FBGRatings Seeks to Recalibrate Madden Ratings, Change Game

[thephenom21]

http://www.fbgratings.com/members/pr...php?pyid=70862

why is Shady McCoy so slow? these ratings are ********.

[DCEBB2001]

Quote: Originally Posted by thephenom21

http://www.fbgratings.com/members/pr...php?pyid=70862 why is Shady McCoy so slow? these ratings are ********.

Slow compared to who? Keep in mind that these ratings utilize a normal distribution that deviates from over-inflation of attributes. So if you are used to EA posting a player with a SPD in the 90s, it is most likely that a speed rating is lower than that. The average 40 time for any player tested at a combine or pro day since 1998 is 4.81s. Instead of having this rating set closer to 78 as EA does, this time is set to 70, so all players will appear to be a bit slower, but more true to life as they reflect the data distribution of real life data. There are oodles of posts in regards to this in this thread. McCoy does NOT possess great speed. He only ran a 4.50 at his pro day on 3/17/2009:

http://www.nfldraftscout.com/members...php?pyid=70862

A 4.50 40 time equates to an 85 SPD in this system. THAT is why he appears to be "so slow" in your uninformed opinion of this ratings system (I say uninformed because you obviously have not read through the entirety of the thread, for this question has already been answered numerous times for many different players).

[Aftershock9958]

DCEBB, I liked what you were trying to do with this system and I'd hoped to eventually play with your rosters, but I've been curious. Now that player editing is largely closed off to anything but play now, what are your plans for FBGratings?

[DCEBB2001]

Quote: Originally Posted by Aftershock9958

DCEBB, I liked what you were trying to do with this system and I'd hoped to eventually play with your rosters, but I've been curious. Now that player editing is largely closed off to anything but play now, what are your plans for FBGratings?

I plan on continuing the project. EA said that they would put it back in, most likely next year, so I will keep things going regardless.

[Big FN Deal]

Quote: Originally Posted by DCEBB2001

Slow compared to who? Keep in mind that these ratings utilize a normal distribution that deviates from over-inflation of attributes. So if you are used to EA posting a player with a SPD in the 90s, it is most likely that a speed rating is lower than that. The average 40 time for any player tested at a combine or pro day since 1998 is 4.81s. Instead of having this rating set closer to 78 as EA does, this time is set to 70, so all players will appear to be a bit slower, but more true to life as they reflect the data distribution of real life data. There are oodles of posts in regards to this in this thread. McCoy does NOT possess great speed. He only ran a 4.50 at his pro day on 3/17/2009: http://www.nfldraftscout.com/members...php?pyid=70862 A 4.50 40 time equates to an 85 SPD in this system. THAT is why he appears to be "so slow" in your uninformed opinion of this ratings system (I say uninformed because you obviously have not read through the entirety of the thread, for this question has already been answered numerous times for many different players).

I really like this question because speed would be relative in the game using a true universal scale. I really like the way you broke that down because in the "Madden world" suddenly 100 is no longer the best ever but just some subjective interpretation of "game speed". SMMFH

Anyway, keep up the good work DCEBB and hopefully this method will eventually catch on with EA/Tiburon.

[DeuceDouglas]

Quote: Originally Posted by Big FN Deal

I really like this question because speed would be relative in the game using a true universal scale. I really like the way you broke that down because in the "Madden world" suddenly 100 is no longer the best ever but just some subjective interpretation of "game speed". SMMFH Anyway, keep up the good work DCEBB and hopefully this method will eventually catch on with EA/Tiburon.

I think there's way too much put into forty times. I think things also become less universal when Pro Days are taken into account as well. A guy like Patrick Willis ran a 4.51 at the combine and then ran a 4.37 at his pro day which just goes to show how different things can be on a given run. If you take his pro day into account, his speed would vault players who out ran him in identical conditions at the combine but didn't participate in a pro day.

[DCEBB2001]

Quote: Originally Posted by DeuceDouglas

I think there's way too much put into forty times. I think things also become less universal when Pro Days are taken into account as well. A guy like Patrick Willis ran a 4.51 at the combine and then ran a 4.37 at his pro day which just goes to show how different things can be on a given run. If you take his pro day into account, his speed would vault players who out ran him in identical conditions at the combine but didn't participate in a pro day.

Every player who runs at the combine has the option to run at a pro day. If you choose not to run at a pro day after you run at the combine it is most likely because you feel that your time is close to what you can run. Otherwise, why would you pass up that opportunity? Many players choose to stand on their combine times only because they don't want to hurt their stock by running slower at their pro day (see DeSean Jackson). Others feel they are faster than how they time at the combine and want another chance to show it (see Devin Hester). To me, if you pass up the option to run again after a few more weeks of training, then you must be accepting of your combine time.

Your statement of seems to allude to the assumption that players are always running faster at their pro day than at the combine. This does not always hold true. Guys like Brian Cushing, Will Blackmon, and David Anderson all ran slower at their pro days than at the combine. To me, the more data we have, the better the idea we have on someone's true speed compared to other players. Once is happenstance, twice is coincidence, and three times is a trend. Give me more trials to see what the trend is.

[Big FN Deal]

Quote: Originally Posted by DeuceDouglas

I think there's way too much put into forty times. I think things also become less universal when Pro Days are taken into account as well. A guy like Patrick Willis ran a 4.51 at the combine and then ran a 4.37 at his pro day which just goes to show how different things can be on a given run. If you take his pro day into account, his speed would vault players who out ran him in identical conditions at the combine but didn't participate in a pro day.

By "universal" I meant the same across the board meaning if an OL and a CB run the same 40 time and 40 times are being used to set speed ratings, then they should have the same speed. As far as using the 40 time alone for speed, again it's the best standard currently, imo but even if they decided some other formula for calculating speed, that's cool with me, as long as it is a universal method that can be applied to every player and their real life data.

I have posted this in other places and maybe even somewhere in this thread but I think the universal formulas should be subjective, not the ratings. The issue I have now is there doesn't appear to be any universal formula(s) for inputting real player data into to achieve player ratings in Madden but there should be, imo.