User Pass
Home Sign Up Contact Log In
Forum > Discuss GLB Issues With Catch22 > NOT A BUG - CLOSED > Wide Open Incompletion Case Study - Intended to answer Borts questions about the frequency of this error
Page:
 
Adderfist
offline
Link
 
Originally posted by Sept 30th Anouncement
Been testing some updates related to the WR catching problem some teams have been seeing (check the bugs forum) to see if I can find something that doesn't affect much beyond those players who are having the problem. Not 100% there yet, but I think I'm pretty close to an update that should work.

 
Mike Rogers
Lead Tester
offline
Link
 
Good news, and as I suggest, Bort generally does try to work on these Bugs, but the more time he spends in here, the less time he spends testing.

(Normally no news is good news, although its not great PR.)
 
tigerbait22
offline
Link
 
Originally posted by Mike Rogers
Good news, and as I suggest, Bort generally does try to work on these Bugs, but the more time he spends in here, the less time he spends testing.

(Normally no news is good news, although its not great PR.)


I'm very happy with this announcement... If it can't be fixed, not a problem, just a heads up is all I ask for... Doesn't take long to say I'm working on it
 
Randy03Texags
offline
Link
 
This particular play never works

http://goallineblitz.com/game/replay.pl?game_id=193655&pbp_id=2858861
http://goallineblitz.com/game/replay.pl?game_id=193655&pbp_id=2858876
http://goallineblitz.com/game/replay.pl?game_id=193655&pbp_id=2859086

Almost always the right WR gets open, the pass gets there, but Ive only seen that play result in a reception like 10% of the time. Its open about 70% of the time, but rarely does it result in a reception. Seems funny to me.

 
caponesgun
offline
Link
 
Originally posted by Randy03Texags
This particular play never works

http://goallineblitz.com/game/replay.pl?game_id=193655&pbp_id=2858861
http://goallineblitz.com/game/replay.pl?game_id=193655&pbp_id=2858876
http://goallineblitz.com/game/replay.pl?game_id=193655&pbp_id=2859086

Almost always the right WR gets open, the pass gets there, but Ive only seen that play result in a reception like 10% of the time. Its open about 70% of the time, but rarely does it result in a reception. Seems funny to me.



Could be a graphical error. The CB actually has the angle on the WR but graphically the dots aren't on top of each other?

Or the play sucks. Just tossin' out ideas.
 
Mike Rogers
Lead Tester
offline
Link
 
I had that problem on a few teams, if you look at it rationally, the QB is trying to float the ball up over the CB, and the WR is running away from him.

So you have a high trajectory, and a moving target.

In real life that gotta be the toughest pass to make? a Touch pass over a DB to the sidelines?

Quick hint, look at the different success rate of a balanced/Possession WR and a Speedster WR for a solution.
 
BloodFarmer
offline
Link
 
Originally posted by Mike Rogers

Quick hint, look at the different success rate of a balanced/Possession WR and a Speedster WR for a solution.


Can you suggest where this information might be found. (Sorry, I don't visit the forums very often so yes I am asking because I don't want to do all of the work to find it myself since there is no search function.)

 
Mike Rogers
Lead Tester
offline
Link
 
Not in public forums, but a lot of people have put in a lot of hard work tracking play succes by build type etc.

Most of that info is in private or team forums.

It all comes down to experimentation
Last edited Sep 30, 2008 07:58:48
 
Sarg01
offline
Link
 
Originally posted by rj414

For my QB, at level 31, a 10% increase would get him up around 70. I think the guys who are currently completing 70+% are throwing more short and medium passes, and thus I don't think this change would affect them as much.


By what logic is it reasonable for a long pass completion percentage to be on par with a shot or medium pass %? It should be less - and frankly, it should be more than 10% less, since the risk on a 60% completion rate is 40%, and 30% on a 70% completed pass. So are 2 to 3 times the yards (plus the breakaway TD chance) worth a 33% increase in risk? Of course they are.

With 3 long passes at even 60% chance (or any pass over 10 yards really) you're looking at only a 0.4 * 0.4 * 0.4 chance of failure - that's just 6.4% chance, because you only need 1 of those to complete to advance, and around 2.5% if you use 4th down long passing as well.
 
DarkVirtue
offline
Link
 
This game has lots of bugs still and the Incompletions are a joke. My team has lost 4 games now due to this. I had to switch to the run and to lofty passing.
 
RAPB
offline
Link
 
Originally posted by DarkVirtue
This game has lots of bugs still and the Incompletions are a joke. My team has lost 4 games now due to this. I had to switch to the run and to lofty passing.


This issue only comes up if the WRs are very fast and the QB got great strength. It's no problem of lv. 17 guys. Try another excuse.

Originally posted by Sarg01
Originally posted by rj414


For my QB, at level 31, a 10% increase would get him up around 70. I think the guys who are currently completing 70+% are throwing more short and medium passes, and thus I don't think this change would affect them as much.


By what logic is it reasonable for a long pass completion percentage to be on par with a shot or medium pass %? It should be less - and frankly, it should be more than 10% less, since the risk on a 60% completion rate is 40%, and 30% on a 70% completed pass. So are 2 to 3 times the yards (plus the breakaway TD chance) worth a 33% increase in risk? Of course they are.

With 3 long passes at even 60% chance (or any pass over 10 yards really) you're looking at only a 0.4 * 0.4 * 0.4 chance of failure - that's just 6.4% chance, because you only need 1 of those to complete to advance, and around 2.5% if you use 4th down long passing as well.


This worries me. And he's totally right. Throw 4 long passes 10+ yards and if the completion rate is 60 %, you'll miss out the first down in 3 of 100 cases. So a regular game needs 33 drives to stop you ONCE.
Last edited Sep 30, 2008 09:08:55
 
BloodFarmer
offline
Link
 
Originally posted by Mike Rogers
Not in public forums, but a lot of people have put in a lot of hard work tracking play succes by build type etc.

Most of that info is in private or team forums.

It all comes down to experimentation


OK, thanks for your response. My team is working on the problem but I didn't know if the general forums had already hashed out the delicacies of the conundrum.

 
j10er
offline
Link
 
Originally posted by Sarg01
Originally posted by rj414


For my QB, at level 31, a 10% increase would get him up around 70. I think the guys who are currently completing 70+% are throwing more short and medium passes, and thus I don't think this change would affect them as much.


By what logic is it reasonable for a long pass completion percentage to be on par with a shot or medium pass %? It should be less - and frankly, it should be more than 10% less, since the risk on a 60% completion rate is 40%, and 30% on a 70% completed pass. So are 2 to 3 times the yards (plus the breakaway TD chance) worth a 33% increase in risk? Of course they are.

With 3 long passes at even 60% chance (or any pass over 10 yards really) you're looking at only a 0.4 * 0.4 * 0.4 chance of failure - that's just 6.4% chance, because you only need 1 of those to complete to advance, and around 2.5% if you use 4th down long passing as well.


60% is too low on wide-open, perfectly-thrown passes. At the higher levels, I'd say it should be about 75%. If two people have a simple 20-yard pass and catch, shouldn't they catch it most of the time, especially if they're actually football players?

However, when you add in CBs/Safeties (deflections and presence), pressure on the QB, the QB's throw, the WR's build and the RNG, you have a varying range of completion percentages. When everything works against the QB/WR, it should be low. But if everything works perfectly, it should be high. Somewhere in between lies the perfect range, and not a single percentage.

The passing game is so complicated and so dynamic that it is impossible to definitively state that the long passing completion percentage should rest at a static percentage for every team/situation.
 
Sarg01
offline
Link
 
Originally posted by j10er

60% is too low on wide-open, perfectly-thrown passes. At the higher levels, I'd say it should be about 75%. If two people have a simple 20-yard pass and catch, shouldn't they catch it most of the time, especially if they're actually football players?

However, when you add in CBs/Safeties (deflections and presence), pressure on the QB, the QB's throw, the WR's build and the RNG, you have a varying range of completion percentages. When everything works against the QB/WR, it should be low. But if everything works perfectly, it should be high. Somewhere in between lies the perfect range, and not a single percentage.

The passing game is so complicated and so dynamic that it is impossible to definitively state that the long passing completion percentage should rest at a static percentage for every team/situation.


I didn't say it should be a set number. I said the numbers quoted as what QBs are actually experiencing in game are too high. Personally, I believe this is because there's a lot of problems with high level coverage, but it doesn't really matter why the numbers are so high. If QBs are honestly experiencing 60% long completion rate regularly against decent Ds of appropriate level, then Bort needs to be finding ways to decrease that rate as a higher priority than fixing what may be a legitimate issue that will have the side effect of further unbalancing the game.

The original post I was responding to was a note that long passing % could safely be increased by about 10% (Bort's estimation of the side impact of the bug fix) because the only guys with 70% completions were short and medium passers. I strongly disagree with that. There's way too much deflected on 1st, dropped on 2nd, 16 yards on 3rd! already.
 
j10er
offline
Link
 
Originally posted by Sarg01
If QBs are honestly experiencing 60% long completion rate regularly against decent Ds of appropriate level, then Bort needs to be finding ways to decrease that rate as a higher priority than fixing what may be a legitimate issue that will have the side effect of further unbalancing the game.


Completely agree with this, but is this happening? I don't think I've seen such a high long completion % on any team I've watched, but of course we don't actually know what constitutes a long pass. Is a 10-15 yard pass actually considered "long" still at level 30+?

And then what is causing high completion %? WR speed? CB builds? Defensive vs. offensive tactics? A combination? It's incredibly complicated.

I guess what I'm saying is that, with no pressure, a good throw and a wide-open WR, long passes should be completed about 85% of the time, assuming long is 15+ yards. Once you factor in the other variables, though, the number should drop quickly.

Probably the best way to do it is by steps. Start with wide-open, then add weak CB coverage, then tighter, then perfect (without deflection), then add pressure, poorer throws, etc. In order to get the passing game right, a long study involving the programming should probably be done.

It's not going to happen. So what can we really use as our template in order to figure out "right" long completion percentages? I'm have no idea, honestly.

Originally posted by Sarg01
The original post I was responding to was a note that long passing % could safely be increased by about 10% (Bort's estimation of the side impact of the bug fix) because the only guys with 70% completions were short and medium passers. I strongly disagree with that. There's way too much deflected on 1st, dropped on 2nd, 16 yards on 3rd! already.


There are defensive settings to prevent the long 3rd down completion though. And I have no problem with the above scenario, assuming it doesn't happen on every series, allowing the offense to easily march down the field.

Three things can happen on a pass, and two of them are bad. I think the best way to handle this is to increase the completion of long passes but to also increase the chance of it being intercepted. So you might see the deflection on 1st, drop on 2nd, 16 yards on 3rd series, but you'll eventually see an interception somewhere in there too. It should be a gamble, with a slightly higher reward AND risk.
 
Page:
 


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