the exact numbers) to inform your decision. The largeness of the ranges these players would have would be determined by team success, the win rate while this player was starting (even though that obviously wouldn't necessarily mean much), and their statistics from the year. Another way to influence this is your suggestion of the team's identity at the time (win now vs. rebuilding) where a team that perhaps is more desperate to contend or feels they are one player away are willing to take a flyer on a player with little data to show how good they really are (Brock Osweiler, most likely bad; Pernell McPhee, turned out to be very good.)
I also think a range is much better than combine numbers (which would quickly become null with aging players or injuries that would sap their abilities) or using letter grades.
I also think that using real player assets or whatever like they do for 2K is worse than using nice real photos: always annoyed when I traded for a player and suddenly I had this glass-eyed, terribly-emoted image of the player while if they stayed on their original team I'd have these cool pictures in their uniforms looking nice. Same thing with the videos for QBs, I'd rather they expand it to as many QBs who they could get videos for and save the real assets for the drafted players.
|
|
|
|
|
|
Yeah, when I was looking at free agency I wanted to focus more on the process more than anything. You can hide or mask the ratings but you'd still be able to get whoever you want without much trouble which was the main thing I wanted to hit on. I totally agree with masking the ratings for free agency but I look at that as more of something that is all encompassing rather than limited to just free agency which I why I didn't mention it.
But within the realm of masked ratings I really like the idea of becoming more familiar with players the longer they've been on your team and/or been in the league. I think having different levels of familiarity would be something that would work extremely well within the masked ratings scope. Players that you've had on your team for long periods of time or veterans could have +/- letter grades giving a better picture of their true ratings. Rookies and young players that your team is unfamiliar with would have the widest possible ranges for letter grades that would narrow the longer they were on your roster. To use your example, while the Broncos might see Brock Osweiler as a C- overall, another team, like the Texans might see him as a B because they don't know as much about him yet. You could even go a step further with this and let's say you hire a new coach or position coach from another team, you also acquire the knowledge that the coach had of his former teams players and/or position groups.
As for the combine numbers, I like the idea of ranges for everything except physical attributes. Physical attributes are something that I think should be a bit more finite and I think using combine numbers allows for that while still maintaining the ability to use wide ranges for other attributes that can be based more on perception. What I mean by this is taking something like speed you may have a player that is an A letter grade but if that's something that ranges from 90 to 99 then you're not sure whether your guy is just typically fast or elite level fast. If he has a 4.29 40 time, you know he's elite level fast but you still don't necessarily know whether it's 99,98,97, etc. You raised a really good point that I hadn't even thought of though in regards to regression of physical attributes but I think if you provided something like this:
to the user in the form of a coaching report or more simply a regression history on their player card, I think combine numbers could still work because you'd be aware of a player losing speed, agility, etc. With hidden or perceived ratings you'd just be told that it's going down while if you were using exact you'd know exactly how much he regressed.