Discussion: RAS, cross post from the browns forum.

[Kiwibrown]

So something that I have been seeing lately is the RAS number. I did a little looking into it. 


I made this in the Browns forum i thought I would share it in NFL Gen. 

It was created by  @mathbomb , RAS, relative athletic score., grades players on a scale of 0-10 based on all their athletic and size metrics. It was developed after the SPARQ score as far as I know, but is somewhat similar to it. 

 

  Quote

"Relative Athletic Scores take player measurements and put them on an easy to understand 0 to 10 scale compared to their position group. A final score is then produced which is also on a 0 to 10 score to show overall athleticism for a draft prospect. This data can be used to chart trends over time, showing that overall athleticism is likely a contributing factor to player success in the NFL."

https://relativeathleticscores.com/
 

Here is some more explanation. 

 

 

  Quote

 

The System

So how does it work? RAS is essentially a ranking system, so it roughly correlates to percentile. While I always say that 5.00 is average, that doesn't quite describe it accurately. A better way to describe that 5.00 middle mark is that it is the score for the average player at that position. Shouldn't that be the same thing as an average, you ask? Why no, no it is not. It's an oddity of studying these measurements so much, but in most cases the actual mathematical average (the mean) tended to be a good deal above what the average player at a position would score. So for instance, a player who had an actually average score may end up with a 7.50 for that measurement. That wasn't what I wanted at all. I used an incredible amount of math to get to the current system, you guys, and I don't mean incredible as in "wow, that's amazing!" but more in the "wow, that's overly complex and cripplingly unexplainable" kind of way. Using the calculations as they are now, we get to see what the average player at a position scored at each measurement (closer to a statistical median than a mean).

 

 

 

  Quote

 

Now it's pretty simple. The actual numbers correspond loosely to percentile, so a player with a 9.87 score for his 40 time managed to be in about the 98.7th percentile of his position group for that measurement. The final number -- the one we actually refer to as the player's RAS -- is gained by averaging the individual scores for each player at a position. This raw average is then compared to the raw averages for every other player at the position to come up with the final Relative Athletic Score for a player.

It isn't a perfect system, and like the measurements themselves, there are plenty of outliers. What it gives us, though, is a way to put a number on a player's measurements as a whole when compared to several hundred other players at their position over the past 17 years (2000-2016). It's a lot of data and I'm continually adding to it. Adding more data does affect existing scores, but not very much. After adding over 1,500 players this offseason, the biggest change was only about 4%, with most scores being affected by about 1%.

 

https://www.prideofdetroit.com/2016/5/16/11678686/relative-athletic-scores-what-they-are-and-why-they-work

Here is an interesting example of one, Christian McCaffery



CMAC gets penalized for height, at a position where height can end up being a disadvantage imo due to the shots that it opens up on your rib and torso area, as an NFL RB you dont want to be taller than 6'3" and generally I wouldnt be surpised that as long as your height is over about 5'8" height is neither a great advantage or a disadvantage. Bench is one of my bug bears as it doesn't relate to playing in the nfl. No one lies on their back and pushed 225 pounds off them. High explosion, agility and speed are more important,  c mac excels at these.

Another interesting Panther, this one from the past





Calvin, unsurprisingly comes out as the Madden create a player with a score of 10. 
 

Grades for the AFC North 2019. 












Drew forbes is at the top of the scale. He is a RAS beast, especially for explosion. If he takes to the coaching, has good bend and balance, we may have one of the steals of the draft. 



C/f another a pro bowl G from the Packers under Campen



 Next Check out Myles




He also has a good RAS




out of 1000 ranked defensive ends, Myles Ranks 2 all time. 

and where ever you are





Sachi,  had a similar draft philosophy to Dorsey when it comes to drafting athletic players. 


RAS through the ages 3 drafts

The 2007 NFL draft was very strong with multiple players achieving multiple all pros. Here is a list of players with multiple APs, their AV grade as per PF and their RAS score.

Calvin Johnson av 78, all pros 3, RAS 10 

Joe Thomas AV 83, all pros, 6, RAS 9.37

Adrian Peterson AV 94, As 4, RAS, 9.66

Patrick Willis AV 95, all pros 5, RAS, 8.5

Marshawn Lynch AV 74, all pros 1, RAS 7.84

Darrelle Revis AV 90, all pros 4, RAS 10

Eric Weedle AV 80, All pros 2, RAS, 5.1

Marshal Yanda  AV 80, APs 2, 9.03

ryan Khalil  AV 74, APs 2, RAS, 9.13

Eric Weddle has a strong career but is only a mediocre athlete according to RAS, while surprisingly Willis scores relatively low for someone who was so dominant in a short period of time. 



1998 NFL draft

Randy Moss AV 123, APs 4, RAs 9.67

alan Faneca, AV 113, APs 113, RAS 3.16

Charles woodson AV 108, APs 3, RAs not recorded

im surpised Moss isnt rated higher, it could be due to his lack weight I guess, Faneca is bottom 40%, well done to him. 

1999

Champ Bailey AV 111, APs 3, RAS 9.94

Edge James AV 114 APs 1, RAS 9.99

Ricky Williams AV 76, APs 1, RAS 9.73


1997

Tony G av 97, As 6, RAS 8.12

Walter Jones AV 97, APs 4, RAS 8.59

Ronde Barber AV 110, Aps 3, RAS 2.86

Darren Sharper AV 87, APs 2, RAS 9.01

Jason Taylor AV 117, APs 3, RAS 9.51

Sam Maddison AV 86, Aps 2, RAS 5.13

From a RAS perspective he top players of 1997 class is relatively weak, Tony G an all time great, along with Walter jones, were low to mid 8's 15-20% of players at their respective positions were 'better' athletes according to RAS. Barber had a long career, is a bottom third athlete. 

scoring over 9 is strongly correlated with top performance in a respective draft class, but isn't necessary. 2 of 20 players were below 4 and achieved multiple all pros. It turns out being an extremely good athlete is beneficial to your chances of becoming an all pro. 

There is also an exhaustive list of players who score 9 or higher that have no business being in the NFL. Ricardo louis, Mingo etc. Vernon Gholston got a 9.78 and by all accounts worked hard at his craft. The inerse is true there are people with a terrible RAs who have prefomred very wel, Calais campbell from the same draft as Gholson gets a 3.4


I hope there was something interesting in there for you



 

 

[Jakuvious]

So, I guess the question I would ask, is what ultimately is the point of this? Typically when you look at a metric that uses a formula to try to combine multiple statistics or measurements, the goal is simplicity. So when you look at the existence of passer rating, for instance, it's goal was to try to find one formula to, for argument's sake, name the best passer in the NFL. Passer rating actually gives you less information than if you just look at the component stats separately, but it makes it easier to look at and simplifies things for viewers.

But I don't feel like that goal coincides with scouting. Why would anyone really concern themselves with a formula like RAS, when they're going to get more information from just looking at the components of RAS? You don't need to know Drew Forbes RAS, you can just like at his combine results. You're getting more information and learning more if you look at his 40, and bench, and vert, and 3 cone, etc. And you want to know in what ways that player is athletic, not just that they are. If you tell me Patrick Willis had a RAS of 8.5, that could mean he had perfect size and speed but so so agility and explosion. Maybe it means he was good but not great in everything. It actually winds up telling me less about the player. It would maybe be convenient for arguments or debates about athleticism, but that really seems like that's about it.

I guess what I'm trying to get at is I don't know what this actually accomplishes that is needed. It feels like a simplification of information in lieu of actually seeing all the information. I don't want a blanket grade of a RB prospect's athleticism. I want his 40, and his 3 cone, and his etc.

[AlexGreen#20]

Analytics have their place but this is low quality analytics that haven't been parsed out yet. The lack of weighting of each of the measurements and the lack of positional specifications have this thing way back in it's infancy. 

It's also completely screwed when you get to the upper extremes just in the way it's compiled.

In addition, the lack of taking into account length causes problems. That's a significant part of the reason that Calais is so low despite anybody with eyes being able to tell he's an excellent athlete at scale.

[jebrick]

But can they play football?

[MWil23]

3 minutes ago, jebrick said:

But can they play football?

I think that RAS is overblown to a degree in the early rounds, but I'd be more willing to roll the dice on high RAS guys on Day 3, hoping to get some type of physically gifted played who could develop/refine their skills at the next level as opposed to a "football player" who is very limited athletically, albeit I'm not talking about guys that were REALLY GOOD players with limited athleticism, more like the "hey he was a decent 2nd/3rd LB/RB/WR on his own college team" type.

[jebrick]

But without context the number is meaningless.  I agree with @Jakuvious in that the components are better than the whole.  The combine just allows you to see if a player is more athletic than they might be seen in their film.  The Combine scores are only for a comparison to previous years for a basis of how a player might do in the NFL.

[MWil23]

15 minutes ago, jebrick said:

But without context the number is meaningless.  I agree with @Jakuvious in that the components are better than the whole.  The combine just allows you to see if a player is more athletic than they might be seen in their film.  The Combine scores are only for a comparison to previous years for a basis of how a player might do in the NFL.

Right, but for guys like Prospect X (the Browns OL drafted referenced above) who aren't invited to the combine, it definitely has value.

[jebrick]

13 minutes ago, MWil23 said:

Right, but for guys like Prospect X (the Browns OL drafted referenced above) who aren't invited to the combine, it definitely has value.

Why?  You look at their Pro day numbers.  And can he play football?

[warrenblue]

IWow I knew Miles Boykins was a good athlete but 9.94 is crazy . This is Megatron/Randy Moss territory. Hopefully we can finally develop an elite WR.His interview is very good too and he looks like a very smart and down to earth kid. Expecting good things from him

[Kiwibrown]

2 hours ago, warrenblue said:

IWow I knew Miles Boykins was a good athlete but 9.94 is crazy . This is Megatron/Randy Moss territory. Hopefully we can finally develop an elite WR.His interview is very good too and he looks like a very smart and down to earth kid. Expecting good things from him

The problem is it is also Ricardo louis, mingo, territory. There are too many false positives for ras to be helpful imo.

[pwny]

I love analytics. But this looks useless. 

[pwny]

10 hours ago, Jakuvious said:

I don't want a blanket grade of a RB prospect's athleticism. I want his 40, and his 3 cone, and his etc.

This is important. And I really don’t see how RAS holds even the same value as the spider chart. 

But I’d also argue that sometimes it’s good to combine stats. However, as you’ve explained, it has to have a set purpose that is helpful. Speed scores are helpful because it puts the player’s 40 time in context of his weight. These pass rusher analytics that have been coming out the last several years are helpful because they put those test results in a context that’s relevant for the position they play. Those are where the value comes from. These are just a mash of everything, pretending that every position relies on physical ability the exact same way. 

[Kiwibrown]

51 minutes ago, pwny said:

I love analytics. But this looks useless. 

I agree, I think it is too broad to be useful. 

Sparq has a focus however, as it looks for explosiveness which is, or should be coveted by every position.

[pwny]

8 minutes ago, Kiwibrown said:

I agree, I think it is too broad to be useful. 

Sparq has a focus however, as it looks for explosiveness which is, or should be coveted by every position.

SPARQ isn’t pass rusher specific, but I see it used so much for pass rushers more than other positions that I was lumping that in as part of the pass rusher analytics I was referring to. So yeah, that’s one I like too. 

[Kiwibrown]

11 hours ago, Jakuvious said:

I guess what I'm trying to get at is I don't know what this actually accomplishes that is needed. It feels like a simplification of information in lieu of actually seeing all the information. I don't want a blanket grade of a RB prospect's athleticism. I want his 40, and his 3 cone, and his etc.

If there was a position specific version that focused on what matters it could be a more useful tool. 

For a corner I am interested 40, 3 cone, splits, height to some extent and weight and bench doesnt interest me.

If there was a corner ras that was build around desired drills I'd be more interested. And the  how this is related to caree success, all pros, hof, etc 

In general 40 is probably overrated and 3 cone underrated as it relates to football.