Ratings and Stats

[Dutch]

I know Jim always wanted ratings to mean more to a player than stats. I wanted to throw an idea out to the gallery and see what you think.

What if FOF were to get rid of scouting errors and masks altogether. And instead of showing the current true ratings (or the scouting error/masked version), it hid them from view?

When you click on a player card, you would still see ratings, but instead it would be a snapshot in time of the players true ratings from the previous season. (Most likely post-training camp with rookies maintaining blue bars until their 2nd year).

The mask then, would be the players current stats as they accumulate during the season. And the scouting error would then be *your* ability to properly predict whether or not the player is still playing as expected.

Perhaps the stats system is possibly too problematic to make this a reality, but maybe it's good enough.

Thoughts?

[gstelmack]

My view is current should always be perfectly accurate (you know how good a guy is right now from practices, game film, etc) but that future should be a less accurate approximation, like say a letter grade.

Or in other words I find the FBCB scouting model / player rating system to be the best one I've seen in any text sim.

[Hammer]

You don't get spot on ratings in the real world, I would sooner keep as close as possible to reality. A bit of scout error is great.

[gstelmack]

You get a lot closer than we have for a guy's current. I agree completely that future is a crapshoot and should be in a game.

[Hammer]

You get a lot closer than we have for a guy's current.


Do you really think so? Was Joe Montana a 78, 83 or a 93? Shaun Alexander at his best...58, 68 or 78?

You ask a range of experts your going to get a wide variation of opinions, some player more than others.

I know some players are heavily masked in FOF, but not many. You get a pair of decent scouts they are usually with 2 or 3 points of each other.

[gstelmack]

Do you really think so? Was Joe Montana a 78, 83 or a 93? Shaun Alexander at his best...58, 68 or 78?

You ask a range of experts your going to get a wide variation of opinions, some player more than others.

I know some players are heavily masked in FOF, but not many. You get a pair of decent scouts they are usually with 2 or 3 points of each other.

You're talking overall, I'm more focusing on the individual attributes like how well he reads a hole, etc. You can bet that Mike Holmgren and Bill Walsh knew exactly what they had with each player. They didn't know how good the player would someday turn out when they acquired them, but they knew what they had on the field from week to week.

You could make a case for some mental attributes that can vary week to week, or perhaps have a consistency or effort rating that controls/indicates how close to those attributes a player plays week-to-week to measure ups-and-downs, but the whole point of ratings in games is to give you information without making you watch every minute of every practice and every game with a stopwatch and having a staff of fellow gameplayers that keep a closer eye on the guys. I expect the scout numbers to reflect that.

[Hammer]

I do agree that the future potential model needs some refinement. The letter grade seems a decent way to do it. When guys lose or gain a few points in his first camp, it kind of makes it a bit obvious where he is going. I'm not sure thats a good thing, it would be nice for there to be a bit more uncertainty as to where a guy is going. The same could be said of the draft, its got a bit too predictable now it has been broken down so much.

I don't really think I see much of a problem with how current is reflected right now though. I think there are a number of issues and ideas that could be worked on rather than this. Team morale, form and greatly improving the importance of cohesion are things I would like to see worked on.

[Jughead Spock]

Don't forget the system. Someone like Alexander was probably a 48, but was behind a pair of 90+s in Jones and Hutcheson*. Stats =/= ratings, and shouldn't.

* not trying to make an NFL argument, just trying to make a point. Take debates elsewhere, please.

[Hammer]

Don't forget the system. Someone like Alexander was probably a 48, but was behind a pair of 90+s in Jones and Hutcheson*. Stats =/= ratings, and shouldn't.


This kind of illustrates my point. I never rated Alexander either, speaking as a Seahawk fan. You can bet there were experts out there who would have him right around 70 when he was at his best. I would have figured his overall to be around 55ish. Very poor receiver, 4.50 speed, soft after contact, good endurance, decent ability to find a hole. Maybe 48 would be better on reflection, he really didn't have many assets you could put your finger on.

[gstelmack]

I don't really think I see much of a problem with how current is reflected right now though.

There's a huge issue with the masking. We have plenty of examples of mid-20s rated guys that play very, very, very well that aren't explained by the Alexander example above, and lots of examples via creeping where guys are much better than the scout is trying to tell you. So now you're forced into playing a puzzle game where you try to read the clues to find out if a guy is masked and actually better (or worse) than your scout is telling you he his, and tracking lots of this outside of the game.

Some may enjoy the masking of current, but I really don't at all.

[gstelmack]

This kind of illustrates my point. I never rated Alexander either, speaking as a Seahawk fan. You can bet there were experts out there who would have him right around 70 when he was at his best. I would have figured his overall to be around 55ish. Very poor receiver, 4.50 speed, soft after contact, good endurance, decent ability to find a hole. Maybe 48 would be better on reflection, he really didn't have many assets you could put your finger on.

I doubt any experts had him rated that high, most acknowledged the blocking in front of him as being key. Alexander had a high hole recognition, that was his key attribute. The rest was average and he managed to overcome that by having a pair of monsters that opened holes for him to hit. FOF actually models that pretty well.

[Hammer]

I really haven't seen any 20 rated guys or even 30 rated guys, that are that good at all in any of my leagues. I have read about a couple on here, but surely it isn't that widespread?

[Dutch]

I really haven't seen any 20 rated guys or even 30 rated guys, that are that good at all in any of my leagues. I have read about a couple on here, but surely it isn't that widespread?

I can't vouch for the prevalency*, but I've seen it. I had a safety in GEFL that played for years as a 20 something rated safety. My backup QB in IHOF has been in the 20's for years and puts up 275 yards and a 90+ QB rating (or better) on the field.

(*FOFC spell check doesn't think that's a word, but I do!)

[Jughead Spock]

Sometimes it depends on where they're marked down too, especially in the case of pass rushers.

I do agree that it's frustrating if the bars are lies. From an immersion perspective, can just chalk it up to them over- or under- achieving or something.

Just as a total shot in the dark, is there any tie between volatility and this behavior? I know it's mostly used in VSOD and VSOL times, but is there something that says, this 30/30 guy with a 90 volatility can play crazy over his head?

[Firefly]

Then there's how ratings are constructed. From bars, I assume. But you see some guys who seem to have pretty good bars, but they get a low rating (and not just from one scout), and the opposite, too. Guys that are in the 60s whose bars seem very mediocre.

For a while I thought one big red bar would get double credit, but now I've seen guys with all even bars, below the poverty line, get decent ratings. I've gone as far as adding up all the red bar points in different players, to compare them "in bulk", and ratings tell a another story than what you see there. So basically scout ratings appear to be a mystical experience sim. A bit like grades for players in the draft.

They should either be improved and made reliable, or eliminated altogether, leaving just bars, like Jim said once he wanted the game to be. I think that would be cool.