Contract Extensions determined by OVR, not rank

[KW77]

Follow up from our previous conversation, believe it needs its own thread.

We're talking over the contracts issue on our CSFL discord. It seems to me that players' extension asking prices tied directly to the value of their OVR, rather than looking at their place in the league.

I created a league with default rosters, and our financial constraints ($140M cap, same mins and maxes per position) and realized that our OVR scales are significantly different than the PF21 defaults. Default has 250 80+ OVR players, our league is around 600 and will rise once more seasons of position skill training happens.

In both leagues, an 81 OVR DT is asking for a salary in the $6-7M range. This 81 OVR DT in the default file is the top DT in the league. There are 24 DTs over 81 OVR in our league, going all the way up to 100 OVR (that guy wants $20M+ to extend). This is why you are not seeing a problem with extensions the default rosters. Because the best DT in your league is asking $7M to extend rather than $20M.

There are examples like that in every position, so more extreme than others.

My rough solution: Do exactly what you do with prospect scouting grades - tie extension asking prices to the league's range of ratings at each position, rather than their OVR. That's obviously not a quick fix because that has to be fine tuned otherwise it would be an even bigger mess. So I'm not demanding that to be in the very next patch (would be cool), but I would be quite disappointed if something like that's not in next year's game. Makes it difficult to run a realistic multiplayer league otherwise.

[brooks_piggott]

The requested salary is already taking a lot of that into account... personality, coaching, team performance, age, experience, individual ratings, positional adjustments (K and P make less than QB, etc.), traits, and then compares to peers based on similar attributes. OVR is mainly used to obey the salary min points.

This is the same system that has been in place since DDSPF16... we're constantly tweaking it to add more effects, but the overall design hasn't changed (especially not for DDSPF21)

Best I can do at this point is look at allowing leagues to set max contracts for individual positions. That way if you only want DT's going up to 7M and not asking for 20M you can. The positions do scale well for 10M max salaries, but obviously don't scale as well with 20 or 30 or 50 mil max contract options.

[zac]

It still seems that ratings in relation to league averages weigh very little and the system is just looking at ratings more in a vacuum.

Below are two players asking for similar contacts. Both players are fairly similar if looking at ratings in a vacuum. Not so similar when comparing their league rankings. Player 1 is from our CSFL online league and player two from the default league/roster created in DDSPF 21.

Player 1: Rating (Lg Rnk) - from CSFL league
Pos: CB
Age: 25
Speed: 80 (96th)
Agi: 79 (95th)
Int: 54 (146th)
Str: 56 (18th)
OVR: 81 (55th)
Asking Price: $5.6 Million

Player 2: from the default league
Pos: CB
Age: 25
Speed: 88 (22nd)
Agi: 76 (65th)
Int: 72(10th)
Str: 53 (7th)
OVR: 80 (4th)
Asking Price: $5.2 Million

One guy ranks at or near the top 20 in almost every category and the other ranks closer to 100th in many categories yet they both are asking the same price. It feels like the system weights ratings more in a vacuum and not as much how they rank in relation to the league. Maybe this isn't the case, but it feels like the case as they only explanation I can currently see for why our league's extensions are so high many times. I understand this is just one example and not a good sample size at all, but this example is repeated again and again when comparing our leagues' extensions to the default roster extension request.

I know you say that ratings in comparison to a player's peers matter but it doesn't always feel like. Maybe there's another explanation for this though.

[brooks_piggott]

I'm not sure what this shows. Default leagues have different contract values than your online league does so comparing the salary demands won't work.

On top of that there isn't one single number they're asking for... their asking price is just a place to start negotiations... it may take a lot more than that to sign, or it may be a bit less depending on the personality traits of the players are, or the current roster status of the team, what their current salary is, and so on and so forth.

In general though we compare attribute numbers to find comparable players... we don't do calculations based on rankings of individual attributes.

[zac]

Thanks for the reply. Just now sure how two leagues set up with the same financial settings get a top 10 CB to have the same demands as a barely top 60 CB.

Unless there's another explanation, it still seems that players seem to be paid based on their ratings (among other factors, yes). But not how their ratings rank in the league. A player should not be paid based on ratings, but how their ratings rank in the league in my opinion. Ratings are all relative.

[AKH]

I will add this:

What players want is entensively based on what similar players are getting paid. One thing - if we know anything at all - is the in online leagues CBs get paid.

[AKH]

and to clarify - what matters mostly is their ratings _not_ their ranks. SO a guy will look at "what is a guy with 78-82 speed getting paid, not ´what is a guy with 100 tackles' getting paid.

[brooks_piggott]

Thanks for the reply. Just now sure how two leagues set up with the same financial settings get a top 10 CB to have the same demands as a barely top 60 CB.

Unless there's another explanation, it still seems that players seem to be paid based on their ratings (among other factors, yes). But not how their ratings rank in the league. A player should not be paid based on ratings, but how their ratings rank in the league in my opinion. Ratings are all relative.

We don't do stack ranking of attributes for salary purposes, nor are there any plans to make this kind of change. I appreciate the thought, but I'm happy with the multiple systems and vectors we have currently in place.

[zac]

Thanks for reviewing. To clarify, when I say rank, I don’t mean stats like tackles. I just mean the having a contract system that does not looking at ratings by themselves but looking at where there ratings rank in the league. So not paying a 78-82 speed guy x amount but paying a guy who ranks 23rd in speed ”x” amount In one league, a 78-82 speed guy might be the fastest CB and another league he could be the 100th fastest guy even though the ratings are the same. Because they rank so different in speed, there demands should be widely different. That doesn’t seem to be the case now in comparing leagues with different distribution of ratings.

Right now, the system seems to pay “78-82” speed guys the same no matter the league even if a 78-82 speed guys ranks top 10 in one league and outside the top 40 in another.

But I’ll leave it at that unless I can get more data on the matter to provide a more comprehensive summary. I still hopeful maybe this is something to look at seeing if improvements are possible for the next version. As of right now, guys sign for much less in free agency than they ask for in extensions. The example above isn’t specific to CB at all.


At the very least, I’d rather have the current system that guys ask for too much than have guys ask for too little leading to teams never losing talent. Thanks for the discussion though!

[AKH]

again to clarify: its not fixed brackets like 78-82. The player looks at his own ratings and says what are people with similar ratings getting paid (and its not just speed he looks at).

Anyhow, we have understood the point, and will look at extension negotiations when we get the time.