ya first round picks should be valued more across all leagues imo

It wouldn't give your Weights. Only your Overall. You can reach the same Overall in a variety of ways.

Depending on the player ratings that would make it easy for people to ball park what attributes you weigh most heavily.

Let me be more specific.

I don't want other teams to be able to see what my overall rating for a particular player is, unless I choose to disclose that information myself (which I do often).

The only thing that should matter to you anyway is how YOU value a player.

My suggestion to show the default AI weights (ratings) for players is just so we can have a better understanding of why the AI does some of the things it does (allows trade A, doesn't allow trade B, etc.)

a
Well, if everybody seems to recognize this as being self-evident, it is curious why this is not being fixed (?) Hmmm. In fact, i think i made this observation myself in another thread, though certainly not as eloquently as GrandadB did here.

To take it a little further:



Yes, i guess it stands to reason that if the trade values are fixed so that trades are actually fair for both teams, then there would be 1), no imbalances to exploit and therefore 2), no cheating would be possible and so 3), duplicate accounts would yield no advantage to anyone and be meaningless and pointless.

But wait!... wasn't that what the 'trade value' and 'trade meter' were implemented to do - impose relatively fair trades?

So let me get this straight, because the logic of this kind of thing escapes me...

...Duplicate accounts are disallowed in order to prevent imbalanced trades (considered "cheating"), but the trade value system actually does not ensure fair trades and actually virtually ensures the opposite (considering that it is human nature to exploit any advantage in one's favor) and so we arrive at a trading system that achieves the exact same result that preventing duplicate accounts was supposed to stop, but NOW it is OK because it is no longer considered "cheating" (?)...

And so we have accomplished... what, exactly?


The logic of all of this escapes me, i have to admit.


i would tend to think a much better way to prevent cheating would be to actually fix the trade values as mentioned above (so as to remove any incentive to ply unfair (now impossible) trades), and add a few minor restrictions to trades such as 1) limiting the number of players that can be traded for a pick, 2) limiting trades to players that are well above those which can be obtained from the FA pool, and 3) limiting how far in advance of their drafts picks can be traded and probably a few other pretty minor tweaks.

The self-evident logic that bogus duplicate accounts need to be policed to prevent unfair trades suggests that the present trading system does NOTHING to ensure fair trades and is, therefore, inherently broken and in need of fixing.

It does need some more work, but it is much better than it has been in the past.

There are a lot of changes that have been made that you are probably unaware of if you have only been here for 2 months. For example, you used to be able to trade picks up to 7 years in advance. Now it is only 3.

Also, some owners consider trades as being "unfair", "unbalanced" or "taking advantage of another user" simply because they wouldn't have made the trade themselves and think everyone should evaluate players and run their team the same way they do.

i'm sure the system is better than it was, but still needs work, as you say.

The ability to trade picks 7 years in advance is really ripe for exploitation simply because you cannot ensure that an owner is going to stay around to reap the consequences he has sown for himself. Things would be different if a GM had a real ($$$) vested interest in the future of his/her team.

********************************************************************************************************

A totally objective method, however, of evaluating the relative values of trade elements is the concept of using a "currency" known as a player-season.

In the real world this concept is relatively complicated and can be translated into an actual cash value and includes such factors as consistent performance level (essentially volatility), injury history, even drawing power (fan attendance, commercial endorsements, etc) among dozens of other variables and factors.

In a game such as this it can be evaluated very simply. A player-season is simply one season's play by one player. To make relative evaluation easy, the player is considered to be a top-level, #1 pick.

For the moment we ignore the ramping that occurs to a player's performance level in his first 2+ years and the drop off that occurs near the end and just consider the number of top level performance (what i call "stellar") seasons the assumed #1 pick is expected to generate. (These high-performance seasons are very valuable and due to cap restraints and availability there is a practical limit of how many of these any team can have.)

So using a very, very simplified evaluation method we can easily compare the value of trading older players, for instance, for #1 draft picks and get a result which is far more satisfactory than anything used in the game. The only input variable we need to use in our rough calculation is the expected longevity (number of player-seasons) our star #1 pick is going to generate for us.

Let's say, for example, a GM wants to trade an 8 year (top-level) RB for one, or several, future #1 draft picks. Let's assume, here, that all the potential players considered in this example will be expected to retire after 12 years, for simplicity. Now, our game tells us that this 8 year RB is worth 2 (possibly 3) future #1 picks. Does this make sense?

The RB will be expected to yield to the recipient 4 "stellar" player-seasons while the receiver of the two picks, after throwing out the first two 'maturation' seasons, receives an expected 20 in return. In some cases 3 #1 picks or 2 #1s and a #2 are considered fair value. 4 stellar "player-seasons" in return for 20-30 is somehow a fair trade?

OK, to be fair and a bit more accurate, there is a chance that the drafted players will be 'busts'. A bust here does not normally mean the player is worthless and has to be cut but instead that these players may turn out to only perform as well a second or third-rounders. Let's assume a percentage of about 30% 'busts' for first round draft picks (this is likely on the high side). Even with this inflated percentage and using the lower number of seasons above (20) then the recipient of the draft picks is still expected to gain 14 stellar player-seasons in return for 4.

This is not subjective opinion. It is not opinion that 20 is five times the value of 4, or that in our very conservative calculation, 14 is 3 and 1/2 times the value of 4. It is fact. It is fact that the giver of picks is giving MASSIVELY more than he is getting in return. Yet the game evaluates '20' as being equal to '4' somehow, more or less.

Obviously the trade-value system in the game is way out of whack and unfairly slanted towards players with 'mature' teams and older high-level players.

As a final, and anecdotal, point, i would just like to mention that in the five leagues in which i presently have a team, no top GM has ever offered a #1 or even a #2 pick for one of my players. Every single one has bombarded me with offers for one of their 10-14 year vets for 1 to 3 #1 picks the moment i joined the league. This was not because these guys are such kind-hearted, giving human beings.

This fact, by itself, is enough evidence of how lop-sided this situation is and i invite all newer GMs to compare and post here if their experience is any different than mine.

Doubtful.

The engine dictates who's good. Tighten what is successful and pair that with the trade values.....it way better, but a low cost qb turned rb that rushes for a 1000 yards but is a 153 on the trade scale is what is subjective.

He doest allow enough sacks or drop enough passes and is too fast. It's getting better, but check beta 87 top qbs.

The qb ratings they get is not in line with trade value. It's very subjective.

Oline / dline is broke and blitz ING players do not need pass rush.

It's the engine not reflecting the trade values. There is no point in changing them much more until the engine is fixed in some of the area I just mentioned. It touchy to say the least.

Familiarity is what should dictate some version of boosting trade value. Cause certain plays are better with a certain type of player . Oh yeah the playbook needs an upgrade also.

bump

How about introducing feedback into the valuation?

Every trade involves an older account and a younger account. Every time a trade is made, make every position or pick the older account trades for worth slightly more points, across all leagues. Likewise, make every item the younger account trades for worth less points, across all leagues.

For example :

Team A (older account) trades:
10th year RG (200 points)
8th year SS (300 points)
12th year QB (100 points)
6th year DE (400 points)

Team B (newer account) trades:
current 1st round pick (520 points)
next year first round pick (480 points)
Rookie CB (800 points)

modifies all 10th year RGs by -.12
modifies all 8th year SS by -.13
modifies all 12th year QB by -.11
modifies all 6th year DE by -.14
modifies all current 1st round by +.152
modifies all next year 1st round by +.148
modifies all rookie CB by +.18

It gradually adjusts every valuation towards fair market values. No particular trades would have any great effect. Gaming the system or cheating would still be possible, but it would take a large scale systematic approach (fighting the tide), would result in a smaller payoff, and become more obvious.

This is just a first attempt, to get the idea out there. The algorithm could be something much more elegant and better adjusted to the actual volume of trading. It could be set to account for a player's best position. It could be fine-tuned with another algorithm to adjust valuations based on attributes. . .

This way is less work for the developer than a top-down solution. Make the system self-correcting, based on the behavior of the entire community.