A methodical approach to understanding sliders. Tests & Resuls

[ZaPPPa]

Auto-Pass / Coach mode slider set at bottom of post

===

Having read many of the slider posts on this forum I have seen many different interpretations for how sliders work and what each slider does. I decided to try a methodical approach to isolate each slider and capture what the exact impact is of each individual slider. This post will hopefully help other people in their quest for the perfect slider settings and it may even clear up some of the confusion about how sliders impact the game.

Being a software engineer of complex systems myself, I have a feeling the individual sliders may not be as complicated as they seem. I figured that if I had to program and balance sliders, I'd make the individual slider pretty straightforward and have the complexity lie in the combination of settings.

So with that said, this is a breakdown of the testing I have done on the slider settings for NCAA 14 today.

Test setup

NCAA 14 with Yukon46 rosters. Practice mode with San Jose State offense vs San Jose State defense

Offense and Defense All-American
All sliders at 50
All penalties at 50
Threshold at 50
Ball spot at middle of own 35

Auto Strafing and Auto Passing are ON
No user interference during play

I will post the offense and defense for each play, or at least the one that had the most clear impact so you can verify the results.

The fundamentals of testing are all based around changing no more than ONE variable at a time and creating a dedicated, reproducable situation for the test. Anybody who tweaks an entire slider set and then makes statements about the functionality of a specific slider is merely guessing. So, to establish a baseline, each play was run at least 10 times on default settings (all sliders 50), then at least 10 times at slider 0 and 10 times at slider 100. In reality I probably ran between 500 and 1000 tests total.

A '+' means 'more' or 'higher'
A '-' means 'less', 'fewer' or 'lower'

Passing

QB Accuracy

Off: Shotgun Normal - Strong Sail
Def: Nickel Normal Man Under

QBA at 0				QBA at 100
+ sacks					+ sacks
+ throw aways				+ catch in traffic
+ short passes				+ into double coverage
- interceptions				- deflections
- into traffic

Notes:
At a high value, the QB is more willing to take risk and throw into traffic. Oddly enough, this value seems to impact receivers as well, making them much better at completing a catch in traffic. At lower settings, the QB makes more shorter passes to the flat and a lot more throw aways. Overall the QB is more indecisive and sloppy.

Impact: High

Sweet Spot: 5-10
I ran QBA at 15 and above for a while, but passing completion got too high (70%). With QBA at 5 the passing completion ends up averaging in the high 50's, low 60's.

Pass Blocking

Off: Shotgun Normal - Strong Sail
Def: Nickel Normal - Man Under

PBL at 0				PBL at 100
+ sacks					- sacks
+ throw aways				- throw aways
- time in pocket			- throw into traffic
					+ time in pocket

Notes:
Time in pocket and pocket formation are mainly affected. An indirect result is less interceptions due to less tendency to throw into traffic. Impact of this setting is quite high; at 100 the QB can stay in the pocket almost indefinitely.

Impact: High

Sweet Spot: 50

Wide Receiver Catching

Off: Shotgun Normal - Curl n' Go
Def: Nickel Normal - Man Under

WRC at 0				WRC at 100

Notes:
Not sure if this slider does anything at all. I tried many different situations such as catch in traffic, easy catches in a screen, but nothing jumped out at me.

Impact: Minimal

Pass Coverage

Off: Shotgun Normal - Outs
Def: Nickel Normal - Man Under

Off: Shotgun Normal - FL Flats
Def: Nickel Normal - Man Under

Off: Shotgun Normal - Strong Sail
Def: Nickel Normal - Man Under

Dedicated test
Off: Shotgun Normal - Strong Flood (left)
Def: Dime - 2 Man Under
Look at slot receiver (Y) and count the times the CB anticipates the route taken.

Tests:
PC = 0 vs PC = 100
9/10 vs. 7/10
6/10 vs. 6/10

PC at 0				PC at 100
				+ deflections
				+ defender looks for ball

Additional tests performed:

SJSU vs SJSU
Ball at own 34, on X

Off: Shotgun Normal - Four Verticals
Def: 4-3 Under - Slant Crash Left

As soon as B receiver crosses the v40 marker, float ball to B. Ball should land on the ^40 marker. If ball is overthrown, ignore attempt.

Test Type:		Look-back	Swat	Int	Catch	Drop
PC 0, Int 0:		5		5	0	9	6
PC 100, Int 0: 		19		19	0	0	1
PC 0, Int 100:		7		0	7	10	3
PC 100, Int 100:	20		0	20	0	0

According to these test results the DB look-back is directly related to the Pass Coverage slider and the Interception rate is directly related to the Interception slider. Also, the interception rate is indirectly related to Pass Coverage because a DB needs to look-back to be able to make an interception attempt. At Int 100 all look-backs lead to successful interceptions.

Notes
At high values the defenders are more aware of where the ball is in the air and will actively deflect and try to intercept the ball. At low value, a defender who has his back turned to the ball will be completely unaware of where it is. It may have a very slight effect on the speed of the secondary, but this requires more testing. I do not believe this setting impacts the reaction speed of the secondary on pass plays. Also, as far as I can tell after having run the same play at least 100 times on different settings, there is NO difference in how the CB and Safeties approach the WR. I have done a dedicated test to see of CB route anticipation is affected by this value and test results show this is NOT the case. I know people are saying this impacts defensive positioning and tightness of coverage, but I simply do not see it in a methodical test. If someone has a specific play combination that proves otherwise, please let me know.

This may be one on those settings that seem very complicated, but turn out to be very straight forward: higher = more swats.

Impact: Medium

Sweet Spot: 75-90
This setting reduces catches in double coverage to less than 25%. If the defender is positioned well, he WILL swat the ball. Note that the higher you go with PC, the lower you should go with INT.

Interceptions

Off: Shotgun Normal - Strong Sail
Def: Nickel Normal - Man Under

Int at 0				Int at 100
- interceptions				+ interceptions

Notes:
Looks to be very basic.. Affects the likelyhood of an attempt at defensive catch being succesfull. No apparent side effects

Impact: Medium

Sweet Spot: 35
A lone defensive player WILL intercept a ball if throw right at him. PC at 90 and INT at 35 will lead to about 1 interception per game per QB. Lobbing a ball up into double coverage is asking for trouble.

Rushing


Running Back Ability

Off: Ace Big Close - HB stretch
Def: 4-3 Stack - 2 Man under

Off: Shotgun Normal - PA SE screen
Def: 4-3 Over plus - Engage Eight

Off: Shotgun Normal - Shark HB Wheel
Def: 4-3 Over - Cover 3

RBA at 0				RBA at 100
- Top speed				+ Top speed
- Acceleration				+ Acceleration
- Agility				+ Agility
- Use of special abilities		+ Use of special abilities

Notes:
I was always under the impression that RBA also affected WR after the catch, but this is NOT the case. The HB wheel test confirmed that the RB is affected at all times, making him highly ineffective as a WR and a blocker at low settings. Use of special abilities (juke, spin, stiff arm) are announced as 'oompf' sounds during the run. This also impact the amount of stamina left.

Impact: High

Sweet Spot: 60
In NCAA 14 running backs start using their special abilities at below 50. Because this slider only affects the running back, it cannot deviate too much from 50 otherwise it will give the RB an unfair advantage as a WR.

Run Blocking

Off: Ace Big - HB Dive
Def: 4-3 Stack - Man Under

Off: Ace Big - HB Counter
Def: 4-3 Stack - Man Under

Off: Ace Big - HB Stretch
Def: 4-3 Stack - Man Under

RB at 0					RB at 100
- Duration of block			+ Duration of block

Notes: Power and push of block does not seem to be affected at all, only the ability of the OL to deal with defensive moves. Overall impact is fairly low on both HB dives and stretches

Impact: Low

Rush Defense

Off: Pistol H Twins - Strong Power
Def: 4-3 Stack - Man Under
Note: Run this play and look at the safety at 0 and 100, take note of his reaction at the instant the HB gets the ball. At zero can take up to 4 seconds for the safety to react to the run play. At 100 he will react immediately without fail.

Off: Pistol Slot - HB Draw
Def: Nickel Normal - Cover 3

RD at 0					RD at 100
- Reaction time LB+sec			+ Reaction time lb+sec
- Conservative in zone			- Aggressive in zone
+ Bite on PA				- Bite on PA

Notes:
Greatly impacts the reaction time of linebackers and secondary for rush defense. Most clearly visible at draw plays where corner backs will be caught flat footed. Line backers are more aggressive at high values, being more forward in their zones. At RD 100 the linebacker will never ever bite on a PA pass. This setting does not seem to have any effect on DL rush defense, only LB and secondary.

Impact: Med-High

Sweet Spot: 15-20
Tested at 10:
Mediocre safeties will mis-identify running plays 15-25% of the time, allowing for stretch plays to succeed.
MLB will bite on PA 50% of the time.

Tested at 15:
MLB will bite on PA 40% of time

Tested at 20:
MLB will bite on PA 30% of time.

Tackling

Off: Pistol Ace Twins - Strong Power
Def: 4-3 Stack - Man Under

Tac at 0				Tac at 100
+ broken tackles			- Broken tackles
					+ Big hits

Notes:
As expected, the ability to successfully complete a tackle

Impact: Medium

Sweet Spot: 30
I want break-away runs and the only way to get them is to lower tackling so tackles from behind fail most of the time. This allows a WR and RB to break through the secondary more easily. Overall I like my games a little sloppy and full of surprises and TAC at 30 gives you exactly that.

Threshold

Test performed:
Off: Pistol H Twins - Strong power
Def: Goal Line 5-3-3 - Jam Cover 1

HB has speed of 92. Snap ball and run backwards, have bunch of defense follow you to bottom right corner of field, turn to left corner and run perpendicular to field to get good view of top speeds. Do not use sprint.

At zero
Good separation between speed 70 and speed 90
Noticable top speed difference between 71 and 78
Stretch/Toss plays to HB ineffective due to OL inability to block OLB

At 50
Barely noticable top speed difference between 72 and 76 speed
Slight top speed difference between 72 and 82 speed

At 100
Barely noticable top speed difference between 67 and 82 speed
OL will beat the HB to the goal line on a 30y stretch play

Sweet Spot: 25
Ability for HB to outrun DL and slow LB, but still enough speed for OL to be useful in blocking at the second level.


Conclusion
While many sliders were straightforward, a couple were surprising to me such as the Rush Defense and Pass Coverage. One of the most annoying issues i have with the NCAA series is that I cannot seem to simulate breakthroughs for RB's. 70y runs are extremely rare, but now that I have a better understanding of these sliders, I should be able to make it happen.

I am not done yet. Next up is testing the impact of the threshold (which I kind of did already), game speed and all penalty sliders.

I may, at the end of all my testing, post a slider set of my own, but that is not the goal of this thread.


=====================

Tweaks:

8/18:
Been going back and forth with threshold 25-35. Finally settled on 25 because it makes outside runs possible.
Have played two entire seasons with these sliders.. they have settled until a tuner/patch is released.
Clipping still at 35. Although this leads to a bunch of uncalled clipping, I pretend it compensates for poor AI offensive blocking.
Switched to slow mode for testing. I avoid outside pitch runs because of the animation glitch.

8/7:
QB completion too high, QBA -5 and PC + 10
Intentional Grounding -25
Trying out the Clipping at 35

8/9:
Going back to Off: All-American and Def: Heisman
Increased threshold slightly (35) to allow for slightly slower players to break through further when they pass the last defender
Offsides and Holding to 99, seems to be more effective than 100 (I hope)
OPI/DPI to 1, in case 0 causes issues
RTP to 54 to keep the penalties down
RTK to 1, for testing

8/11:
In an attempt to bring passing completion down, QBA now at 5. Passing Completion percentage is now between 50 and 65%
Testing player threshold at 22
OPI set to 50 to improve passing defense.
Clipping to 50
WRC catching to 40. Since drops are not used in any stats or calculations, I consider them poor passes and use them to simulate better defense.
Pass Blocking improved slightly to reduce sacks

Slider set I am currently using in my Coach Mode (play calling with audibles) offline dynasty:

Offensive Skill: All-American (8/9)
Defensive Skill: Heisman (8/9)
Quarter Length: 9 mins
Game Speed: Slow (8/18)
Threshold: 25 (8/18)
HFA: On
Ice: On

Both User/CPU

QBA: 5 (8/11)
PB: 50 (8/11)
WRC: 40 (8/11)
RBA: 60 (8/11)
RB: 35 (8/9)
PC: 90 (8/9)
Int: 35 (8/11)
RD: 20
Tac: 40
FGP: 40
FGA: 35
PP: 45
PA: 65
KP: 40

Offsides: 99 (8/9)
False Start: 80
Holding: 99 (8/9)
Facemask: 55(8/6)
OPI: 50 (8/11)
DPI: 1 (8/9)
KR/PR int: 99 (8/9)
Clipping: 35 (8/18)
Int Grounding: 75 (8/7)
Rough passer: 54 (8/9)
Rough kicker: 1 (8/9)

[The_Gaming_Disciple]

Thank you for all of your hard work. Keep in mind that practice mode plays differently with regards to slider functionality in play now, and dynasty (on, offline).

[TheBleedingRed21]

Very interesting, we play CPU QB ACC at 5 and see all day short routes or throw aways, I see why now. They still are deadly so maybe a higher QB ACC would help the CPU pass better but not be Peyton Manning. Need to find out more about that!

[ZaPPPa]

@The_Gaming_Disciple
I see what you're saying. It does not seem to affect normal sliders, but it does the penalties... Which is a shame, because it will make testing those so much more difficult.

@TheBleedingRed21
I was in the same boat. There are a couple of options:

Increase Pass Coverage.. Let the QB throw, but have more passes swatted. I am not sure if increasing PC also increases the number of interceptions, but it would not surprise me. If so, lower the INT slider accordingly.

Decrease Pass Blocking. Give the QB less time and force him to make bad decisions. I have not gotten to testing penalty sliders yet, but I have tried increasing the Intentional Grounding slider to 75 in my dynasty and it seems to encourage throw-away passes. My QB went from 75% passing completion to barely over 50%.

[charter04]

Even though practice mode plays slighty different than other modes I still think its usefully for what your doing. All your findings seem similar to what I see in the real games. Keep working its good stuff. I would love to see your threshold results.

[MRGM420]

Great post and hope to see your future efforts! Appreciate your approach and repping San Jose St in the tests hahah. From nitty gritty san jo city myself lol

[ZaPPPa]

I went to college in Europe, but when I moved to the US and later the Bay Area I kind of adopted SJSU as my college to root for. Really hoping that Fales is going to make a late 1st round pick. He deserves it.


I looked at threshold settings and will add my initial findings to the first post.

I also looked at Rush Defense some more and decided to create a 'sweet spot' entry for sliders. This 'sweet spot' is for that specific slider only. For example, for Rush Defense I looked at reasonable mis-reads and reaction times for defensive players. With human reaction time being between 100 and 200 milliseconds and adding another 300ms to identify the play, I figured a reasonable safety would take approximately 500ms to determine whether he has to stay in zone or rush the HB on a stretch rush play. I tweaked the sweet spot to hit that number. Then I did the same for LB biting on PA. I have not checked real-life stats (and if someone has those numbers I would love to see them), but I figure a reasonable MLB will bite on a PA about 25% of the time.
Obviously EA wouldn't be EA if one slider setting wouldn't get both stats correct, so the end result is a compromise. MLB bite may be a bit high and Safety reaction a little low, but it evens out and gives semi-realistic gameplay.

[gatorsinRENO]

I think this adds a lot of credibility towards the notion that sliders affect how the players respond and what happens as a result of that. The INT slider is one that I was always afraid to raise because I thought that would automatically give the CPU more ints, but having done testing like this actually proved me wrong and taught me how to be a better player.