In P75, I drafted a low Accuracy QB because he had high vol and decent other attributes. As a result, he "boomed" except for his Accuracy, which stayed low. The DEFAULT WEIGHTS currently put him at 61/80: https://private75.myfootballnow.com/player/6831

In two preseason games, he has gone 2 / 51 passing: TWO COMPLETIONS and 4% COMPLETION PERCENTAGE.

Why does this really matter?

In game 1, I used a set of plays that he had no play knowledge of, and he completed ZERO passes in 25 attempts.

In game 2, I used only passing plays he had 90+ play knowledge, and he completed ... 2 passes in 26 attempts.

We all know the Default Weights are not accurate, but don't you think that an 80 rated player in the Default Weights should go better than 2 of 51 in two preseason games. An 80 rated player should be a 5-10 year starter on most teams.

1) It is time to change the Default Weights so they are ACTUALLY MEANINGFUL to beginning players (and so I and others don't have to waste time anymore coming up with weight sets that beginning players can use).

2) We MUST have minimum attribute ratings in our weight scales that pre-filter players out. If we need the minimum Passing Accuracy for QBs to be 80+, for instance, we shouldn't see any QBs with an Accuracy rating of < 80. And they SHOULD NOT be rated with a default rating of 80.

Defaults drive trade scores and contract logic. The ai would benefit from this would as well, but what would the new defaults be? I know I would up speed.

Each update should carry a change to defaults as well.

As far as I can tell, the default weights for QBs should be Accuracy 100 and a max of 20-25 for the next most important attribute.

In general, the problem with weights is that weights are percentile (linear) and attributes are now parabolic. It's really hard to say how important INT is compared to Accuracy for instance. If I want Intelligence to function 90% as effective as Accuracy, then linearly along the x axis INT would be ~(-2.25, -10) from the vertex. Does that mean that Accuracy is 2.25x as important as INT?

It's been awhile since I've had to do parabolic equations.

No RB,WR,LB's or DB's with less than 75 speed.
No OL or DL with less than 75 strength and under 275 lbs

Just for starters.

setherick, you know far more about the weights and gameplans than I probably ever will, but I worry your post will be quite misleading to the average player.
First, you don't have an 80-rated QB, you have one currently rated 61. And as far as I can tell a 61-rated QB wouldn't start on any team. That doesnt' mean he should go 2 for 51 either, which leads me to my second point...
Second, how much of your QBs poor performance is really attributable to issues with version 0.4.4 and not his ratings?

I don't understand this point. His rating is meaningless based on changes that were made in 0.4.4.

In 0.4.4, accuracy calculations are parabolic, which means that Passing Accuracy is exponentially more important than any other attribute. The problem is that the weight scale is percentile based, and essentially linear. If you want an attribute to be twice as important as another one, you set the slider to 50.

So the default weights set the overall on that QB to 80, but really, he's more like 0 since he'll never complete a pass because of his accuracy.