A Player Development Tell

[Ben E Lou]

Originally Posted by JMO

In my recent draft in the Imperial League, I had 5 picks. None in the first round. 3 of those picks had over 40 pop(at least now ten weeks into the season they do..none of them have produced enough to signifcantly effect their pop). I picked the first two on pure combine/% dev and the last one because i liked his strength and endurance bars and I had seen guards like that blossom before. He has 72 pop now although he hasn't played a snap except on special teams. I had no knowledge of the popularity tell before and i only use apps such as change tracker and the draft analyser.

Moral of the story..it would be nigh on impossible to tell who was using this tell as it will coincide way too often with good combine/creeper guys. Be careful before pointing fingers.

This is precisely my concern about not making the list public--especially when it comes to all of those little bar tells like the one you mention. It seems like it's just a more sophisticated/complex version of the old FOF2K1 redliners--different for each position, and sometimes multiple tell-tale patterns per position, but people who draft a lot will begin to see these patterns and use them to their advantage. And in the meantime, the guys who don't see the patterns will just see "JMO drafted a guard in a later round who had 72 popularity," and be suspicious of you. It's very easy for any league member to "check up" on their other fellow owners, and suspicions will arise. That's bad for leagues.

[gstelmack]

Originally Posted by Dutch

I'd be surprised that was the case. Everybody, including Jim, recognize that 3rd party utilities have helped extend the life of this product.

That may be true, but the point still remains that the utility is exposing something that the game does not, and that was NEVER the intent.

[gstelmack]

Originally Posted by Vinatieri for Prez

4. gst, that would be great if you could expose yearly HOF votes. It would add some nice immersion, and I see no harm in it. Pretty please.

Now I'm sorry I let that slip. Let me be clear: I am not talking about info that is not already in FOF. I am talking about info that is exposed in the data file before it is exposed in FOF. When a guy retires, FOF runs the HoF calculations immediately, then stores them for later use. When the guy is finally eligible, those results are exposed. If I let these through in Interrogator early, all you'd get is an early HoF list. There is NOTHING in the file that is not already shown on the HoF screen, or at least nothing I've found. We're talking about a timing issue, that's all. And I like having to wait until the guy should be eligible to find out.

So, no

[cuervo72]

Oh...so it doesn't store vote totals for guys who don't get in?

[gstelmack]

Originally Posted by cuervo72

Oh...so it doesn't store vote totals for guys who don't get in?

Not any place that I have found. Which is not the same as saying it doesn't, but the point is I don't have it to show.

[ShaneTheMaster]

Originally Posted by Ben E Lou

The highlighted portion has been an interesting sidelight to this discussion for me. It makes complete sense to me that popularity should correlate with talent. Are there any guys in the NFL who we'd categorize as Pop>90 who haven't been at least very good players at some point in their careers? Maybe one or two stray guys who do a bunch in the community might have localize popularity, but FOF doesn't claim to try to model that sort of thing.

I think what you saying is true for veterans - but not necessarily true for rookies. Ryan Leaf would have been very popular his rookie season, but in year 2 I would say he wasn't. Rookies are such question marks that the fans popularity rating should not directly correlate to how well a player will eventually be.

If anything, fan popularity should correlate to the PERCEIVED (OR MASKED) value of a player - not his HIDDEN potential.

The best solution by far is for FOF to be fixed so that this direct correlation does not exist, or it has some other factors put into the popularity calculation.

[Ben E Lou]

Originally Posted by ShaneTheMaster

I think what you saying is true for veterans - but not necessarily true for rookies. Ryan Leaf would have been very popular his rookie season, but in year 2 I would say he wasn't.

Players change pop with playing time and performance....

Rookies are such question marks that the fans popularity rating should not directly correlate to how well a player will eventually be.

If anything, fan popularity should correlate to the PERCEIVED (OR MASKED) value of a player - not his HIDDEN potential.

Terrell Davis. Tom Brady. They're the exceptions, not the norm. Guys like them are represented by the Volatility Stick Of Life in FOF. Ryan Leaf? The Volatility Stick Of Death. Massive obvious combine stud creepers don't really represent any type of NFL player, if you think about it. NFL Scouts know what they're getting in the vast majority of cases. Masking is a mechanism that makes FOF a more challenging computer game. (I suppose one could call it a proxy for a coaching/scouting staff who works overtime to understand the talents of the players they have, or that creeping is a proxy for seeing the player in practice and watching more film, but I don't see it that deeply.) Point being, when an obvious superstar with monster combines shows up in FOF with mediocre bars, he's a star player, and I'm of the opinion that his popularity should reflect that. He's a guy who was a star player in college and the fans are excited about having him.

The best solution by far is for FOF to be fixed so that this direct correlation does not exist, or it has some other factors put into the popularity calculation.

Or hide popularity with the real player ratings in encrypted places, only revealing it after the draft. *shurg*

[ShaneTheMaster]

Originally Posted by Ben E Lou

Player change pop with playing time and performance....

Point being, when an obvious superstar with monster combines shows up in FOF with mediocre bars, he's a star player, and I'm of the opinion that his popularity should reflect that.

Yes, I agree, but not until the fans actually see him display it on the field and he proves himself - not before he is even drafted. The fans are basically looking ahead in time 2 or 3 seasons down the road - giving the user an indicator he should not have.

By the game's logic, Marques Colston, Dan Fouts, Sammy Knight would all have a fan popularity of 80-90, before they were drafted. Doesn't make sense. In real life, The fans have no way of knowing who is going to boom ahead of time. They may prefer players that were winners in college or who played for the hometown college team.

I can't completely blame Jim because he probably didn't intend on that number being available before the draft. But I can't blame Greg either, because there is no way to know of the weird correlation. I don't think there is anything Greg can do, but I think there is something Jim can do.

[Celeval]

Originally Posted by ShaneTheMaster

By the game's logic, Marques Colston, Dan Fouts, Sammy Knight would all have a fan popularity of 80-90, before they were drafted. Doesn't make sense. In real life, The fans have no way of knowing who is going to boom ahead of time. They may prefer players that were winners in college or who played for the hometown college team.

Not necessarily. The game has plenty of talented players who aren't popular at the outset.

And the key with the hometown college team is to realize that Popularity is league-wide (i.e. it doesn't change based on team, via trade or whatever). To use an example SkyDog and I talked through, Thomas Brown may be a very popular Falcon (since he grew up in Tucker and played for UGA), but RaiderFan couldn't care less.

[Ben E Lou]

Originally Posted by ShaneTheMaster

Yes, I agree, but not until the fans actually see him display it on the field and he proves himself - not before he is even drafted. The fans are basically looking ahead in time 2 or 3 seasons down the road - giving the user an indicator he should not have.

We're not talking about a ton of players here with high pop. I remember exactly where I was when I heard the Falcons had acquired the 1(1) and the rights to Vick. The Falcons couldn't sell out the Dome in a 14-2 season, but they sold it out for the first preseason game he played. I scrambled to watch Herschel's first pro games ever. I couldn't WAIT to see Deion in the NFL. I would argue that these guys were wildly popular before they'd done a dadgum thing on the field. The fans wouldn't necessarily have said it the way we say it in FOF...but they were excited about these guys' potential ratings, not their current.

By the game's logic, Marques Colston, Dan Fouts, Sammy Knight would all have a fan popularity of 80-90, before they were drafted. Doesn't make sense. In real life, The fans have no way of knowing who is going to boom ahead of time. They may prefer players that were winners in college or who played for the hometown college team.

All players above x pop are good players. Not all good players are above x pop. There are significant numbers of quality players who do not have high pop in FOF.

I can't completely blame Jim because he probably didn't intend on that number being available before the draft. But I can't blame Greg either, because there is no way to know of the weird correlation. I don't think there is anything Greg can do, but I think there is something Jim can do.

Again, it's simply not a weird correlation.

[ShaneTheMaster]

Originally Posted by Celeval

Not necessarily. The game has plenty of talented players who aren't popular at the outset.

And the key with the hometown college team is to realize that Popularity is league-wide (i.e. it doesn't change based on team, via trade or whatever). To use an example SkyDog and I talked through, Thomas Brown may be a very popular Falcon (since he grew up in Tucker and played for UGA), but RaiderFan couldn't care less.

Does the game have players that were popular at the outset but did not boom? From what I understand, the answer is no.

[ShaneTheMaster]

Originally Posted by Ben E Lou

Again, it's simply not a weird correlation.

So, you think Fan Popularity should correlate to a) the actual hidden potential rating of the players or the b) perceived rating that everyone sees before/during the draft? Option "a", to me, is weird.

[Celeval]

Originally Posted by ShaneTheMaster

Does the game have players that were popular at the outset but did not boom? From what I understand, the answer is no.

Absolutely the answer is yes. A guy could be very popular and rated 77 out of the gates, but nega-creep to 52. In addition, random busts can hit anyone.

From the data I posted in the other thread, the underlying potential of any player who has at least a 50 popularity is at least a 50 potential. Boom/bust doesn't matter - that's what he is at the time of the draft under the covers.

He could come out viewed as a 30 player who creeps to 50 or an 80 player who nega-creeps to it.

[Celeval]

And really, the term we should use is creep - the booms/busts are volatility-based, have no bearing on popularity, and happen during Training Camp.

[ShaneTheMaster]

Originally Posted by Celeval

Absolutely the answer is yes. A guy could be very popular and rated 77 out of the gates, but bust to 52. In addition, random busts can hit anyone.

From the data I posted in the other thread, the underlying potential of any player who has at least a 50 popularity is at least a 50 potential. Boom/bust doesn't matter - that's what he is at the time of the draft under the covers.

He could come out viewed as a 30 player who booms to 50 or an 80 player who busts to it.

Ok, I follow you.. so I will rephrase the question:

Does the game have players that were popular at the outset but did not become a solid player?

[Celeval]

One quick example from the league where I pulled the draft files from:

The most recent draft had a QB go at 1.5 overall to the Colts. The player was rated 38/69 in the draft pool, is currently (post-draft) 39/69 on the Colts roster and has a popularity of (oddly enough) 69. However, his under-the-hood potential is a 60; and he'll creep down to that over the next few seasons.

[Celeval]

Originally Posted by ShaneTheMaster

Ok, I follow you.. so I will rephrase the question:

Does the game have players that were popular at the outset but did not become a solid player?

Based solely on FOF2004 data, the only players who were popular (>50) in the draft pool had 50+ potential in the draft pool as well. Possibilities for them not being solid players would include the random-volatility-busts (which could and do happen to anyone) and injuries.

FYI, I'm copying this post to the other thread to continue discussion there.

[Ben E Lou]

Originally Posted by ShaneTheMaster

So, you think Fan Popularity should correlate to a) the actual hidden potential rating of the players or the b) perceived rating that everyone sees before/during the draft? Option "a", to me, is weird.

Re-read my post on the matter. The masking has nothing to do with realism whatsoever. For the purposes of this discussion, it should be just ignored. What I'm saying is that in most cases from a design perspective, popular players should be good players, and that's precisely what's being modeled here. The fact that FOF has a fog-o-war is an entirely different discussion. Just because a player is masked doesn't mean that he's not a good player.

[stretch]

Originally Posted by Celeval

One quick example from the league where I pulled the draft files from:

The most recent draft had a QB go at 1.5 overall to the Colts. The player was rated 38/69 in the draft pool, is currently (post-draft) 39/69 on the Colts roster and has a popularity of (oddly enough) 69. However, his under-the-hood potential is a 60; and he'll creep down to that over the next few seasons.

N00b question: How'd you see the "under the hood" potential?

Or was that a hypothetical?

[Subby]

Originally Posted by stretch

N00b question: How'd you see the "under the hood" potential?

Or was that a hypothetical?

In 2k4 the true player ratings aren't encrypted (like they are in 2k7) so they are easy to see if you use a hex editor or some other similar program.