User Pass
Home Sign Up Contact Log In
Forum > Test Server Discussion > Test Server: Worrying About Run Game Ineffectiveness
Page:
 
PP
offline
Link
 
Originally posted by smeberen

O-Lineman need to move forward on running plays.


That is a VERY big key right there! If the G is free (OT or TE, for that matter), he should be going out to hit a LB or safe, not waiting for them to come to him.
 
Adderfist
online
Link
 
http://test.goallineblitz.com/game/replay.pl?game_id=686&pbp_id=171113

Poor pathing by the RT. Should go forward with the HB and try and -lead- block.

EDIT: And on the pitch plays, don't nerf it unless you test it with a proper pitch D.
Last edited Apr 11, 2009 18:26:37
 
Sulfadias
offline
Link
 
Originally posted by Adderfist
http://test.goallineblitz.com/game/replay.pl?game_id=686&pbp_id=171113

Poor pathing by the RT. Should go forward with the HB and try and -lead- block.

EDIT: And on the pitch plays, don't nerf it unless you test it with a proper pitch D.


Wierd, looks like that RT was heading towards his "man assignment" versus the LDE.

I hope that doesn't happen alot!

 
odg62
offline
Link
 
Originally posted by thehazyone

....Again, every run was shut down. OK, now I'm a bit worried so I move it to long/pass focus. Teams SHOULD be able to run against this type of defense. It was actually worse than short/run as there was 141 carries for 151 yards....

http://test.goallineblitz.com/game/game.pl?game_id=649

137 carries for 267 yards. A little better but still way under where it should be for a Long/Pass focus defense.

Maybe I need to input a whole set of rosters but I was assuming that the base teams were meant to be relatively even against one another and should provide a balanced, realistic game. Regardless, I see signs that show that the run game is way underpowered now and I can only foresee the large number of complaints that will arise when RB's are averaging 2 yards per carry. If I can have my team playing long/pass focus every down, the game will be easy.

I'd list individual plays but there's so many of them I don't know where to start.

Am I overreacting here or do I have a right to be worried?


First off, i appreciate the work the testers do despite the many ethical issues that arise from the whole situation, what will follow in this post is (mostly) not about the ethics behind the test server and is not designed to spark discussion about said ethics.


That said, I just got done watching all of the first half and a lot of the 2nd half of the game linked above. Every single play i watched the defense was on run focus except for one which was balanced. every play. Most of the plays were on long coverage, but not all, there were some on short and medium distance. The coverage distance dosent play a big factor anyway since it only affects the safeties (on run focus that still leaves 7 players in the box right up on the line).

I would expect that a team would average about 2 yards a carry against an about equal defense that focuses on the run every play and would go as far as to say i think that is how the sim should work.

Please dont allow the misinformation quoted above to influence any changes to the run game. Had i actually seen a pass focused D holding a run game like that, i would of agreed with the sentiment of the op somewhat, not the suggested fix method, but definitley the sentiment of "to hard to run." However, the D is on run focus, and run focus absolutely should contain the run.

This brings up a few other issues. Issues with the testers, test server, and how it is being used.

Obviously, a lot of power comes with being a tester, i.e. the oppurtunity to find exploits and see how new changes work, respect from the GLB community who view testers as people who know a lot about the game, and most importantly the ability to bring what you learn from the TS directly to Borts attention while making high profile threads about it in the TS Discussion fourm.

Here is the issue - How does a tester not know the difference between run and pass focus? Any half decent DC can tell you in 1 second if a play is a run, balanced or pass focused. Any half decent DC can also tell you that the diffrence between coverage distance and play focus. Yet, the tester who started this thread seemed to miss the fact that 99% of the plays in the game he linked as "long/pass" were run focused. Did he actually watch the game? or did he just look at the numbers? Maybe he put the wrong link? this is a problem.

Here is the problem - He made a high profile thread claiming that the run now dosent work even if the D only looks for the pass. The information in the thread was wrong and has now misguided many GLB members and possibly caused unwarranted concern/action on the part of the person who codes the game.

Back to the issue - No tester should be making this kind of mistake and then make a thread about it, for whatever reason it may have occurred, it shouldn't happen. It still brings up the question, Why? Why did this mistake happen? (Im sorry hazy, this isnt an attack on you, you just kinda put yourself in the line of fire here) Was it because of ineptness? Did the tester not know the difference between run and pass focus? I'd have to say it would be a poor reflection on all testers the whole test server program and GLB in general if this was the case. Maybe the tester does know the difference in focuses, but that would have to mean he didnt watch the game. Again, dosent reflect real well on the way the test server is being run, just looking at the box score dosent tell the whole story. Next, (i hate to go down this road but this post wouldnt be complete without it, im not accusing anyone of anything im just trying to make a point, i cant stress enough that the following is in no way an actual accusation, just trying to make a point), considering the background of the tester, you have to wonder about ulterior motives. Anyone who pays attention knows that the tester who started this thred is a big fan of the run game and his awsome pro championship team relys heavily on the run. Is it possible that a tester tried to pull a fast one and use their influence to create a sentiment among a group of users and the creator of the game in order to have changes in the game go the way they would perfer? Its not out of the question until someone can come up with a better explination of how we are on page 6 and just now its getting pointed out that the D in above replay is in fact on run focus.

If this thread is so off base, is it that much of a stretch to wonder how many other "findings" on the test server are just as misinformed?

This thread shows that there are major holes in and problems with - the testing process - the knowledge, methods, motives, and influence of the testers - tester selection - the process by which the test server creates change.

A lot of people (customers, owners, staff) have a lot of time and money riding on this game developing into a complete, succesful, reliable, enjoyable, fair, and FINISHED product. Those who have invested their time, money, lives into this game should without a doubt expect nothing less then near perfection from the people who have the most influence and control over the direction and state of this game.

If you wanna flame me fine, but to save space and time my stock answer to any flamer is fuck you.

If you wanna discuss legitimately this im all ears and more then happy and willing to talk about and reasonably discuss what i typed.








 
smallbugger
offline
Link
 
I have to agree with the above post in that Cover 1, balanced focus is more of a run stopping formation.

or here going cover 0 doesn't scream pass focused to me
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164490
that looks like working as intended

Here you have the power back running outside
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164506
you would assume the power back would be better running up the middle and the speed/elusive backs would be the threat outside.... this again looks like working as intended to me. A slow HB doing an outside run should be caught for a loss usually

Another one with a faster back
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164507
the HB does get some yards, but gains even though this is a play going away from the strong side while the defense isn't even playing the strong side..... looks good to me. The defense should be able to run down the HB in this situation

Again a weakside rush (away from the majority of your blockers) even though the defense is in a center shift....
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164508

All this looks like bad play calling against the defenses called, not reasons for huge concerns. If more plays went to the strong side because the defense was always in center shift more yards would have been gained and that average would look better.

Thanks odg62... I was taking the word for it that hazy knew what he was doing... the issue with testers is they have their own players... maybe hazy thinks his HB's 5.2 ypc career average should be typical.




Last edited Apr 12, 2009 06:26:18
 
smallbugger
offline
Link
 
Here is a strong side run working... the defense is in a center shift so the offense has an extra blocker and the ROLB gets cought up on the weak side.... looks good to me
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164509

Here the DT beats the LG/C.... that might be an issue with the O-line or this match-up, but this looks like a DT making a play not a issue with the HBs
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164510

Again cover 0... this is working right IMO
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164511

A strong side outside run for 4 yards... might have been more if the FB blocked a bit better, that play looks fine to me. Because the LBs are center shifted they are not a factor in this play and its the SS and FS that have to make a stop
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164512

A slower back running weakside outside away from their blocking against a center shift.... this looks fine to me
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164513

What is pass focused about this play? If this was pass focused the SS wouldn't be blitzing...it would be in cover 2 against the Shotgun.... again nothing wrong with this play... just bad play calling.
http://test.goallineblitz.com/game/replay.pl?game_id=649&pbp_id=164514


I'm going to try to stop here... but this defense is far from pass focused and would in an even game be passed against fairly well as there would be 1 on 1 coverage with no help over the top in most cases WR to CB... and even havethe #3WR against a man coverage FS on 3rd down in single back formation....

Looks like you ran a 50/50 strong side and weakside runs and 50/50 outside and inside running without setting for your power back to do inside runs and your speed HB to do outside runs, ect. This bad of playcalling and set-up should earn 2ypc or less against a center shifted cover 1 defense.

My respect for testers just went out the window.... show me a real game plan...

Bad gameplanning should give you 2 ypc on average or less... good gameplanning should give you 3+ ypc average
Last edited Apr 12, 2009 07:42:05
 
thehazyone
NFL Replacement Refs SUCK
offline
Link
 
I've run various tests, using different types of game plans/AI's, some of them are basic AI, some are advanced AI with basic settings, other are more specific. Defensively, I've been running standard defenses that I've seen used at the Pro level against those formations with the only thing changing being the play type focus (run/balanced/pass) and the distance focus (short/medium/long). Really my initial goal has been to identify that ALL aspects of the running game are working, not just a power back running up the middle or a perfect play call against a defense. I want a running game where there are stuffs, average gains, and the occasional long run.

If I said the defense was on long, pass focus - it was on long, pass focus. No misinformation being given here. I'll do another test sim and take screen shots of the AI - maybe the problem is with how the focuses are working. I thought they looked a bit off when I was reviewing the game tape myself but thought to myself "nah, I set them to long/pass, that must be right."
 
smallbugger
offline
Link
 
Originally posted by thehazyone
I've run various tests, using different types of game plans/AI's, some of them are basic AI, some are advanced AI with basic settings, other are more specific. Defensively, I've been running standard defenses that I've seen used at the Pro level against those formations with the only thing changing being the play type focus (run/balanced/pass) and the distance focus (short/medium/long). Really my initial goal has been to identify that ALL aspects of the running game are working, not just a power back running up the middle or a perfect play call against a defense. I want a running game where there are stuffs, average gains, and the occasional long run.

If I said the defense was on long, pass focus - it was on long, pass focus. No misinformation being given here. I'll do another test sim and take screen shots of the AI - maybe the problem is with how the focuses are working. I thought they looked a bit off when I was reviewing the game tape myself but thought to myself "nah, I set them to long/pass, that must be right."


don't need to see the AI... only the results to see what happened... at least the game I saw everything seemed to be right. You didn't like the results, but that was because of bad play calling against the defense.... if you want to test against "pass focused" then put defenses in cover 2 and see how that does.

As well as run 70% strong side runs if the defense is going to be in a LB center shift... I'd except your going to see a increase in your HB effectiveness
Last edited Apr 12, 2009 08:00:33
 
OttawaShane
offline
Link
 
Cover 1 long pass focus is still going to do decently vs the run.

In another thread there was cut/paste of an anon tester advising bort the backs shouldn't take a stamina hit till they break "5 or 6" tackles and some other stuff that was unrealsitic buff for powerbacks - ODG's points are valid..
 
thehazyone
NFL Replacement Refs SUCK
offline
Link
 
I'm running a test right now with no blitzes and some different base formations (3-3-5, 4-2-5 against 3 WR formations and 4-3 Cover 2 against 2 WR formations). I've also made some depth chart adjustments and offense AI adjustments to where power backs will get inside runs, elusives will get outside runs, and combos will get a bit of both. I also removed the stupid Shotgun HB Draw since it sucks - I still want to see how the remainder of the plays work though.

My problem with the logic mentioned above is that EVERY team in Pro runs pretty much the same defense. It's always 4-3 Cover 1 against the 2 WR set, and always 4-3 Cover 0 against the Single Back - there are some teams that will vary, but those are easily the two most predominant play calls against those formations. And yes, I know, that you should be able to pass against these defenses because they are so geared to stop the run, but the thing is those defenses stop the pass very effectively.

I'll post the test results as soon as it's done running and I've had a chance to look it over.
Last edited Apr 12, 2009 08:16:30
 
thehazyone
NFL Replacement Refs SUCK
offline
Link
 
While I'm at it, I want to address some of the points odg brought up.

Originally posted by odg62

Here is the issue - How does a tester not know the difference between run and pass focus? Any half decent DC can tell you in 1 second if a play is a run, balanced or pass focused. Any half decent DC can also tell you that the diffrence between coverage distance and play focus. Yet, the tester who started this thread seemed to miss the fact that 99% of the plays in the game he linked as "long/pass" were run focused. Did he actually watch the game? or did he just look at the numbers? Maybe he put the wrong link? this is a problem.


I know the difference between run and pass focus. I've been setting my own AI and the AI of other teams since season 1 and my teams have had plenty of success, so it's not a matter of me not knowing what I'm doing. I double check every AI before I submit the game to the test server to make sure it is matching up with what I am wanting it do. I assure you AI for these defenses was set to long pass. The problem might have been as some people mentioned that the base defenses themselves lend to stopping the run (i.e. 4-3 Cover 0 and 4-3 Cover 1). And yes, I watch the game. I notice the good runs and bad runs. You have to remember in a particular test run, I'm not necessarily looking for the game itself to average 4 yards per carry. I'm looking for a mix of stuffs, average runs, and long runs. Thus far there have been far too many stuffs and not enough of the other stuff, and that's the issues that we're trying to address.

Originally posted by odg62
Here is the problem - He made a high profile thread claiming that the run now dosent work even if the D only looks for the pass. The information in the thread was wrong and has now misguided many GLB members and possibly caused unwarranted concern/action on the part of the person who codes the game.

Back to the issue - No tester should be making this kind of mistake and then make a thread about it, for whatever reason it may have occurred, it shouldn't happen. It still brings up the question, Why? Why did this mistake happen? (Im sorry hazy, this isnt an attack on you, you just kinda put yourself in the line of fire here) Was it because of ineptness? Did the tester not know the difference between run and pass focus? I'd have to say it would be a poor reflection on all testers the whole test server program and GLB in general if this was the case. Maybe the tester does know the difference in focuses, but that would have to mean he didnt watch the game. Again, dosent reflect real well on the way the test server is being run, just looking at the box score dosent tell the whole story.


I have no problems putting myself in the line of fire. You'll find that I am going to be as honest and open about what I am testing. Probably more than any one else that is doing it (not a knock on the other testers, they are excellent in their own right, but I know that I'll probably communicate more about what I am doing than anyone else). Again, I know the differences in focuses - that's what they were "set" to. And I watch the replays of the game, you do see me posting individual plays that I think need addressing. I don't just get those plays from looking at the box scores.

Originally posted by odg62
Next, (i hate to go down this road but this post wouldnt be complete without it, im not accusing anyone of anything im just trying to make a point, i cant stress enough that the following is in no way an actual accusation, just trying to make a point), considering the background of the tester, you have to wonder about ulterior motives. Anyone who pays attention knows that the tester who started this thred is a big fan of the run game and his awsome pro championship team relys heavily on the run. Is it possible that a tester tried to pull a fast one and use their influence to create a sentiment among a group of users and the creator of the game in order to have changes in the game go the way they would perfer? Its not out of the question until someone can come up with a better explination of how we are on page 6 and just now its getting pointed out that the D in above replay is in fact on run focus.


I'm not trying to pull a fast one. As I stated in my initial thread, I was looking for possible exploit plays like the Strong I Slam prior to season 9 starting so that we could nip those in the bud before they happened. How I did that was by running tests of each individual run play against defenses that I knew were effective against the run. When I noticed how weak the running game looked, I changed the focuses of those defenses to medium/balanced and the problem was still there. I then changed it to long/pass and the problem was still there. Granted, I was running a variety of runs, some that aren't that effective, but on a whole the run game was problematic and so I brought it to Bort's attention. It wasn't until I'd run some ten test sims that I did this though, because I wanted to make sure.

While it is true that my team heavily focused on the run last season, it was because we had to in order to be competitive. Prior to this season we have always been a very balanced team. I'm not going to waste time going back through the seasons to prove this point, you can do that for yourself. I'm trying to bring the users of GLB a balanced game where both the RUN and PASS are effective. I was planning on doing the same thing for the pass game as for the run game as soon as I am done with the run game.

 
thehazyone
NFL Replacement Refs SUCK
offline
Link
 
Ran two test sims with the following AI:

Offense:

30% Elusive Outside
15% Combo Outside
15% Combo Inside
40% Power Inside

The only running play removed from the playbook was Shotgun HB Draw. All other plays were used.

Defense:

All defenses were set to Man to Man Coverage, Long Coverage Distance, Pass Play Focus, Center Shifts

3 WR:
25% 4-3 Cover 0 Blitz SS ROLB
25% 4-3 Cover 0 No Blitz
25% 3-3-5 Cover 0 No Blitz
25% 4-2-5 Cover 0 No Blitz

2 WR:
30% 4-3 Cover 1 Blitz SS ROLB
35% 4-3 Cover 1 No Blitz
35% 4-3 Cover 2 No Blitz

http://test.goallineblitz.com/game/game.pl?game_id=692
http://test.goallineblitz.com/game/game.pl?game_id=693

For the two games, the RB's had the following numbers:

Elusive backs 66 carries 303.5 yards 4.6 ypc
Combo backs 53 carries 165 yards 3.1 ypc
Power backs 102 carries 341.5 yards 3.3 ypc

Breakdown by play type (pitch or handoff) for each of the types of RB's was as follows:

Elusive Pitch
<2 yards 9
2.5-4 yards 6
4.5-7 yards 4
7.5-10 yards 3
10.5+ yards 2

Elusive Handoff
<2 yards 10
2.5-4 yards 13
4.5-7 yards 11
7.5-10 yards 3
10.5+ yards 4

Combo Pitch
<2 yards 11
2.5-4 yards 3
4.5-7 yards 3
7.5-10 yards 2
10.5+ yards 2

Combo Handoff
<2 yards 11
2.5-4 yards 14
4.5-7 yards 3
7.5-10 yards 2
10.5+ yards 0

Power Handoff
<2 yards 28
2.5-4 yards 42
4.5-7 yards 21
7.5-10 yards 6
10.5+ yards 0

Summary:

Much better overall balance in the run game, although some of the defenses were intentionally soft on the run. The elusive backs had better success than any of the other backs but it wasn't so out of whack to be a major cause for concern, especially since most of the defenses weren't outside run type of defenses. A good balance of stops, average runs, and the occasional long run.

Next up, I'm going to start testing the individual run plays one by one again against a multitude of defenses - one, defenses that should stop the play; two, defenses that should limit the play, a balanced defense; and three, defenses that should give up yardage against the play.
Last edited Apr 12, 2009 10:14:36
 
smallbugger
offline
Link
 
things look pretty good... if bort has the passing game worth something things might work out. Forcing defenses to have to make decisions.... I'm seeing blockers only blocking 1 guy at a time... so a free defense man can make plays. Not sure the builds... bt honestly wouldn't mind the power backs breaking a couple more tackles, but overall I'm enthused.
 
Jaxx
offline
Link
 
I have always thought it worked out as

Cover 0 -- Run coverage while maintaining some pass
So you should be able to burn this setup with a decent passing attack since it sets man coverage and has the possibility of creating mismatches

Cover 1 -- Brings SS up in the box to assist in run and short passing game
Strong against the run with the SS suport but leaves a gap in the middle of the field for pass protection and little help for deep crossing routes over the top

Cover 2 -- Best set up for pass defense on first or second down without selling out on the run coverage
You have plenty of over the top coverage excellent pass defense weak or should be weaker than the other two against the run

I think most of the current problems people are upset about would be cured with an improvement on zone coverage AI.
 
tautology
offline
Link
 
Originally posted by thehazyone

My problem with the logic mentioned above is that EVERY team in Pro runs pretty much the same defense. It's always 4-3 Cover 1 against the 2 WR set, and always 4-3 Cover 0 against the Single Back - there are some teams that will vary, but those are easily the two most predominant play calls against those formations. And yes, I know, that you should be able to pass against these defenses because they are so geared to stop the run, but the thing is those defenses stop the pass very effectively.



Highlighting the very true and somewhat over-looked part of this post...
 
Page:
 


You are not logged in. Please log in if you want to post a reply.