Jump to content
Maynard_G_Krebs

Changing QB WR Priority

Recommended Posts

A few people have sent me messages asking to help explain how to alter the QB WR priority selection. Here's a brief tutorial.

Go to the destination HEX addresses below.

    Pass Play Pointers:


PRO T WAGGLE L = $A4ED (x84FD)
R AND S FLARE C = $B287 (x9297)
PRO T WAGGLE R = $ADA0 (x8DB0)
ROLL OUT R = $A808 (x8818)
ROLL OUT L = $AB6E (x8B7E)
T PLAY ACTION D = $B2BF (x92CF)
PRO T SCREEN L = $B2EF (x92FF) and $B7EE (x97FE)
PLAY ACTION = $B0C2 (x90D2)
PWR FAKE Z POST = $B7FF (x980F)
WTE F;FLICKER = $A5DE (x85EE)
SHOTGUN X CURL = $B32B (x933B) and $B7F4 (x9804)
R AND S Z FLY = $B363 (x9373)
PRO T FLARE D = $B39F (x93AF)
OFFSET FLARE E = $B3C1 (x93D1)
ONEBACK Z CROSS = $B3F5 (x9405)
ONEBACK FLARE A = $B436 (x9446)
T FLEA FLICKER = $B471 (x9481)
PWR FAKE X FLY = $A68F (x869F)
SHOTGUN X DRIVE = $B4AC (x94BC)
R AND S 3;WING = $B4F5 (x9505)
PLAYACTION Z IN = $AFE1 (x8FF1)
FLEA FLICKER = $B53C (x954C)
PRO T FLARE C = $B582 (x9592)
SHOTGUN 3;WING = $B5BD (x95CD)
SHOTGUN XY BOMB = $B5EF (x95FF)
R AND S Y UP = $B629 (x9639)
X OUT AND FLY = $B666 (x9676)
REV;FAKE Z POST = $AA26 (x8A36)
SLOT L Z DRIVE = $B69B (x96AB)
NO BACK X DEEP = $B6DF (x96EF)
SHOTGUN Z S;IN = $B715 (x9725)
REDGUN Z SLANT = $B743 (x9753)

> credit Jstout



    Additional QB Priority Information:

Example Play:
R&S FLARE C QB
$B287 (x9297): d4; c0 00 dc; e4; c0 00 f0; f0 80; 2c $B2AC; c7 $B2A6; c1 1e c8 b3;
THROW (x92AB): 94 31 25 44 32 f3 $B842; ff $B832
$B2A6 (x92B6): c1 1e c8 f3; fe THROW
$B2AC (x92BC): c1 23 2d fd; e2 04; e3 04; d8 10 e8; d8 1a f0; 27 $B838; fe THROW

Common Designs:
$B832 (x9842): STAND FOREVER
$B838 (x9848): RUN
$B842 (x9852): SWITCH CONTROL TO RECEIVER

Key:
91-94 = PASS RECEIVERS
20-2F = RANDOM GOTO (20 is likely to 2F is not likely)
C0 = DROPBACK
C1 = PASS TIMING
C7 = GOTO POINTER IF CPU (CPU and COA/MAN are different)
D4 = TAKE SNAP
D8 = MOVE TO LOCATION
E2 = BOOST RP
E3 = BOOST MS
E4 = PLAYER TAKES CONTROL
F0 = FACE DIRECTION
FE = GOTO
FF = GOTO POINTER

C1 has 3 values. The first is START TIME, the second is END TIME, and the third is TAKE A SACK CHANCE. The QB will throw the ball anywhere inbetween the start and end times (so a small range 23 2d will not vary much but a large one like 1e C8 would). The final value is the chance of taking a sack or immediately throwing it (higher is more likely to throw) even if not in time frame.

91-94 has values for the odds of throwing to a receiver and which receiver order. 91 has 2 players, 92 has 3 receivers, 93 has 4 receivers, and 94 has 5 receivers. From the play above (94 31 25 44 32 f3) 31 is 3 chance to 1 (RB1), 25 is 2 chance to 5 (TE), 44 is 4 chance to 4 (WR2), 32 is 3 chance to 2 (RB2), and F3 is F chance to 3 (WR1). The chances are 0 is very unlikely to F who is very likely and F will always be the final receivers value. The QB will look in order from first to last on the list one by one checking whether to throw or not.

Note: the QB will only throw the ball once the 91-94 command has been reached.

> credit Jstout

Share this post


Link to post
Share on other sites

the take snap hex D4 is for under center. In the formation, the center points to a location that is 4 bytes, the last byte is either D2-snap to under center QB, or D3-Snap to Shotgun qb. The Qb's shotgun take snap is D5. D6 take snap is the Fake FG. WR order is detemined in the play pointers.....A0 thru A4. A0 qill be the first receiver and so on...Have to have 2 minimum of 2 receivers in route.

Share this post


Link to post
Share on other sites
91-94 has values for the odds of throwing to a receiver and which receiver order. 91 has 2 players, 92 has 3 receivers, 93 has 4 receivers, and 94 has 5 receivers. From the play above (94 31 25 44 32 f3) 31 is 3 chance to 1 (RB1), 25 is 2 chance to 5 (TE), 44 is 4 chance to 4 (WR2), 32 is 3 chance to 2 (RB2), and F3 is F chance to 3 (WR1). The chances are 0 is very unlikely to F who is very likely and F will always be the final receivers value. The QB will look in order from first to last on the list one by one checking whether to throw or not.

Is there a known % for the 0 - F chance to throw to each receiver?

And do these numbers need to add up to anything...or can you just put every receiver at F?

Share this post


Link to post
Share on other sites

I believe its


0 100%
1 93.75%
2 87.50%
3 81.25%
4 75.00%
5 68.75%
6 62.50%
7 56.25%
8 50.00%
9 43.75%
10 37.50%
11 31.25%
12 25.00%
13 18.75%
14 12.50%
15 6.25%

Now maybe it is different for the COM plays and it seems like it could be because by my calculations the chances of throwing to the last 3 WR's is almost 0.

Chance to throw to RB1 3= 81.25%.

Chance to move on to next WR = 18.75%

Chance to throw to TE= 2= 87.5% x remaining chance = 87.5% x 18.75% =16.4%

So there is already a 97.66% chance it will throw to the first WR unless my assumption of the probabilities is wrong.

Share this post


Link to post
Share on other sites
I believe its


0 100%
1 93.75%
2 87.50%
3 81.25%
4 75.00%
5 68.75%
6 62.50%
7 56.25%
8 50.00%
9 43.75%
10 37.50%
11 31.25%
12 25.00%
13 18.75%
14 12.50%
15 6.25%

Now maybe it is different for the COM plays and it seems like it could be because by my calculations the chances of throwing to the last 3 WR's is almost 0.

Chance to throw to RB1 3= 81.25%.

Chance to move on to next WR = 18.75%

Chance to throw to TE= 2= 87.5% x remaining chance = 87.5% x 18.75% =16.4%

So there is already a 97.66% chance it will throw to the first WR unless my assumption of the probabilities is wrong.

Actually, this makes sense....but your list would be reversed, right? 0=6.25% and F=100%

Share this post


Link to post
Share on other sites

No 0 actually mean VERY likely to throw F means not likely. I THINK mgk had it reversed. Since from jstouts play info post:

(Random) = Random Chance (x00 is likely to x0F isn't likely)

(Take Sack Chance) = Take Sack (x00 is likely to xFF isn't likely)

Share this post


Link to post
Share on other sites
No 0 actually mean VERY likely to throw F means not likely. I THINK mgk had it reversed. Since from jstouts play info post:

(Random) = Random Chance (x00 is likely to x0F isn't likely)

(Take Sack Chance) = Take Sack (x00 is likely to xFF isn't likely)

Damn. That would have made sense too since the lower % were listed first.

Now I'm totally confused (WHAT ELSE IS NEW!)

Looks like I have more testing to do.

Share this post


Link to post
Share on other sites
No 0 actually mean VERY likely to throw F means not likely. I THINK mgk had it reversed. Since from jstouts play info post:

(Random) = Random Chance (x00 is likely to x0F isn't likely)

(Take Sack Chance) = Take Sack (x00 is likely to xFF isn't likely)

False.

post-3789-13413728965453_thumb.jpg

For QB WR priority, F is most likely (almost every throw). 0 is almost never.

Share this post


Link to post
Share on other sites

Sweet - I fully understand it now.

Thanks guys!

....I love how I asked MGK for this info in Dec 09 and am finally getting around to looking at it and understanding it. 2010 was a busy year!

Share this post


Link to post
Share on other sites

The C1 and 91-94 values make sense to me here, but what if you wanted to create a pass play where the QB ran 50% of the time (or more)? I'm not understanding the $B832 (x9842): STAND FOREVER

$B838 (x9848): RUN part.

Share this post


Link to post
Share on other sites

I suggest reviewing the command info document by jstout.

Basically you would do

2x yy yy ( pass commands)

Where 2x = random 20= always ... 2F = 1/16

YY YY = where the random will jump to execute the commands

Example 27 F0 AE will jump to

Swap AE and F0 bytes to get AEFO. Add x10 for nes header. AFF0.

First A= 8 for offensive commands= 8FF0

Share this post


Link to post
Share on other sites

Common Designs:


$B832 (x9842): STAND FOREVER


$B838 (x9848): RUN


$B842 (x9852): SWITCH CONTROL TO RECEIVER


 


Key:


20-2F = RANDOM GOTO (20 is likely to 2F is not likely)


--------------------------------------------------------------------------------


 


So I did some testing with the above.  I think the 20/2F is opposite.  20= not likely, 2F= likely.


On pass plays, I changed the Stand & Switch to Receiver to 20 and Run to 2F and the COA or COM controlled QB would bolt for the LOS 100% of the time and take off with the ball about 1/3 of the time.  Even better, before taking off, he would do a pump fake.  More testing to come....

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


  • Recently Browsing   0 members

    No registered users viewing this page.

×