The new user interface is in preview!

Want to check it out? Click here! (If you don't like it, you can still switch back)

The draft is underway!

Click here to go to your war room, or visit the war room item in the draft menu.

League Forums

Main - General MFN Discussion

Serious Talk About Speed (with Maths)

By setherick
12/26/2020 11:17 pm
I'm posting this to open up a frank discussion on speed and how speed works in the game. I'm finding far too many owners don't understand the speed rating, and it's leading to a lot of arguments.

Let's get this out of the way right now. There is a lot of hidden math in this game, which is great if you can do estimate complex equations in your head, but terrible if you think that game cards actually show a linear difference between players.

Speed is one of these complex equations.

Definition

Here's JDB's definition of speed: "It's actually just a downward-opening quadratic as opposed to exponential or logarithmic, where the vertex of the parabola is at 100. Basically it just slows down the impact of increasing numbers as you near 100 and increases the impact of decreasing numbers as you approach 0."

But, Seth, I have no idea what that means?

If we want to think through what this parabola looks like when compared to the Speed listed on the player card, we need to think about what the displayed speed compared to a linear scale. This means we need to think about a parabola that intersects at the origin and has its line of symmetry where x = 100 with a max y of 100.

Not to take you back all the way to junior high algebra, but that gives us an equation like this:

-((0.1X-10)^2)+100 = y
2X - 0.01X^2 = y

Note: This is not the actual speed equation unless I can guess really, really good. This is used as an example of comparing linear distance to distance on the parabola.

Also, if I have gotten this wrong, someone let me know. It's been a long time since I was in junior high algebra myself.


Great, but what does that really mean?

It means exactly what JDB says above. The closer you get to 100, the less of a gap there is between Speed ratings.

Let's take the following 5 players:

Player A: 64 Speed (Displayed)
Player B: 75 Speed (Displayed)
Player C: 84 Speed (Displayed)
Player D: 91 Speed (Displayed)
Player E: 96 Speed (Displayed)

On a linear scale, each player has a 10 point gap between them, which means the player would need to be 10 Speed points faster to equal the next player.

On the parabolic scale, A actually needs 11 speed points to reach Player B; but, Player B need 9 to reach Player C. Player C needs 7 to reach Player D, and Player D only need 5 to reach Player E.

Caution: This works the other way too!

If anything, what this should show you is to never play a LB at below 80 SP ... ever.

Let's take a 90 Speed RB. An 80 SP LB would need 12 linear points to equal the RB. An 85 SP LB would need 6 linear points. An 88 SP LB would need 2 linear points (this is the intersection point), and a 89 SP LB would need 0-1 linear points.

This means that a max SP LB at 237# is equal or faster than 95% of RBs.

It's only until an RB gets to that magical 95 SP number, which is only possible by playing an RB at WR, that an 89 SP LB needs 10 linear points to equal it.

Re: Serious Talk About Speed (with Maths)

By CrazyRazor
12/27/2020 4:30 pm
So you're actually saying that speed is the most important attribute for just about every position, but especially important for the defense?

This explanation, about how drastic the variation of speed is from one end of the spectrum to the other, is reinforcing the fact that speed kills in this game.

I'm not complaining. I'm merely pointing out the fact that this just proves how important it would be to have specific parameters concerning attributes for each position.

Re: Serious Talk About Speed (with Maths)

By setherick
12/27/2020 4:47 pm
For the record, I have never not said Speed was not important.

But, when you really start digging into things, there are a few positions where having a slower player won't hurt you. The clearest example is CB1. You can shove a 40 SP, 100 everything else CB at CB1 and he'll be just fine.

You cannot play the same guy at CB2 without him getting torched.

Re: Serious Talk About Speed (with Maths)

By Mcbolt55
12/27/2020 6:11 pm
Ok that’s just weird, where’s the old madden option to flip a play to expose that side? Why wouldn’t any routes or plays to the strong side require speed in that direction?

Re: Serious Talk About Speed (with Maths)

By setherick
12/27/2020 6:38 pm
Mcbolt55 wrote:
Ok that’s just weird, where’s the old madden option to flip a play to expose that side? Why wouldn’t any routes or plays to the strong side require speed in that direction?


There are, but the WR1 isn't currently targeted often on them or at least the primary target. (See my post about primary targets in the Help Forum.)

And then it's still hard to exploit the CB1 because the WR1 is often running into coverage.

Re: Serious Talk About Speed (with Maths)

By Meatmen
2/06/2021 7:00 pm
I love when you do the Maths, I don't know my goesinta's from my takeaways, but you sure can cipher! lol! Thanks this is awesome sethrick. I honestly appreciate the effort. wish I would have seen this sooner.......................MM
Last edited at 2/06/2021 7:06 pm

Re: Serious Talk About Speed (with Maths)

By setherick
10/15/2021 8:04 am
Time to revive this thread since 4.6 is going to confuse people even more with what speed actually means. I don't have the time to fully write out my notes, but here's a preview.

In 4.6, JDB "raised the 0 point", which in terms of a parabola which is what speed is means that he adjusted the Y intercept to some non-zero number. Assuming that (100, 100) is still the vertex of this parabola so that 100 linear points = 100 speed, this will have the affect of flattening the parabola so that differences in speed are actually more exaggerated than they are now.

So let's go back to the first post:

Player A: 64 Speed (Displayed)
Player B: 75 Speed (Displayed)
Player C: 84 Speed (Displayed)
Player D: 91 Speed (Displayed)
Player E: 96 Speed (Displayed)

Now let's assume that JDB raised the zero point a reasonable 10 points on the Y. Here's the same players:

Player A: 67.6 Speed (Displayed)
Player B: 77.5 Speed (Displayed)
Player C: 85.6 Speed (Displayed)
Player D: 91.9 Speed (Displayed)
Player E: 96.4 Speed (Displayed)

Everything looks good right? No, it actually means that players are getting slower because the display value needs to be impossibly high in terms of display value to reach the highest speed.

Now, let's assume that JDB raised the Y intercept to an unreasonably 25.

Player A: 73 Speed (Displayed)
Player B: 81.25 Speed (Displayed)
Player C: 88 Speed (Displayed)
Player D: 93.25 Speed (Displayed)
Player E: 97 Speed (Displayed)

If that's the case, the fastest running back (display value) doesn't even qualify as Player D. And Player B from the first example is barely faster than Player A if the Y intercept is raised to 25.

All of this is conjecture of course (nod to fred). But it is the math behind why flattening a parabola means that players are now slower in 4.6.

Re: Serious Talk About Speed (with Maths)

By TheAdmiral
10/15/2021 8:12 am
setherick wrote:
Time to revive this thread since 4.6 is going to confuse people even more with what speed actually means. I don't have the time to fully write out my notes, but here's a preview.

In 4.6, JDB "raised the 0 point", which in terms of a parabola which is what speed is means that he adjusted the Y intercept to some non-zero number. Assuming that (100, 100) is still the vertex of this parabola so that 100 linear points = 100 speed, this will have the affect of flattening the parabola so that differences in speed are actually more exaggerated than they are now.

So let's go back to the first post:

Player A: 64 Speed (Displayed)
Player B: 75 Speed (Displayed)
Player C: 84 Speed (Displayed)
Player D: 91 Speed (Displayed)
Player E: 96 Speed (Displayed)

Now let's assume that JDB raised the zero point a reasonable 10 points on the Y. Here's the same players:

Player A: 67.6 Speed (Displayed)
Player B: 77.5 Speed (Displayed)
Player C: 85.6 Speed (Displayed)
Player D: 91.9 Speed (Displayed)
Player E: 96.4 Speed (Displayed)

Everything looks good right? No, it actually means that players are getting slower because the display value needs to be impossibly high in terms of display value to reach the highest speed.

Now, let's assume that JDB raised the Y intercept to an unreasonably 25.

Player A: 73 Speed (Displayed)
Player B: 81.25 Speed (Displayed)
Player C: 88 Speed (Displayed)
Player D: 93.25 Speed (Displayed)
Player E: 97 Speed (Displayed)

If that's the case, the fastest running back (display value) doesn't even qualify as Player D. And Player B from the first example is barely faster than Player A if the Y intercept is raised to 25.

All of this is conjecture of course (nod to fred). But it is the math behind why flattening a parabola means that players are now slower in 4.6.




Is there anything you've written above which isn't going to confuse people?

I'm being somewhat facetious with that comment but it's just the irony of you're opening paragraph made me laugh.

Re: Serious Talk About Speed (with Maths)

By setherick
10/15/2021 8:40 am
Short takeaway. The players are slower due to math.

Re: Serious Talk About Speed (with Maths)

By TheAdmiral
10/15/2021 8:46 am
setherick wrote:
Short takeaway. The players are slower due to math.


Thanks for not being offended, I genuinely wasn't trying to do that. However some people have a different sense of humour to me.