• Welcome to the New BGO! We know you will have questions as you become familiar with the new software. Please take a moment to read our New BGO User Guide which will give you a great start. If you find potential glitches, please report those in the Possible New Site Glitches and Issues thread. If you have questions, post them in the New Site Questions? thread.
  • The 2020 Season will be the last for our little community. Following the final WFT game of the 2020 season, the site will close it's doors. We wanted to give you a little advance notice so that you could retrieve any photos, content, or other material from the site before it became unavailable, so that you could exchange contact info with anyone you may desire, and ensure that folks would forgo any site donations going forward. We have had a blast being your favorite Redskins and WFT watering hole for the past decade. We had a great run and each of you were a part of it. Thanks to all of you for your amazing contributions and camaraderie. You made this a special place.

Capital Obsession: Delving Into a Numbers Game

One of many experimental iterations ...

KDawg

The 1st Round Pick
Joined
Jul 24, 2009
Messages
590
Reaction score
0
Points
0

College at Brockport


Delving Into A Numbers Game Recently, I began reading The Hidden Game of Football by Bob Carroll, Pete Palmer and John Thorn. It’s an older book,...

More...
 

Hog Fever

Vescere bracis meis
Joe Gibbs Club Member
Joined
Jul 15, 2009
Messages
1,656
Reaction score
47
Points
48
Location
Bryson City, NC


Very interesting. Like you mentioned “what’s good?”. I understand their grading but without a context it’s hard to make and reasonable judgment from the data.

If you are taking requests for comparison, how about LT from a few years back? You could use that as one of the best seasons for a RB in recent memory. How about Arian Foster and Michael Turner for contemporary guys?
 

KDawg

The 1st Round Pick
Joined
Jul 24, 2009
Messages
590
Reaction score
0
Points
0

College at Brockport

Very interesting. Like you mentioned “what’s good?”. I understand their grading but without a context it’s hard to make and reasonable judgment from the data.

If you are taking requests for comparison, how about LT from a few years back? You could use that as one of the best seasons for a RB in recent memory. How about Arian Foster and Michael Turner for contemporary guys?
Are you volunteering your services to help figure this out? ;)

And by the way, the system they proposed is MUCH simpler. I changed it and added to it to make it more "accurate". I use quotes because the truth is that I don't know if it made it any more accurate, but I sure think it did :)
 

fansince62

The All-Time Great
Joined
Jul 22, 2009
Messages
16,349
Reaction score
156
Points
343


this is gonna take some time to digest.

btw...you are amazing poster!
 

China

The Team Captain
Joe Gibbs Club Member
Joined
Aug 3, 2009
Messages
2,671
Reaction score
21
Points
68
Location
Falls Church, VA

Michigan State

Interesting, I see a lot more fails than wins in Torain's runs, maybe that has something to do with why they got away from running so much?

KDawg said:
A touchdown is always worth 4 points.
And here, I always thought touchdowns were worth 6 points.


I also think it would be helpful to have something to compare these numbers to, but thanks for the effort nonetheless. It is greatly appreciated.
 

Goaldeje

The Legend
Joined
Feb 1, 2010
Messages
18,383
Reaction score
47
Points
328
Location
Waynesboro, VA

James Madison

I wonder if this speaks as much to the line as to Torain. Play calling factors in as well. How many of those fails at the 3-4th plays were because of obvious play calling? If I have time next week, I will look at our line and see if they progressed at all as the year went on.
Posted via BGO Mobile Device
 

Lanky Livingston

The Commissioner
Joined
Jul 15, 2009
Messages
27,017
Reaction score
0
Points
0
Location
Houston, TX

Florida Atlantic

Not to bag on you, KDawg, but this is kind of useless without some sort of comparison or baseline. Very interesting numbers, but don't say much without something to compare them too! Need to analyze someone like Jim Brown or Walter Payton or something to get a baseline of what's "good." And no, I'm not volunteering. :p
 

KDawg

The 1st Round Pick
Joined
Jul 24, 2009
Messages
590
Reaction score
0
Points
0

College at Brockport

I wonder if this speaks as much to the line as to Torain. Play calling factors in as well. How many of those fails at the 3-4th plays were because of obvious play calling? If I have time next week, I will look at our line and see if they progressed at all as the year went on.
Posted via BGO Mobile Device
Of course it does. You're spot on.

Football statistics are not like baseball statistics. Most of baseball's stats are based on individual performance, so it's easy to isolate. Sure, there are some stats that depend on others, but even then it's the effort of a few people. In football, all eleven players are accountable on every play. And even the players on your defense and special teams are responsible for your offense... And your offense and teams are responsible for your defense. So on and so forth. And then your coaches factor in. It's almost 100% impossible to isolate a single player on a football team.

It would be interesting to look into the offensive line, though I'm not sure how we'd isolate them. Perhaps just look at them as an entire offense?

This system, by the way, is not designed just for the running backs. It should be able to be used on all players.

fansince,

Thank you sir.

China,

I'd say that theory has some credence. But then if we broke down McNabb/Grossman's wins/fails I bet we'd see a ton of fails there as well.
 

KDawg

The 1st Round Pick
Joined
Jul 24, 2009
Messages
590
Reaction score
0
Points
0

College at Brockport

Not to bag on you, KDawg, but this is kind of useless without some sort of comparison or baseline. Very interesting numbers, but don't say much without something to compare them too! Need to analyze someone like Jim Brown or Walter Payton or something to get a baseline of what's "good." And no, I'm not volunteering. :p
Seems like you take issue with every one of my entries. That's fine. And I understand. I think I even said as much in the piece. What I didn't say was that I intend on breaking down a few other backs for comparison's sake in my next blog.

I would have liked to have done it for this one, but breaking down Torain took enough time. :rockstar:
 

servumtuum

The Owner's Favorite
Joined
Jun 30, 2009
Messages
7,138
Reaction score
0
Points
116
Location
Raleigh, NC

Indiana

KDawg, you might want to check out Advanced NFL Stats website-their methodology of football statistical analysis is based on The Hidden Game of Football to which they've added probability analysis, game theory, multiple regression types, and, possibly most importantly, they've done a lot of the number crunching you may be looking to do yourself-it might both save you some "re-inventing the whel" and offer further insights for your own research. I'm looking forward to your further posts on this as I'm a statistical analysis geek (I used to do it for a living as a quality technician).

They also have a huge database of play-by-play records of regular season games going back to the 2002 season-they have just finished compiling the 2010 regular season.

Here's the link: http://www.advancednflstats.com/
 

KDawg

The 1st Round Pick
Joined
Jul 24, 2009
Messages
590
Reaction score
0
Points
0

College at Brockport

Excellent web site. NFL.com's play by play works for me. I like some of the changes I made, but these are outstanding stats. Will be fun to stack these up to what I come up with for now... And in the end, these may even be what I use when I no longer feel like going through each stat.

The system I'm using is something I'm going to extend to the team that I coach.

Thanks for the link. Great stuff.
 

Lanky Livingston

The Commissioner
Joined
Jul 15, 2009
Messages
27,017
Reaction score
0
Points
0
Location
Houston, TX

Florida Atlantic

Seems like you take issue with every one of my entries. That's fine. And I understand. I think I even said as much in the piece. What I didn't say was that I intend on breaking down a few other backs for comparison's sake in my next blog.

I would have liked to have done it for this one, but breaking down Torain took enough time. :rockstar:
No no, I think these numbers are very cool - but Torain's don't mean anything by themselves. Just the math nerd in me, I suppose.
 

Hog Fever

Vescere bracis meis
Joe Gibbs Club Member
Joined
Jul 15, 2009
Messages
1,656
Reaction score
47
Points
48
Location
Bryson City, NC


Are you volunteering your services to help figure this out? ;)

And by the way, the system they proposed is MUCH simpler. I changed it and added to it to make it more "accurate". I use quotes because the truth is that I don't know if it made it any more accurate, but I sure think it did :)
YOU brought it up :D I'm merely taking your interesting idea to the next logical level.

Besides, I pretend to have a life outside of football fandom.....stop laughing....If I have some time later this week I'll see if i can put together some numbers.
 

fansince62

The All-Time Great
Joined
Jul 22, 2009
Messages
16,349
Reaction score
156
Points
343


turnovers relative to time and score make a huge difference.....end of the game on the final drive down by 2.....bigger buzz killer than turnovers when ahead by 21.

how is the system normalized across players who play against different quality defenses?
 

China

The Team Captain
Joe Gibbs Club Member
Joined
Aug 3, 2009
Messages
2,671
Reaction score
21
Points
68
Location
Falls Church, VA

Michigan State

China,

I'd say that theory has some credence. But then if we broke down McNabb/Grossman's wins/fails I bet we'd see a ton of fails there as well.
No doubt, but I don't think it requires much in-depth analysis to see the fails with McNabb and Grossman. :)
 

KDawg

The 1st Round Pick
Joined
Jul 24, 2009
Messages
590
Reaction score
0
Points
0

College at Brockport

Hog,

I'm planning on using Barry Sanders, LaDainian Tomlinson. Trung Canidate and Redskin's version Clinton Portis to see how their numbers stack up. Interesting group, and I may change that group by the time I do it. But I want a little bit of everything

turnovers relative to time and score make a huge difference.....end of the game on the final drive down by 2.....bigger buzz killer than turnovers when ahead by 21.

how is the system normalized across players who play against different quality defenses?
Indeed those numbers would add to the system, but I'm not sure that those things happen enough to seriously jolt it.

And it's not normalized. I take the liberty of assuming that the defenses played will average out a bit in order to make it more accurate.

No doubt, but I don't think it requires much in-depth analysis to see the fails with McNabb and Grossman. :)
Touche. :)
 

KDawg

The 1st Round Pick
Joined
Jul 24, 2009
Messages
590
Reaction score
0
Points
0

College at Brockport

Sanders would be intriguing - he routinely had negative runs followed up by huge-gainers.
Couldn't find any of his game logs, so I went with Chris Johnson from last year. Halfway through he has ALOT more fails than wins but his points are through the roof due to big plays. He had more points in one game than Torain had through the entire season.
 

Lanky Livingston

The Commissioner
Joined
Jul 15, 2009
Messages
27,017
Reaction score
0
Points
0
Location
Houston, TX

Florida Atlantic

If you wanted to get really nerdy, you could do some sort of statistical correlation between a RBs points & wins vs. team wins. You could find a sweet spot of sorts for the production of a RB.

Actually, that would be more of a team running game thing, probably.
 

KDawg

The 1st Round Pick
Joined
Jul 24, 2009
Messages
590
Reaction score
0
Points
0

College at Brockport

Yeah, but you could do it using this model, but it would probably be a bit more of a pain.

I think I'm going to be making some tweaks to the formula, though.

What's your guys opinion on this one:

If a runner gains yards OVER the necessary yardage, I give them bonus points (.2 on 1st/2nd down and .5 on 3rd/4th down) for every extra yard gained. But for every yard UNDER what's needed, they only receive the fail negative point.

I think, for example, I should change it to this:
1st and 10, Gain of 3, Fail (-1 pt), 1 yard under (-.2) Total Loss: 1.2 points

OR

3rd and 3, Loss of 5, Fail (-1 pt), 8 yards under necessary yardage (-4.0) Total Loss: -5 points

Currently it would work like this:
1st and 10, Gain of 3, Fail (-1 pt) Total Loss: 1 points

OR

3rd and 3, Loss of 5, Fail (-1 pt), 2 yard loss or more (-2) Total Loss: -3 points

Opinions? It's not that difficult a change to make, just recalculating points. I just don't know how beneficial it would be.
 

Users Who Are Viewing This Thread (Total: 1, Members: 0, Guests: 1)

Private conversations
Help Users
    You haven't joined any rooms.
    Chat 0
    Top