Run Defense Slider affects DE option read?

[Rudy]

I read this on another board by a member who knows what he's talking about. He likes Oregon and that offense. This is what he said, "In testing some sliders (every slider set seems to be for pro sets, and so I wanted to adjust it to the spread) and found that the run defense slider impacts how often the Defensive End crashes down or stays at home on the read option. At 50 it's 50/50, and the DE stays home more the lower the slider. "

Has anyone noticed this? I don't run the spread but find this interesting. Maybe I should leave run defense at 50 and adjust the run blocking sliders accordingly.

[JeffHCross]

Don't run the Zone Read often, but the few times I've tried it, I haven't seen evidence of this. Of course, there are other factors at play with how often the DE crashes down, i.e. some of the ratings.

[AustinWolv]

IMO, the only way to determine this is to chart it.

The other thing I would say to that is to look at other plays as well.......does the defense just go more aggressively after the ball in general with a higher slider versus just playing the 'reads' more conservatively?

[morsdraconis]

The best way to test it, in my opinion, would be to change the DE's awareness and play recognition ratings to 99 each on a separate roster file (to eliminate that as a factor) for whichever DE would be constantly being the read guy on the play, run the same read option play against the same defensive man coverage play (human player controlling both sides so that the CPU can't individually audible players on the defense to blitz or jump the snap instead of just going straight by play recognition/awareness) and then run it 20-50 times with each setting and chart the differences (if any).

Really damn time consuming and seriously something I might try to start doing tomorrow when I get a chance (cause I'm crazy like that and enjoy the hell out of doing stuff like this).

[morsdraconis]

Ok, just finished my test.

Setup:

• Practice Mode
• 2 player (me controlling hike of offense with no other control besides immediately diving after handing the ball off so that yards gained wouldn't have any effect)
• Offensive play: Shotgun Y-Trips HB Wk Read Option (WVU Playbook)
• Offensive team: WVU
• Defensive play: 4-3 Normal 2 Man Under
• Defensive team: Akron (with modified DE with 99 awareness and 99 play recognition)
• Sliders all at 50 on All American difficulty (besides changing Run Defense Slider)

Results:

[Rudy]

That didn't show any real tendency. I guess the run defense slider doesn't affect it. Nice work.

[AustinWolv]

Well, from 0-50 on the slider yields the same percentage.....mid-30% of staying on the QB.
At 65 and 100, the percentage shifts higher at 42 and 50%. So there is a slight difference but not drastic, not linear clearly, and not enough data points to draw a definitive conclusion.

[Rudy]

Mors tested 200 observations and the total result came to 40.5% (use 0.40 = p). We can test each case separately (50 observations = n) to see what a significant result would be. Testing at the 95% confidence level (this is the accepted standard and why every poll is accurate 19 times out of 20) we would expect a normal result to range from 14 to 26 with the average being 20. All 4 tests are within this range so nothing conclusive can be derived. We must accept the fact that the run defense slider has no effect.

[Rudy]

Mors - did you chart the actual results of all those plays? I'd love to see how bad the cpu rushing numbers are while executing the spread along with the percentage of plays that result in negative yardage. This area needs a LOT of work and is priority #1 for NCAA 12 imo. They need to improve the cpu handling of the spread as well as promote more scrambling.

[morsdraconis]

Mors - did you chart the actual results of all those plays? I'd love to see how bad the cpu rushing numbers are while executing the spread along with the percentage of plays that result in negative yardage. This area needs a LOT of work and is priority #1 for NCAA 12 imo. They need to improve the cpu handling of the spread as well as promote more scrambling.

No I didn't chart actual numbers just whether or not the CPU DE choose to follow the RB or stay home with the QB. I never actually got anything positive out of any of the plays cause I didn't want positive yards to skew the scientific part of it (getting more with the RB as compared to the QB or vice versa).

I think I may need to redo the test though cause going back, I might have been changing the CPU sliders instead of the human sliders, which, if that was the case, that makes it even more interesting that there were such varying results with the slider at 50 the whole time...