Interesting revelation regarding AI interceptions

[fistofrage]

I think the AI QB accuracy/interception problem is tied to the Human QB Accuracy.

If Human QB accuracy is below 50, the AI will throw many more interceptions than if Human QB Accuracy is left at default 50.

I tested this on Heisman. Actually I was just playing default Heisman because I was tired of playing with sliders and wanted to see how my guys would perform in an online match-up againts a human team in dynasty because those games are played at default.

Any, to make a long story short, on Heisman, I put all AI sliders to 75 and left human at default, except put interceptions to 0. The AI rarely threw any interceptions. Called a good gameplan and had decent stats.

When I moved the Human QB accuracy slider to 40, the AI started throwing Interceptions in bunches.

My question is, for those of you guys who are seeing too many interceptions thrown by the AI, do you have your QBA below 50?

[crques]

Quote: Originally Posted by fistofrage

I think the AI QB accuracy/interception problem is tied to the Human QB Accuracy. If Human QB accuracy is below 50, the AI will throw many more interceptions than if Human QB Accuracy is left at default 50. I tested this on Heisman. Actually I was just playing default Heisman because I was tired of playing with sliders and wanted to see how my guys would perform in an online match-up againts a human team in dynasty because those games are played at default. Any, to make a long story short, on Heisman, I put all AI sliders to 75 and left human at default, except put interceptions to 0. The AI rarely threw any interceptions. Called a good gameplan and had decent stats. When I moved the Human QB accuracy slider to 40, the AI started throwing Interceptions in bunches. My question is, for those of you guys who are seeing too many interceptions thrown by the AI, do you have your QBA below 50?

I seriously believe you are on to something here. I was using a lot of different slider combos the first few months that I had the game. Most of these had Human QB Accuracy anywhere from 20 - 40 and the computer was throwing 4 - 5 interceptions per game, even with Human Int at 0. After becoming frustrated with continuous slider adjustments I decided to play with a lot of them near default on Heisman level. Since doing so, I've had much better games and the computer usually only throws 1 or 2 interceptions per game. I've even played a few games where the computer didn't even throw one.

[fistofrage]

Quote: Originally Posted by crques

I seriously believe you are on to something here. I was using a lot of different slider combos the first few months that I had the game. Most of these had Human QB Accuracy anywhere from 20 - 40 and the computer was throwing 4 - 5 interceptions per game, even with Human Int at 0. After becoming frustrated with continuous slider adjustments I decided to play with a lot of them near default on Heisman level. Since doing so, I've had much better games and the computer usually only throws 1 or 2 interceptions per game. I've even played a few games where the computer didn't even throw one.

I swear I'm onto to something big time. Michigan State had 0 Interceptions in 2 games with Human passing at default. Played a classic 17-16 thriller win over Cal and a shootout win over Eastern Michigan 40-35. Believe me, thats something the SPartans would do in real life. . Anyway, I moved the human slider to 40, what did I see, 4 picks, 3 picks with 2 pick 6's, 2 picks both pick 6's. Moved it back for 3 games and saw 1 Int.

Can't be coincidence.

I don't think EA tested any of the sliders or anything so moving something could foul it up.

[crques]

Quote: Originally Posted by fistofrage

I swear I'm onto to something big time. Michigan State had 0 Interceptions in 2 games with Human passing at default. Played a classic 17-16 thriller win over Cal and a shootout win over Eastern Michigan 40-35. Believe me, thats something the SPartans would do in real life. . Anyway, I moved the human slider to 40, what did I see, 4 picks, 3 picks with 2 pick 6's, 2 picks both pick 6's. Moved it back for 3 games and saw 1 Int. Can't be coincidence.

I agree with you. I guess I never really made the connection to that one particular slider but I have noticed that the game plays much better without making a whole lot of adjustments. Adjusting 1 slider seems to screw 4 or 5 things up on this year's game.

[youALREADYknow]

Need a much larger sample size, especially considering the human element.

Would be interesting if true, but the passing game is too east with hum qba at 50 to me.

[fistofrage]

Quote: Originally Posted by youALREADYknow

Need a much larger sample size, especially considering the human element. Would be interesting if true, but the passing game is too east with hum qba at 50 to me.

It is for me too. But if I only use an ask Corso play and don't hot route to receivers, it becomes challenging.

I know you are big into this game and this probably belongs in the slider section, but try these out....

QBA 50/75
PB 0/75
WR 50/75 ***Move human WR catch to whatever you want, but I see drops at 50
RBA 50/75
RB 50/100

AWR 50/100
KD 50/75
INT 0/75
BB 50/100
Tackle 50/75

****Use ask Corso and choose any of the 4 plays on the 1st page only. No hot routing WR routes.

I was only controlling the DT, so no user Int's.

Anyway, any testing under these conditions would be great to get a larger sample size, because I'm getting some great games.

Also, if you want to see realistic FG results, super sim FG's and Xp's.

[youALREADYknow]

I'll give them a try and post some stats so you have more data.

[youALREADYknow]

I just saw that you only control the DT and that changes my opinion on this. I've been able to limit INTs if I only control the DT/DE in the past, so I don't know how much of this is due to having a specific slider at default. Have you still had INT problems by only controlling the DT using some other slider set?

Anyways, just for fun I played a game on 6 min Heisman with your sliders and some custom special teams sliders and it was highly competitive (but very boring at times).

I chose Florida State (86 OVR) against Texas (91 OVR). Texas (CPU) won 16-10. The game was highlighted by two completely inept offenses in the first half, then a bit of scoring in the 2nd half highlighted by Robo QB Colt McCoy who threw 21-23 for 190 yards as he quick passed my secondary to death.

I don't know if I'll get back into playing using DT only and this was my first time ever using strictly Ask Corso plays, but I could see how you play with these in a dynasty.

My only suggestions on these specific sliders would be the following:

1. Lower HUM TKL a bit to help the CPU running game.
2. Raise HUM KND to around 75 to create more incompletions (we all know the QB's will never truly throw incompletions).
3. Lower CPU QBA to 70 to get rid of Robo QB and allow the WR routes to develop more downfield.
4. Increase HUM RBK a bit to allow for more between the tackle running (I wouldn't decrease CPU BB because the pass rush is needed to combat user QB cheese).
5. Raise CPU KND a bit to defend the deep ball better.

I think the above changes would result in more accurate stats to align more with the absurd offensive sim engine and actually should help the CPU more (I should probably be losing by 10+ pts using Ask Corso, handcuffing myself on defense, and using a playbook I'm not very familiar with against a 91 OVR team).

But on the bright side.... NO INTERCEPTIONS! I think that's what you wanted to know.

I don't think this is any reflection on the CPU passing game because they had the ultra-robo QB out there. If you complete 91% of your passes, then of course you're not throwing INTs. I believe there was only one play the entire game where my defender even made a play on the ball.

What I did notice, however, was that the interceptions by the CPU defense actually became a lot more realistic. I'm not sure if that is because I've only been playing my ridiculously hard Online Dynasty for the last month and have somehow become a composed pocket QB or whether the sliders have anything to do with it. Either way, I was very satisfied with the way the user passing game felt and didn't see much of the CPU defender warping and super jumping to get INTs.