User Pass
Home Sign Up Contact Log In
Forum > Test Server Discussion > Some testing results
Page:
 
tragula
title
offline
Link
 
Originally posted by bhall43
As far as speedsters...its thee same build as the slot guys with slot machine being the only difference. Probably why slot 2 on t2 only got 22% instead of around 30%. I don't know. Wasn't my test.


Slot2 on team 2 got less CF since he played less against the 1st type of D. (My guess)
 
bhall43
offline
Link
 
I ran a lot of testing on it as well with run test team and tt1. Those are marked mostly as deep zone testing. Run test team has a single % piece of avoid fake. Tt1 is mostly double stacked.

But that test was mostly to check whether it fires appropriately and it served its purpose.

Avoid fake aeq has certainly shown to drown the effects of catch fake. Think a couple seasons back when cf was thought to be maybe not worth a lot. That's kinda the effect it is giving off towards af%. Tho every now and then you get a couple good ones still.
 
bhall43
offline
Link
 
Originally posted by tragula
Slot2 on team 2 got less CF since he played less against the 1st type of D. (My guess)


Maybe. I just assumed since I didn't see either of the speedsters in his conclusions that the slot guy might not have been in the slot.
 
gbororats
Bugs Moderator
offline
Link
 
The tests are not only testing the rate of fire, it is also testing its affect on the defenders as well. IE. Are multiple dots being taken out at the same time by the same catch fake. There's a lot more being looked at than just how many times it fires.
 
bhall43
offline
Link
 
Originally posted by gbororats
The tests are not only testing the rate of fire, it is also testing its affect on the defenders as well. IE. Are multiple dots being taken out at the same time by the same catch fake. There's a lot more being looked at than just how many times it fires.


Yes sorry that as well.

My tests were mostly the affects vs single and double stacked af. But we are also testing the zones too.
 
Dr. E
offline
Link
 
Originally posted by tragula
Talking about the CF test :

I am saying that in day -6 there will be an adjustment to the change which was just tested .

The CF test is a bunch of random test parameters showing little to none thought. The measured variables are no much different. There is no test methodology, and any change that will be made/approved based on that will have nothing to do with 'balanced' sim. So unless GLB is lucky after two days of wide sim it will turn out that the formula's need to be adjusted again.

Examples :
(1) Team 1, Speedster WR Two- Catch writes : 26% of routes CF fired
No : It is closer to 50% when QB has time, and close to 10% when he doesn't. The 25% average has nothing to do with the sim, it is just tell you how many game with the 1st type of D was used and how many of the other one.
(2) Team 1, Slot WR Two : he played 51 plays against the cover defense, which is below his average number of plays per game vs the blitz defense.
(3) Team 1 and 2 are so different that trying to compare between then is just
a. Team 2, doesn' t have a speed WR at all.
b. Team 2 one receiver less then team 1.

Real conclusion:
- Catch fake fires a lot if you give it time.
- Avoid fake may not work. But then again it may work.
- Fake % AEQ seems important (comparing TEs and WRs). But then again it may not.



Actually there is some methodology. Any time I took part in a test, I also went back and looked at my own teams games and included those results as a "control"/comparison. Also, when testing advances, the older test serves as a "control". And, those posted results aren't the only ones, just from what you could call, the "primary" on that subject. Overall, there are generally a lot of games run and input from pretty much everyone on the test server. Which by the way, anyone can apply to become. I've seen it posted a few times over the time I've played GLB. But back to the number of games run, doing it that way pretty insures the average will be what to expect. It is after all what is done by statisticians, throw away the two extremes and keep the middle. Not saying there isn't a better method out there, but I think there is a time constraint on the testers who for the most part are messing with GLB after work. I mess when I'm not playing with my grandson. TEs, hmm, mine don't have catch fake, both are 1000 yard receivers, don't think they need anything. Their routes are the bomb. Just my opinion mind you.
 
Catch22
offline
Link
 
Originally posted by gbororats
The tests are not only testing the rate of fire, it is also testing its affect on the defenders as well. IE. Are multiple dots being taken out at the same time by the same catch fake. There's a lot more being looked at than just how many times it fires.


That was not the purpose of the CF test results posted in this thread. The purpose of the CF test was a simple one. To see the rate at which CF fires and to see if it was within the ranges we wanted it to be. That's it. I only posted the results so that no one could claim testers had "inside" knowledge.

Now other testers were indeed testing what you mentioned, but I want to make it clear that this test was being done solely for one reason only.
 
bhall43
offline
Link
 
Originally posted by Catch22
That was not the purpose of the CF test results posted in this thread. The purpose of the CF test was a simple one. To see the rate at which CF fires and to see if it was within the ranges we wanted it to be. That's it. I only posted the results so that no one could claim testers had "inside" knowledge.

Now other testers were indeed testing what you mentioned, but I want to make it clear that this test was being done solely for one reason only.


Thats what I thought when you said that after that test.
 
Dub J
offline
Link
 
What I gather is confidence is a waste of SPs.

 
gbororats
Bugs Moderator
offline
Link
 
Originally posted by Catch22
That was not the purpose of the CF test results posted in this thread. The purpose of the CF test was a simple one. To see the rate at which CF fires and to see if it was within the ranges we wanted it to be. That's it. I only posted the results so that no one could claim testers had "inside" knowledge.

Now other testers were indeed testing what you mentioned, but I want to make it clear that this test was being done solely for one reason only.


I know but people here are panicking thinking that's all that's being tested
 
evileyez
Tester
offline
Link
 
Glb user base never overreacts
 
bhall43
offline
Link
 
Originally posted by evileyez
Glb user base never overreacts


If I listed everything I was looking at in my description box people would think im bolicking them.
 
Longhornfan1024
HOOD
offline
Link
 
Originally posted by Catch22
That was not the purpose of the CF test results posted in this thread. The purpose of the CF test was a simple one. To see the rate at which CF fires and to see if it was within the ranges we wanted it to be. That's it. I only posted the results so that no one could claim testers had "inside" knowledge.

Now other testers were indeed testing what you mentioned, but I want to make it clear that this test was being done solely for one reason only.


Please link the equivalent thread from last season.

Also, I wonder why GLB is saying that Avoid Fake % is going to be the answer to preventing these OP'd fakes. Wouldn't that just mean that you're just weighting the roll slightly more in favor of avoiding the fake? All AF% does is give an equivalent percentage boost to the avoid fake attributes (vision and SV). Should a DB really be required to have +15% to vision and SV to prevent an otherwise OP'd fake? Why not just change the ratio more so that a DB who invests in SV and vision at the cost of other attributes can do a decent job of preventing the OP'd fakes and a DB who adds avoid fake % can do a great job preventing the OP'd fake along with other fakes. By doing what you are doing, you are making avoid fake % a requirement.

And on that topic, what are the plateau DBs who don't currently have avoid fake % (because it wasn't needed until season 30) supposed to do to prevent fakes? Most of the DBs that would have been the best out there avoided avoid fake % because it wouldn't have much of an impact (until this next season). Now they are screwed because it becomes mandatory and they can't work it into their build.
 
bdnannac
offline
Link
 
You know someone's going to say CEQ!
 
Longhornfan1024
HOOD
offline
Link
 
Originally posted by bdnannac
You know someone's going to say CEQ!


And I'll point out how that just gives GLB more money. And even then it might be a trade-off that an agent doesn't want to have to make--a trade off that wouldn't have been required if not for OP'd fakes.
 
Page:
 


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