no content
AdBlocker active?
It seems you are using software to block advertisements. You could help us if you could switch it off when visiting redzoneaction.org.
The reason is very simple: Advertisements help us running the site, to offer you the game in a good quality for free. So if you like the game, please support us by purchasing a Supporter Account or disabling the AdBlocker on this site.
Thank you very much!
| Main / Discussions / Shotgun 4 Search Forum | |
| Navigation: |< < 1 2 3 9 10 11 > >| | |
| Poster | Message |
|
thomastem
|
posted: 2017-07-19 23:28:40 (ID: 100110065) Report Abuse |
|
Buddy Ryan wrote:
Gambler75 wrote:
Not that much more effective? Picking up blitzes (w/no DE BT) 100% of the time, versus 0% of the time in the others ... amounted to triple the sack rate in the others? 3.5% vs 10% is a pretty nasty difference. thomastem wrote:
The FB blocking and sack rates bring up an interesting point on Superpippo's data set. RZA does not back out negative yards from sack in total passing yards while the NFL does. Further it doesn't even show up as an attempt in the RZA box score. hmm yes that would explain the disconnect between sack-rates and Superpippos numbers... but with that in mind, would it at least allow us to conclude that the impact of pressure (without sacks) is about the same for Shotgun and Non-Shotgun? Blitz % per formation is a variable that would need to be tracked. If you have to blitz 70% of the time against SG to get within .5 yards of other formations that are blitzed 50% I'm not sure if that is close to real life or not. I think if you get the blitz % down you get a better picture. If Supperpippo did not back out sack yardage and add sack attempts and then does so and we see a substantial difference in SG vs other in avg yards per play what we have done is isolated a single problem, FB blocking, as the singular fix that would bring SG to a more realistic formation and then it is just a question of if Pete can tinker with the engine to improve. Data and facts are productive, emotion and complaining not so much. |
|
| Quote Reply Edit | |
|
Buddy Ryan
|
posted: 2017-07-19 23:44:35 (ID: 100110066) Edits found: 1 Report Abuse |
|
thomastem wrote:
...isolated a single problem, FB blocking, as the singular fix that would bring SG to a more realistic formation and then it is just a question of if Pete can tinker with the engine to improve. yeah, FB blocking is a core issue... it would be ideal if its success could be brought down from automatic vs. the blitz in the SG to being more dependend on whatever mix of FB vs. Blitzers-skills Pete feels comfortable with. at the same time the FB in other formations should be able to pick up blitzes at a reasonable rate too - without making the Shotgun redundant and obsolete. the problem might be (could be BS!) that as the illustrations in the manual show, the FB is lined up next to the QB in SG, but behind him in the other formations. maybe he cant pick up blitzes from there by definition of the engine? Last edited on 2017-07-19 23:59:12 by Buddy Ryan |
|
| Quote Reply Edit | |
|
thomastem
|
posted: 2017-07-19 23:55:06 (ID: 100110067) Report Abuse |
|
If we are talking about getting the FB to be realistic he should also be involved blocking in the running game and the HB should be the back blocking in the passing game at least that is what you see in pro teams most of the time.
Your typical SG formation would have the best pass catching/blitz pick up HB in the game. If no blitz or break down of OL in front of him the he either goes in the flat or 5 yards in the middle of the field. Being able to program that would be amazing toward the realism. If FB blocking turns out to make SG OP I'd settle for that being adjusted. |
|
| Quote Reply Edit | |
|
Superpippo
|
posted: 2017-07-20 00:00:36 (ID: 100110069) Edits found: 1 Report Abuse |
|
Sacks are counted in for their negative yds value.
Btw there is also a lot of difference between LB blitz and SF blitz depending on which defense you run, like I said, lot of variables to play with... Last edited on 2017-07-20 00:03:21 by Superpippo |
|
| Quote Reply Edit | |
|
thomastem
|
posted: 2017-07-20 00:03:36 (ID: 100110070) Edits found: 1 Report Abuse |
|
Superpippo wrote:
Sacks are counted in for their negative yds value. What about attempts? If one formation is getting sacked at a higher rate it makes a difference on avg. Last edited on 2017-07-20 00:04:01 by thomastem |
|
| Quote Reply Edit | |
|
Buddy Ryan
|
posted: 2017-07-20 00:09:25 (ID: 100110072) Report Abuse |
|
Superpippo wrote:
Sacks are counted in for their negative yds value. Btw there is also a lot of difference between LB blitz and SF blitz depending on which defense you run, like I said, lot of variables to play with... interesting, thx! what would be your explanation for the non-impact of the 3 times higher sack rate for non-SG-formations? are sacks still such low-numbers-events that over all they dont matter? |
|
| Quote Reply Edit | |
|
Meitheisman
|
posted: 2017-07-20 07:29:09 (ID: 100110079) Report Abuse |
|
passionatelad wrote:
I wish people would get it through there thick sculls. It's not about my game blah blah blah. It's the use and effectiveness of this formation. I wish you would get it through your thick skull that nobody cares what your opinion is until it is supported by data. The data so far does not support your opinion so either you provide better data that supports your opinion or you change your mind on the subject and stop talking about it. It's like if you said "I wish people would get it through there (sic) thick sculls (sic) that it's cold in Miami." While the thermometer is showing that it's 110 degrees in Miami. You'd look like an idiot, wouldn't you? It doesn't matter how strongly your posts are written, the data will always be more valuable and important than anyone's opinion. |
|
| Quote Reply Edit | |
|
hollyhh2000
|
posted: 2017-07-20 09:19:44 (ID: 100110086) Report Abuse |
|
First off great discussion, lets not ruin it with fighting those 'I know everything from gut feeling' guys, please just ignore those posts as the fightings ruin good discussions (too a good percentage a message to myself
 )
Superpippo wrote:
Well, I've done exactly this for around 150000 plays... the plays from games 1-9 this season so far, all elite and division 1 matches not containing bots. Then split up runs/passes against each formation and honestly, SG is not a far superior formation, if anything it's more effective when used as both a passing and running play because of the way people defend against it. 100% passing is actually taking away from it's strength. It's like Thomas says there are many variables that needs to be taken into account, but at the end of the day the average pass from SG4 goes 0,5yd further than the worst passing play. So in 100 passes it might gain you 50yds against someone throwing 100% from a crappy formation. I can absolutely agree to Superpippos comments. I will post some data, but cant support too much as I will be on vacation for the next 3 weeks and my wife will probably kill me if I crunch data instead of spending time with my family
One issue are the many variables. I have 179000 plays from season 25 available, but when I want to compare passing formations with 7-10 yards to go throwing over the middle when the opponent is blitzing, just 34 of those plays from Goalline O were among those 179000, so that number is just too low to make a statement for that offense. And I believe you need to break down that data that deep because each one of those variables has an impact. And you still have different skills having an impact in real game data. So only one man can really make accurate statements how formations match up. Yoda is a tough cookie  so he probably will not share his full wisdom 
However you can't analyze pass direction without losing sack data (no direction of the intended receiver is known, when the QB is sacked), when you compare completion % you have to eliminate screen passes as they have a far higher completion % than normal throws. I |
|
| Quote Reply Edit | |
|
Superpippo
|
posted: 2017-07-20 09:48:04 (ID: 100110091) Edits found: 1 Report Abuse |
|
See below
Last edited on 2017-07-20 09:48:39 by Superpippo |
|
| Quote Reply Edit | |
|
Superpippo
|
posted: 2017-07-20 09:48:10 (ID: 100110092) Report Abuse |
|
Yeah I'll see what I can do too, but it's a lot of work, A LOT! Because like Holly says to really see the undelying differences you need al data split among each defensive formation against each offensive formation, then split Pass/RUN then INT's/Blitzes/Sacks/Pressure, L/R/C and there only so much time I can spend looking at excell fiules
That being said, since you guys asked I ran some extra numbers: so to answer some questions: 1) When it comes to % of sacks compared to blitzed passes, the shotgun formations are around 6-7%, other formations are around 13% so it's more like double instead of triple like said here before (sample size from 2 matchdays = 42357 plays of which 24067 where passes) 2) Sack% is around 1,3% of plays. Statiscal significance of sacks is small, but does need to be taken into accouunt. example: there were 5122 passes from the SG4 in my sample size, but average yardage goes up with 0,35yds if you take negative yardage from sacks out of it. 3) The pass completion percentage: Shotgun formation are the worst when it comes to completion %. 58% for 2WR, 59% for 4wr. Average of all others combined is 61,5% 4) Wasn't asked but quite interesting: Shotgun 2WR and 4WR are ranked 2&3 when it comes to percentage of passes intercepted, only one formation does worse. So a small recap: The shotgun 2WR and 4WR plays are statiscally the best passing plays when it comes to average yardage, they give up least sacks, but have a lower completion rate and a higher interception rate. Sounds pretty good to me. I know I've been hammering on this for quite a bit, but there's a shitload of variables in this game a lot of you seem to overlook, which leads you to make pretty dumb generalisations. Few examples: - We know QB's can feel pressure, in Holly's FB research a lot of you seem to come to the conclusion blitz but no sack = no influence. Why? - - We all know that every defenseive formation has pro's and cons against each offensive formation. On top of that they're all influenced by the strengths and weaknesses of the players utilized, distance to the first down and goal line and then all of this gets put through a RNG. So assumptions made from one game or the few games you remembered something like this or that happened are pretty much useless. - Speaking of pro's and cons, the reason I started keeping this data is because I read a few posts in which people stated i'd rather give up a few yards on the run than on the pass, which is why I always use (insert formation here) against (insert other formation here). I you're one of those people you might want to rethink that... Would you still say that if you would use the exact numbers for that? something like: I'd rather give up 4 yards on each run than to risk giving up half a yard on a pass! 
|
|
| Quote Reply Edit | |
| reply Mark this thread unread | |
| Navigation: |< < 1 2 3 9 10 11 > >| | |
| Main / Discussions / Shotgun 4 | |
